Let’s Talk About Customization

One topic that has come up frequently in our private beta newsgroups as well as here in blog comments from time to time is the issue of customization.

As with every component of the Office 2007 user interface redesign, we put a lot of thought into how much customization to provide; today I'm going to try to walk you through our thought process.

Many of you have been passionate in conveying feedback that you wish the UI had absolute customizability. As in my article on the size of the Ribbon, I'm going to lay the facts out on the table and hopefully it will help you to at least understand the rationale behind the decisions we made (even if you wish we had made different ones.)

What is Customization?

There are many aspects to customization in a software user interface. The ability to change the visual appearance, to change preferences, and to turn pieces of the UI on or off are all aspects of customization.

Most frequently among power users, the term "customization" is used to represent the ability certain programs have to add, remove, and relocate commands within the UI.

The History of Customization in Office

Command Bars, introduced in Office 97, were kind of a nirvana of customization capabilities. With Command Bars, you could change virtually anything imaginable within the organization of the menus and toolbars: create new ones, move buttons from toolbars to menus and back, use a built-in icon editor to directly edit the pixels of the icons, etc.

Unfortunately, this flexibility came at a price in terms of the complexity of Command Bars and the kinds of layouts and controls it could support. One of the reasons that many of the prior attempts to simplify the UI were unsuccessful was that any feature had to work within this ultra-customizable framework where you could never predict where a control might live or how it might be presented to the user.

There were downsides for normal users as well. When we go on site visits to watch people use Office 97-2003 in their place of business, we often find that Office has been ravaged by the effects of accidental customization.

In fact, one of the most frequent questions we are asked by people during on-site usability research is: "How can I get the menus back to the top of the window?"

Because of the ultimate flexibility of Command Bars, you can make one small misplaced click and suddenly the menu bar is docked to the left side of the screen or floating in space. Of course, this could have been improved somewhat by some sensible measures such as locking the UI by default, but it does illustrate the different ways a power user and a more typical user think about the same feature.

How Many People Customize?

When we started designing the Office 2007 user interface, one of the first and frequent discussions we had was: "what is the right kind of customization to include in the UI?"

We started collecting research by talking to some of our expert users within large companies, who in several cases assured us that "everyone" customizes their UI to optimize it for the most efficient use possible and, furthermore, that any new UI needed to be at least as customizable as Command Bars (preferably even more.)

An interesting perspective, but it can be dangerous to base decisions on just a few opinions (especially when the first word in the opinion is "everyone.")

The next step was to see what was actually happening in the real world: were people customizing as much we assumed?

Given that we already had one of the most customizable user interfaces of all-time in Office 2003, looking through the real-world data could help us to confirm objectively how many people were customizing their UI and exactly how they were customizing it.

Looking across a hundred million or so people using Office 2003, here's what we found:

  • In fewer than 2% of sessions, the program was running with customized command bars.

  • Of the 2% of sessions with customizations present, 85% included customization of four or fewer commands.

Needless to say, we were surprised, but when we looked at the statistics in detail, this data matched that which had been collected from other sources and in historical research.

It breaks down like this: in ~1.9% of sessions, buttons have been added, removed, or moved between toolbars and menus. (Changing the docking position or location of an entire toolbar is not counted as a customization. Buttons added by add-ins or templates are also not counted.)

Of the customized sessions, around 85% of them had only what we'd call minor customizations: four or fewer buttons. Most of these are added toolbar buttons, either from the command well or from a toolbar people don't want to keep up all the time. And even within these 85% of 2% of sessions, there are patterns that emerge.

The most popular single customization? Removing the "Read" button from the Standard toolbar in Word.

The PowerPoint team helped by collecting hundreds of screenshots of heavily-customized versions of PowerPoint from professional slide designers. It turns out that many of the same customizations are widespread among these users, such as adding Send to Back as a top-level command so that it's not buried four-levels deep in the Drawing toolbar.

Finding so many patterns in the customizations was a heartening revelation, because it implied that if we got enough of the details of the command organization right based on these common customizations, many fewer customizations might be necessary.

Furthermore, because Office 2003 is still relatively new, it is deployed disproportionately among power users--the very people who most likely to customize. In all of our analyses, we try to be aware that the raw statistical data skews slightly towards early-adopting power users. This is probably much less true today than it was more than two years ago when we first compiled customization statistics.

(One metric that causes us to believe that the Office 2003 data slightly skews towards power users even today: Every month, the average screen resolution of people using Office 2003 decreases as more of the core installed base adopts it.)

Using the Data to Drive the Design

Even though Microsoft is a big company, we don't have unlimited design or development resources. Given how ambitious our plans were for reinventing the Office user interface, we had to be realistic and optimize for the most common scenarios first.

So, we took a pragmatic approach and decided to focus on the 99.7% case: people who don't take advantage of customization or only use it to customize four or fewer commands. Out of this goal was born the Quick Access Toolbar.

The Quick Access Toolbar is designed to make it easy to add controls, galleries, and groups from anywhere in the Ribbon: just right-click the thing you want to add and choose "Add to Quick Access Toolbar" from the context menu. We designed the customization model to be efficient but with the goal of "zero customization complexity"; it would be unacceptable for customization to cause the user interface to degrade as it did so often with Command Bars.

Adding controls to a persistent toolbar allows people to have one-click access to the features they choose from anywhere in the product, and can help eliminate problematic command loops as well.

We also paid close attention to common customizations in Office 2003, making sure that commands are organized together for maximum efficiency. This helps to further reduce the need for customization.

Command Location Customization in Office 2007

What command location customizations are supported in Office 2007?

Here's a list of the major capabilities:

  • Add any control in the Ribbon to the Quick Access Toolbar
  • Add an entire Ribbon group to the Quick Access Toolbar as a single icon
  • Add a gallery to the Quick Access Toolbar
  • Add individual menu items to the Quick Access Toolbar
  • Add any command from the command well to the Quick Access Toolbar
  • Add macros to the Quick Access Toolbar and choose an icon and label for them
  • Add separators to help organize the Quick Access Toolbar
  • Reorder controls within the Quick Access Toolbar
  • Full-width Quick Access Toolbar mode below the Ribbon
  • Auto-assigned keyboard shortcuts given to customized Quick Access Toolbar commands
  • Customize the content of many galleries (especially in Word)
  • Customization based on use of many galleries (Recent Documents, Margins, Shapes, Themes, etc.)
  • Complete customization of status bar
  • Customize the Ribbon via XML in document template
  • Use RibbonX XML to customize Ribbon content through a COM Add-in

As you can see, there's quite a lot of customization available in the Office 2007 UI. The laser focus on the Quick Access Toolbar is evident as a result of mapping the design of the product to real-world use.

That said, the last two items on the list will be of interest for expert users who are craving more control over customizing Ribbon content. RibbonX, which has been written about in this space many times, provides an XML interface for describing Ribbon content, including repurposing built-in controls and groups.

You can build custom tabs and groups in RibbonX already; it's just a matter of loading the XML into the program you're using. There are a few ways to do this, including saving the XML into your default document template.

Undoubtedly, people will write tools to help take the XML coding out of using RibbonX to customize Ribbon content. Patrick Schmid has illustrated many of the techniques necessary to make this work on his blog. We've added some additional capabilities to the object model post-Beta 2 that will help these tools along, such as the ability to query for the icon of a command or to execute it directly from code.

Beta Feedback

Nevertheless, throughout the beta cycle we've received requests for additional customization capabilities. There have been discussions about the desire for it in private beta newsgroups, and people have posted comments on the topic here as well. One reader in particular has graced me with at least five personal flame-mails, complete with speculation about my ancestry and theories about the difficulty I will have finding future employment once I've been fired.

If we really optimized for the 99.7% case (98% + (85% of 2%)), why are we hearing so much about this issue?

Because 0.3% of the 450 million paid Office customers still represents 1.35 million people.

And given the strong correlation between expert users and customization, one would expect that these 0.3% are precisely those who download early betas, participate in private beta programs, visit enthusiast web sites, read technical blogs, and generally are interested in and participate in the software development process.

On the other hand, in our long-term broad deployments of Office 2007 Beta 1, Beta 1 Technical Refresh, and now Beta 2 at customer sites, customization has not been a particularly hot issue--nor has it been in Send a Smile feedback.

I have definitely heard those of you who spend the time and effort necessary to build totally custom menus and toolbars in Office 2003, and I hope that a combination of the Quick Access Toolbar and RibbonX-based customization will provide you with a similar level of flexibility in Office 2007, albeit using different technologies.

Addressing Feedback

In a way, all of this attention around customization is a bit like a TV show getting canceled.

Even the least successful network shows attract a lot of viewers in absolute terms--just not relative to the opportunity cost of keeping them on the air. Financially, it doesn't make sense for a network like NBC to keep around an underwatched sitcom like "Joey," but if you're one of the people who like "Joey" it doesn't sting any less when it's canceled. The fact that you're in a small minority doesn't console you.

In other words, the decisions we made around customization in Office 2007 weren't based on absolutes ("no one should be allowed to customize the Ribbon") but instead based on pragmatic use of resources ("let's start by getting the core aspects of the design right for everyone's benefit.")

Is additional customization against the design philosophy of the Ribbon? No, provided that we could add it in such a way that it added no additional complexity for the vast majority of people who aren't interested in it.

Within our resource and schedule constraints, we've acted on the expert user feedback by continuing to add additional customization throughout the beta cycle.

In Beta 1, we added the ability to move the Quick Access Toolbar below the Ribbon to allow many more controls to be added to it. In Beta 1 Technical Refresh, we introduced the ability to add whole groups to the QAT, the ability to add galleries to the QAT, separators, and new command well functionality. In Beta 2, we added the ability to assign custom keyboard shortcuts by adding controls to the QAT.

Are there other customization features we'd like to add eventually? Of course. One example is the ability to customize the contents of the Mini Toolbar--this was an affordance we included in the original spec, but ultimately didn't get done for Office 2007. In future versions, could I imagine adding built-in facilities for customizing Ribbon content? Yes, I could imagine it.

At the same time, I honestly believe that Office 2007 as it stands will be a great user experience for people of all skill levels. If you're a power user (as I am), there's a lot we designed just for you, and even more coming post-Beta 2.

The fundamental principles and constructs of the new UI--galleries, Ribbon layout and command organization, Live Preview, Contextual Tabs, the Mini Toolbar, and all the rest--have been designed to benefit everyone who uses Office, power users and novices alike.

Other Aspects of Customization

A different kind of customization people ask about occasionally is the ability to relocate certain parts of the user interface. The two biggest feature requests in this area are for a vertical version of the Ribbon (to take advantage of widescreen monitors especially in Word) and the ability to "float" and drag around certain parts of the UI.

Some day in the not-too-distant future, I plan to write a post about why we built a horizontal version of the Ribbon instead of a vertical one. There are several compelling reasons, but needless to say we looked at prototypes of both aspect ratios in great detail before making a decision.

Could I imagine a hypothetical version of the Ribbon designed to dock to the side of the screen instead of the top? Of course. Will we build it in a future version? Nothing's for sure, but if it's the right thing for the UI platform moving forward, we'll definitely consider it.

The story with floating UI is similar. Floating toolbars in Office 97-2003 caused a lot of problems, primarily because they were forced on people as the primary means of accessing many features. As a result, toolbars were always popping up over top of what people were working on, needing to be dragged out of the way, or mistakenly docking to the side of the screen.

Our design mantra for Office 2007 was that default feature access wouldn't rely on floating things popping up on top of the document; the UI would be in a single, consolidated, consistent place.

But there are a few repetitive action scenarios in which it would be useful to be able to float UI just to make it closer to the area of the document you're working in. While the data shows that the vast majority of people don't take advantage of this functionality in current versions of Office, we've definitely prototyped a more expert mode in which you could "tear off" groups from the Ribbon or the Quick Access Toolbar, or even a whole tab of the Ribbon to move to a secondary monitor.

Again, were we to build such a feature in a future version, it would be in a way that had no possibility of accidental misuse, and in a way that added minimal complexity to mainline cases.

Last but not least, there's always a lively discussion to be had around choosing a visual appearance for any piece of software.

Unlike in past versions of Office, we allow you to choose your color scheme directly in Office 2007. You can change between the three schemes from the first page of the Options dialog box. People always want more color choices, but I hope with the addition of the third scheme in the final product that most everyone will be able to find a look for the product they're happy with.


As with everything in the Office 2007 user interface redesign, we informed the design both by analyzing the usage data and then by adding a sprinkling of anecdotal feedback. We focused first on the most common scenarios which benefit the broadest set of customers. Then we've used the feedback from the beta process to drive iterative improvements to the design to help satisfy more of the specialized uses of the product.

In the next version of Office, we will undoubtedly look at all of the feedback we get on the completed product (which, as more people use it, will be more fully representative of the entire user base) and decide how to continue to move the UI platform forward.

One goal won't change however: to empower people to do great work in Office as easily as possible.

Comments (114)

  1. brad says:

    Interesting post. The main reason I avoid customizing (and I think this applies to a lot of people) is that I know I’m going to switch computers one day, or upgrade to a new version of Office, or both. I’m sure there is a way to transfer your customized settings to a new machine or a new version of Office, and maybe it’s easy….I see a "Save My Settings" wizard in my Office Tools folder (how many users are aware that this exists?). But still the process of migrating my settings to a new machine feels intimidating and unlikely to succeed perfectly, and so I’ve never bothered taking time to customize any of my Office applications.

  2. Stevbe says:

    So the 1.3 million power users, who help people use office the most and typically use office themselves more than most … well thanks for the support but you are out of luck. We retained the most obscure commands but no customizing the ribbon if you are not a developer. Oh … and those of you who built apps using office as a platform … well that just got a lot harder. While a mugger can explain that they need the money for drugs doesn’t change the fact that they stole your wallet.

  3. Lukas says:

    Hey, I commented back on an earlier post, begging you to maintain customizability.

    Well it sounds like you’ve struck just the right balance.  Optimize for the "average user" while still making customization possible (don’t worry about making it easy) for the power user.  Nice.

  4. Jensen,

    Thank you for your explanations. I now understand better the decisions about enduser customisation (thats not to say I agree with all of them <g>), and your statement that programmed customisation has been excluded  from the customisation stats removes a lot of confusion.

    I observe that all my customers except for some smaller companies use custom Excel addins which create their own toolbars and menus, usually several addins at a time.

    Do you have any statistics showing if this is typical for large customers or not? (I have not yet found a sensible approach for migrating these kinds of customers to Excel 2007 …).

  5. A User says:

    So, being a power user I am not a valued customer. Ok, being a power user I can figure out a way to deal with it. Still, I am disappointed:

    One person’s essential feature is another person’s bloatware, and vice versa. What may be a natural default for you may be an option I would never use under any circumstance, and vice versa. That is why customizability is valuable.

    It is sobering to consider that for 98% the most important customization may be a choice of color schemes.


    "The enjoyment of one’s tools is essential to the successful practice of one’s craft."

    — Donald E. Knuth, The Art of Computer Programming

  6. Messenia says:

    My biggest problem in Office 12 has been the inability to customize the right-click menus.  I don’t want to have to add all of my preferred commands to the QAT and run up to the top of my screen whenever I need one.  I want to right-click and get at the command *I* want.  

  7. Andy Wilkinson says:

    Nice article yet again Jensen…

    But maybe Office users have been spoilt in the past with an over-customisable UI? The question perhaps is, if previous versions had no customisation at all, what would the feedback on the beta be?

    (a) I can’t customise my UI how I want to.

    (b) You should have put command X in ribbon Y in the "out of the box" UI.

    I’d bet on option (b).

    I accept that you can argue that everyone uses Office in a different way, but do they really? How many people can *really* say that more than 10% (or even 1%) of the time they are doing something more than writing a letter/report/etc., changing some styles and inserting a table/picture/chart? And even if you get more advanced with tables of contents, etc., I hope nobody is wasting 10% of the entire document time on the task of inserting them!

  8. LGFN says:

    So, I see that customization is a battle already lost.

    The only thing to do now is, wipe away the tears and say good bye, and, wipe away the sweat and start working the old, hard way.

    Well, I’m not giving up altogether so soon; I’m still looking forward for it by its next birthday (becoming a teenager-version 13).

  9. [ICR] says:

    I aplaud the customization desisions chosen, and am glad to have some statistics to back it up.

  10. Lex says:

    It’s hard to try and argue with those numbers. Personally like the guy who posted before I tend not to customise due to swapping machines.

    In fact I’d say the prevous poster who complained about the needs of a power user who will teach the masses – well it’s easier to help people if your both using the same interface – and if your answer to a problem is first customise this… then your probably on the wrong route already.

    One thing that I felt the product lacked (tho I may be mistaken in it’s non existance) is a reset to defaults option – something that the Repair feature never seemed to do. That and lockable UI… but you alredy mentioned that 🙂

  11. Ruben says:

    "There are three kinds of lies: lies, damned lies, and statistics."

    That’s what I am reminded of whenever people revert to citing their own statistics to prove a point. I mean, 98%? With what margin of error? How large was the sample? How did you measure the results? Where did you do these tests? Etc.

    Believe me: 36.78% of readers don’t buy this. And 98.53% of them are spot on. Don’t believe me? Numbers don’t lie; honest! I tell ya, the next person that’s citing a statistic to prove their point…

    Come to think of it, I’ve never heard of anyone being visited by MS to watch their use of Office. Ever. Except from MS blogs, that is. Where do I sign up? Let’s skew the statistics! US-only, right? Damn.

    Couldn’t you just say: "We couldn’t find anyone that did that. Well, except for Larry three doors down the hall."

  12. Eli says:

    Customization is totally overrated, it just serves to calm people who like to think they’re in control of everything.

  13. I don’t think customization is the problem. I think the problem is that the Office team has put the features in the wrong place. Things are just too difficult to find.

    I found the Office 2007 beta so hostile that I’ve uninstalled it.

  14. Stefan KZVB says:

    Well, I do not think those statistics represent the real world! We’ve got some customizations of the Office 2000 GUI and when rolling out (Outlook) 2003 later on we decided AGAINST using the customer experience program – and we disabled it by policy and customized settings.

    But even if we let the customer experience program enabled and already had updated completely to 2003 when you created the statistics you would NOT have received all the results from our company – simply because only few of our clients do actually have internet access. Also I doubt that the company firewall would have let the traffic sent.

    I think there are many companies with similiar environments – results from all of those are missing in the statistics.

    For that reason I believe the release after Office 2007 will be more customizeable – just to encourage these companies finally to update their Office to the new GUI world…

  15. Ruben:

    As I believe Jensen mentions, the "statistics" came from CEIP data, presumably from *millions* of users. If you didn’t have CEIP enabled on your machine, then in effect you were registered but didn’t bother to vote.

    I use a highly customized version of Word (though not so highly customized as some other users of my acquaintance), and I will be scrambling to fill up the QAT when I start using Word 2007 (and there will be some commands I still won’t entirely be able to replicate that way), but I do think Jensen has done a masterly job of explaining how and why the customizability decisions were reached, and I can see the point even though I will be one of the 0.3%/1.35 million users who will be adversely affected.

  16. Joe says:


    I think the customization decisions you’ve made are spot on. But PLEASE tell me that you’re going to let developers add buttons to existing groups by the final release.

  17. I think this is a great post, Jensen. The power users who are making comments about how you’ve let them down seem to be missing the point: you have to concentrate resources on satisfying the majority of people.

    Power users are intelligent enough to work around the issues, and the really sensible ones will make a living from working around these issues.

    Sure, I’d like more customisation, but I think this post, one of the best you’ve written, is a great justification for making tough business decisions.

    Remember that, everyone. Microsoft, evil or not, is a business. It cannot afford to satisfy every last customer. It must make tough decisions on where to allocate resources. The power users can cope, so stop moaning!

  18. Patrick Schmid says:

    Joe: I highly doubt it. The entire automatic resizing of groups as the window size changes (e.g. for lower resolutions or when you make the window narrower) is hand-coded and hence depends on those groups being static. That’s probably just one of the reasons why it won’t happen.

    The add-in I am working on that gives (power) users the ability back to customize the ribbon itself is almost done. I’ll see if I can find the time tonight to implement the last feature bits and get rid of a few bugs. Then I’ll make an alpha version accessible to the tech beta users, followed by a beta a week later or so.

  19. Joe says:

    Patrick: Thanks for your reply. I understand your explanation, but I don’t think it’s a good reason. Custom buttons don’t resize, so there’s no reason why other buttons shouldn’t just resize around them. OK, so maybe it’ll be difficult when two buttons that would normally combine to a menu have had a button placed between them. Solution: either don’t combine those buttons in that case, or at the least, let me add buttons on the end of a group that won’t interfere with the others. If Microsoft don’t allow us to add buttons to built in groups, it totally contradicts their own guidelines to place buttons where users should expect to find them. i.e. if i have a button that inserts a predefined comment, it should go in the Comments group.

  20. Joe says:

    Andy: I’m happy with the customization available in office 2007, but don’t forget, in Office 2003, you didn’t have to be a super duper programmer to customize it. There will be plenty of people who dragged and dropped buttons to get them just the way they want them, who are now completely unable to acheive those same results because learning XML or creating a comm addin is beyond their abilities.

  21. GregM says:

    "Unlike in past versions of Office, we allow you to choose your color scheme directly in Office 2007. You can change between the three schemes from the first page of the Options dialog box."

    Please explain how forcing the user to select from a very small fixed set of color schemes is better than using the color scheme that the user has *already* defined through the operating system.

    I would much prefer this to read:

    "Unlike in past versions of Office, we allow you to set your color scheme directly in the operating system, so that it applies to all applications."

  22. Vorn says:

    Once upon a time (I think it was like ten years ago) I turn on /every single menu option in existance/ for Office 5 for Mac.

    That was dumb, i could never find anything ever again.

    Nowadays the main extent of my "customization" is adding half a dozen standard toolbars to Visio’s opening configuration.


  23. Tim says:

    I really don’t think that you can complain about not being noticed when you decide to hide yourself. If you didn’t choose to be part of the customer experience program, and you blocked yourself off from providing any info about your usage to Microsoft, you shouldn’t be surprised when they don’t include you in their changes.

    You sound like th child that hides from his mother and then gets upset when she doesn’t notice him.

  24. Jensen, this has to be the single best post to date on your blog.  It reads like gospel.  In my opinion, one of the strongest points of the Macintosh has always been a patent limit on the degree to which customization is allowed.  While the parallels between office applications and operating systems may not be perfect, the logic of your case is wholly inescapable.  I’m not the least bit surprised to hear that you’ve been bombarded by customization complaints.  These are likely the same people that think multiple inheritance was an essential feature of a programming language.  😉

    Lastly, I’d like to say that posts like this are what really bring the value of Microsoft’s new more inclusive development and beta processes to full fruition.

  25. Jon Peltier says:

    I’ve heard this number before:

    "In fewer than 2% of sessions, the program was running with customized command bars."

    And I always suspected, but never knew for sure, that:

    "Buttons added by add-ins or templates are also not counted."

    Thanks for clarifying.

  26. Patrick Schmid says:

    Tim: I think the perception of the Costumer Experience Improvement Program among power users and IT in general was that it is just another Microsoft phone home feature that better should be turned off. The importance of it never really got through until the beta of Office 2007 started. I am among all the people who deliberatly switched this feature off as I was concerned for my privacy.

    Andrew: What would we do without multiple inheritance? 😉

  27. Jon Peltier says:

    "Furthermore, because Office 2003 is still relatively new, it is deployed disproportionately among power users–the very people who most likely to customize. In all of our analyses, we try to be aware that the raw statistical data skews slightly towards early-adopting power users."

    I’d say most of the power users I know fit into one of two categories.

    1. They work for large companies whose IT departments are very slow to upgrade applications like Office. In my last corporate job, which I left in 2004, I was still using Office 97.

    2. They work as consultants or contractors to the companies listed in #1.

    I suspect that a lot of the "early-adopting power users" are the moms and pops who have finally upgraded their Windows 95 machines for something newer to read their email and browse the web. The replacement machines came bundled with the trial version of Office 2003, and when the trial period ran out, they merely upgraded the license.

    The first couple of computers I bought with the Office 2003 trial on it, I uninstalled the trial version of 2003 and installed Office 2000, because that was what most of my clients were using. More recently I run a hodgepodge of Office versions on a hodgepodge of machines, and I will admit that I really do prefer Excel 2003 to prior versions. It crashes less frequently, and the data list feature is worth the price of the upgrade.

  28. Jon Peltier says:

    "…toolbars were always popping up over top of what people were working on, needing to be dragged out of the way…."

    Yes, but I recall with amusement when I first noticed the ‘mouse-aware’ floating toolbars in 2003 (or was it 2002). I think I spent an hour moving them around  by selecting the underlying cells. Having to move the toolbar that is occasionally in the way is to me a small price to pay tfor the ability to locate it right on the 50 yard line where the action is.

    "…we’ve definitely prototyped a more expert mode in which you could "tear off" groups from the Ribbon or the Quick Access Toolbar…"

    I’m hoping B2TR, but I really don’t think so. This would be an awesome recognition of the comments of a small but admittedly vocal fraction of your users.

  29. Jon Peltier says:

    "On the other hand, in our long-term broad deployments of Office 2007 Beta 1, Beta 1 Technical Refresh, and now Beta 2 at customer sites, customization has not been a particularly hot issue…."

    We grew tired of flogging that horse. It wasn’t getting any deader, and meanwhile I was seeing progress made by Patrick and others that I might be able to steal, I mean, borrow.

  30. Patrick Schmid says:

    Jon: thief! 😉 Sure go ahead, that’s what my blog is for 🙂

  31. GaryK says:

    you mention that .3% of the users may be power users and that’s why you may get a lot of requests for customization during betas and previews.

    well, i would think that power users declined your invitation to the customer experience improvement program, and therefore those usage stats you mention may be skewed the other way. i know i never joined.

  32. Dylan Greene says:

    People don’t customize Office because it’s too complicated.  

    Questions that should have been answered:

    1) Why is it so difficult/impossible for a documentation team or a team of lawyers or an author and editor share the same customized toolbar?  

    2) Where’s the offical Microsoft community site for sharing customizations, toolbar recommendations, and safe macros and templates?  

    3) Why is it so easy to move a toolbar where it shouldn’t be but so hard to discover how to remove toolbar icons I never use?  

    4) I just made a change to my toolbar and i don’t like it.  Why doesn’t undo work?  I should be able to undo toolbar changes just as easily as document changes.

  33. keepITcool says:

    Well…  I’m the developer for a commercial excel based tool. I was / am critical, BUT I do like the Ribbon.

    There’s no design tool yet. It’s like building HTML with notepad. It takes time to select appropriate controls and containers, and getting familiar with the hierarchy and the properties in the schema. Then typing, saving, opening, editing, evaluating. Alas.

    You DESIGN a Ribbon’s Tab, where you implemented a commandbar.. It took several days to decide which controls to use where and how they should appear on the tab.. I’ve spent the weekend creating my icons.(14done, 8todo)

    This week I’ll be starting on a ton of  callbacks. Most of it is just tedious repetitive implementation. (all labels, screen and supertips need table lookup to implement multilanguage). The "Push to Pull" change reqs a change of mind but it makes sense. I was very worried about the impossibility of "insession creation". As I go and learn I’ve found that I CAN fill my dropdowns and galleries at runtime. I probably wont even NEED Dynamic menus.

    It’s tedious.. but you can make it work and look terrific.

  34. Centaur says:

    > Unlike in past versions of Office, we allow you to

    > choose your color scheme directly in Office 2007. You

    > can change between the three schemes from the first

    > page of the Options dialog box.

    Whatever happened to the good old Desktop control panel applet, blessed by the Designed for Windows Logo specification?

  35. Phillip Reeves says:

    This is the post I really needed to read.  Thanks Jensen.

    But as an Excel developer who builds highly customised applications using Excel as a base, these decisions make things extremely awquard for me.  And before I’m accused of moaning as a member of a tiny proportion of users, I should add that my application is used daily by 150+ people in my company, and I very much doubt that I’m alone being in that kind of position.  It is highly unlikely that my bank will upgrade to 2007 precisely due to it’s lack of customisation.  And that’s 10000+ users.

    It seems to me that 2007 will cater superbly for home office/small business use, but far less so for professional corporate use where the majority of users will at least have several tailored proprietory add-ins, and many will have full Excel-based applications that totally tailor the UI (like mine).

    If I could just get rid of the ribbon/QAT/minibar, build custom menus and tailor right-click menus PROGRAMMATICALLY, then I’d be a happy camper.  Office 2010, perhaps?

  36. Jim Rech says:


    Re customizing context menues, this still works:

    Sub AddClearAll2CellMenu()

     On Error Resume Next

     With CommandBars("Cell")

       .Controls("Clear All").Delete

       With .Controls.Add(, 1964, , _

        .Controls("Clear Contents").Index + 1)

        .Caption = "Clear All"

       End With

     End With

    End Sub

  37. Stacia says:

    Jensen, do your numbers include LORG customers and their customizations? Is there a breakdown of the number of customizations made by end users vs. those made by administrators deploying Office to <n> users?

    In other words, if one IT department makes 15 (non-programmatic) customizations and deploys that configuration to 10,000 desktops, does that count as 10,000 customizations? Or are you just counting end user customizations with these numbers?


  38. MDS says:

    We build cars.

    Only 2% of our cars have a manual transmission.  Only 15% of those are turbo-charged.  Building turbo-charged, manual-transmission cars makes it difficult for us to build standard cars quickly and cheaply.  For these reasons we will stop making turbo-charged, manual-transmission cars.

    Oh and by the way, we have a monopoly on cars.  Those professional drivers who make a living racing cars will just have to use an automatic transmission like everyone else.

  39. Jeff says:

    Excellent post, Jensen.

    I guess I am a power user; I have customized my Word environment extensively, but in all cases this was to bring deeply buried commands up front-and-center where I could find them and access them quickly. In some cases I have added buttons for commands I couldn’t locate in the menu structure at all (e.g. the Style Separator).

    I have not been able to try the Beta, but from what I’ve seen it looks like the new interface makes a sincere effort to make all commands more accessible. Between this and the Quick Access Toolbar (brilliant idea, btw) I think the new interface will make me very happy, and I will be relieved not to half to deal with the tedium of customizing my UI (and the difficulty of explaining how to perform routine tasks to other users who do not share my customizations).

  40. Ruben says:

    Ah, I must have missed that tidbit about the Customer Feedback function.

    But, which SANE sysadmin would allow a rogue application to phone home just to let TrustUs Corp collect user data?

    Does anyone really know what the apps send back? I mean, what they *really* send back? I know I would never, ever allow that to happen. And yes, I would block Office without a blink. I *have* blocked Office, come to think of it. That’s a sane thing to do if you give a rat’s *ss about privacy.

    So that pretty much proves my point. Lies, damned lies and the Customer Feedback Program.

  41. Mark says:

    I’ll add my voice to the small chorus here of those who think your data might be skewed slightly in opposite direction to the one you assume, because power users are the most likely to disable the CEIP. I know I did it on my own 3 copies of Office because I had privacy and performance concerns. Average users are far less likely to share those concerns and, I’m guessing, are far more likely to leave CEIP activated.

    Perhaps a little rehabilitation PR campaign for CEIP is in order. If I could be convinced there’s no downside, of course I’d like my usage patterns taken into consideration.

    But to convince me there’s no downside, Microsoft would need to stop doing things like Windows Genuine Advantage (to name just the most recent doubt-inducer).

  42. A User says:

    Jensen – I really do appreciate your explanation of the thinking here. I find the 98% measure approximately believable because it is close to my anecdotal observation of the number of Excel users who have never written a formula with parentheses. For my totally impractical two cents worth, I have often wished there was a separate application for the large number of people who just want a grid-based report layout tool. That way they might not hold back the application for those of us who, you know, work with numbers.

    John – This power user has not upgraded to 2003 either, but for a different reason: it is incompatible with a third party add-in that I would rather not spend money replacing while it works perfectly well. Methinks 2007 will encounter even more resistance for the same reason.

    Tim – Microsoft is not my mother, thank God. I do call my mother regularly, but I would not let her put spyware in my office either. This is not because anyone in the world suspects her of malice, but because she is demonstrably not competent to manage the risks.

  43. Frank Eskridge says:

    Good grief! I had no idea so many people customize and feel so strongly about it! I think I and several other of my fellow employees (we are tech writers) are pretty advanced Word users, but no one I know has customized more than one or two commands. For one thing, we often work on different machines. For another, we are often expected to support the less expert users in our office, whose copies of Word are not customized at all.

  44. Carrie says:

    I will refrain from flaming, no matter how much I want to. I will repeat as a mantra, "It’s okay, I can learn to use the RibbonX-based customization." Breathe. Breathe.

    Apparently, 98% of people are sheep, and 85% of those who are not sheep are not easily annoyed. That doesn’t mean that what you’ve done deserves self-congratulation.

    I was moderately gratified, on viewing the home ribbon the first time, by how much it resembled my custom toolbar. Except that my custom toolbar has all the functions I want, and yours doesn’t. Why in God’s name would I want to format a number that needs commas for formatting to two decimal places? Why? Why? If it needs commas, it doesn’t need decimal places, except in exceptional circumstances. So to get commas with no decimal places using the ribbon, I need three mouse clicks instead of one. My menu has commas, no decimal places as a format on a drop-down menu with the other number formats I use frequently. I could put that button on the QA toolbar, but then it wouldn’t be grouped with the other number format buttons. That sucks.

    I’m done now.

  45. Tim says:

    A User: Aw, muffin, don’t want to be included in the survey, but also want to be counted. Pick one, and suck it up.

  46. A User says:

    Tom – I respond to surveys. This is not a survey, it is spyware.

    Jensen – Do you have a policy on trolls?

    Bye bye.

  47. GregM says:

    If the dialog asking you if you wanted to enable it had said:

    "Microsoft is planning to completely rewrite Office, and is going to use this data to determine which functionality can be dropped because it is very rarely used."

    then there would probably have been a lot more power users enabling it to make sure that their favorite functionality didn’t get dropped.

  48. Ed Bott says:

    My chief complaint about the Quick Access toolbar is that the icons are way too small. My eyes aren’t as sharp as they used to be, and I really have to squint to see what’s what on the QAT. Everything on the ribbon is clear and relatively large and easy to read. Why is the QAT so small? Why not a small/med/large icon option?

  49. Tim Dawson says:

    What did you expect them to do with the CEIP data? Just pass it around the office for amusement at what some people do with Office?

    Seriously, a lot of "power users" think they are so clever by informing those that misguidedly listen to them that there is some big conspiracy, Microsoft is evil, yada yada. Of course Microsoft isn’t stealing your personal information – you think a company of that size could get away with doing that? Well, you probably do. Conspiracy theories are fun, after all.

    Ah, sysadmins. Why do I picture the comic book guy when I think of you?

  50. Francis says:

    I second the calls for the Windows color scheme to "show through" in Office.

    Why not give us a "non-themed" option in Office? I.E., title bar, window frames, etc.–all UI elements (with the possible exception of the ribbon)–drawn by Windows instead of Office.

    Using Office 2007 with Windows Classic leads to an uncomely hodgepodge of window styles.

  51. Mikeslax says:

    The idea that Office 2003 data is skewed towards power users seems, well, skewed; it is turned on by default so *all* users who have it enabled have no idea how to turn it off.  The comments that by turning off Microsoft’s spyware (what else would you call a service which monitors keyboard & mouse input?) power-users deserve the new UI’s lack of customization seems short sighted.  For example, enterprise users will not want thousands of users sending corporate information over the Internet to Microsoft, not only because of potential security concerns but because of the increased network traffic and support calls when the service causes Office to crash.

    I understand wanting to limit easy customization – I work at a help desk as a specialist in Office and actually like the way 2007 manages the UI.  I know a number of people have expressed concern about the lack of floating toolbars, but I, for one, will not miss them.  Jensen, I’d also like to be fair to you – you are doing a great job that, as Abe Lincoln pointed out, is going to annoy a number of people no matter what you do.

    However, with almost 1,000 users I know without UI customization 2007 will not work in our environment.  Not only do groups within our company have specific needs that cannot be addressed, users with accessibility concerns are left out in the cold with the new UI.

    Another concern is the inability, even with RibbonX, to customize the Ribbon.  That’s hubris, in my book, because you are assuming the organization of the Ribbon – placement of groups on the ribbon, inclusion/exclusion of tools in the groups, etc. – is perfect and needs no adjustment.  Users are not all the same and access features in different ways.  

    The sad thing about locking down the UI against any change is it makes the user an incovenience to be kept in the corner while it dolls out features it feels like.  Everything mentioned regarding why people do not customize the UI has one simple solution that does not involve locking everyone out – training.  Having provided Office training for 6 1/2 years, I have seen the positive response of inexperienced users when they learn how to customize Office, and how to avoid accidentally customizing it.  


  52. Patrick Schmid says:

    Mike: You can customize the Ribbon using RibbonX. There are a few limitations though:

    1. You cannot modify MS groups. In a corporate scenario though where you have total control over a user’s Office, there is nothing that stops you from making a new group that replaces an existing MS group. The issues arising from this (incompatibilities with newer versions, add-ins) can be dealt with sufficiently in such a corporate environment.

    2. You cannot reorder MS groups on their original tab. Let’s say you’d want to change the order of groups on the Home tab. You cannot do that. However, you are free to place the Clipboard MS-group anywhere on any other tab.

    I can’t think of anything else right now that you cannot do with RibbonX. What do you want to do, but seem unable to do with RibbonX?

  53. Soren says:


    I have been looking at the new XL2007, and I like the extra rows and colums very much – especially if it will also handle larger files than today…

    The new UI is however not my taste – and worse it seems that the ability to change and add to the menus (all menus are now gone!) is no longer available! Maybe good the the mouse users, but not for a number cruncher like me, who mainly uses the keyboard (much faster, no pain in my hand or shoulder) – its a disaster! You can still use the keyboard (now as easy, but you can), but all my enhancements to be able to work faster and with more extra commands/functions is gone!

    It seems XL2007 will be able to take advantage of extra CPU cores – fine! And a lot of other functionality as well!

    But if the UI and the ability to enhance a menu structure doesn’t come back in the final version, I think I will stick to XL2003!


  54. Mikeslax says:

    Patrick:  I understand I can add items to the Ribbon, but by customize I mean modify the structuring of the Ribbon – as you state, you cannot move groups around, or replace items within a group.  This makes the Ribbon a one-size fits all solution.  

  55. Patrick Schmid says:

    Mike: As I said, the limitation in terms of moving only applies if you want to reorder groups on tabs. If you want to change the structure, you can do that. For example, I don’t like the WordArt Styles group on all the contextual tabs (e.g. Drawing Tools) of PPT. Instead, I rather have the basic text formatting tools there. That is easy to do with RibbonX. What you cannot do e.g. though is to put the Clipboard group at the far right of the Home tab.

  56. Mikeslax says:

    Patrick: I’m not only talking about the inability of rearranging groups on the Ribbon – already a serious problem – but the inability to customize the built-in groups themselves.  For example, the font group in Word 2007 has 14 items, any number of which may be useless for a given user, not to mention impossible for a user with vision problems to read because the icons are so tiny.  The only way to customize the group is to hide the built in group (which is strongly discouraged) and recreate a new group from the ground up.  For example – your style add-in (which I really like!) will, every once in awhile, just turn off; I can re-enable it easily enough, but it highlights the problem with attempting to hide a MS ribbon group.

  57. Patrick Schmid says:

    Mike: Interesting that the add-in will turn off. I wonder what the problem could be. It could be something the add-in does itself and not something Word does. Hence I wouldn’t take it as illustrative of problems with add-ins.

    I am trying to remember right now whether you can remove controls of an MS group. I think though you can set the visibility of MS controls in an MS groups. So you could hide some of those 14 items.

    I am well aware of the implications of being unable to truly customize an MS group. I have come to just accept it by now though.

  58. Kraig says:

    Jensen, Microsoft must not forget about thier users!  2% customization you say?  I am proud to be part of that number! You need to make ALL your users happy.  It doesnt hurt to leave things as they are for us customizers. And the people who don’t?  Well then fine…it works the way it is!

  59. Jamie Hanrahan says:

    No, Jensen, I don’t agree.  I don’t think the rationales presented are valid, either.  

    The Office app I spend the most time in is PowerPoint.  I use five toolbars (two along the top, two down the left edge, one on the bottom).  They hold over 100 buttons.  I use at least 70 of these very regularly.  Why are they on the toolbars?  Because I want them just one click away at all times.

    (Incidently, I think PPT is a materially different sort of app from Word or Excel. How do those "customization" stats look if you just consider PPT users?)

    FAR too many of these commands will now take multiple clicks.  The QAT is nowhere near big enough to hold them all.

    To add insult to injury, the default "Ribbon" in PowerPoint makes it very easy for me to get to… WordArt.  WordArt!!!  I’ve been using PowerPoint for over ten years and I’ve used WordArt maybe four times total. And not because I didn’t know it existed, I assure you.  

    What are your typical users creating with PowerPoint, that you promote WordArt to such a front-and-center location?  Talks aimed at gradeschoolers?  

    The Ribbon concept has another major loss vs. customized toolbars, besides number-of-clicks:  much of the kinesthetic sense of "where things are" is gone.

    With my current toolbar setup I know that object formatting commands are in this area, object arrangment and grouping commands in that area, etc.  This saves time.  No, I can’t pick out an individual button without looking, by kinesthetic memory, but I can most certainly get the mouse to within a group of maybe 2×2, 3×3 at most.  Then I don’t have to look at a large area in order to find the right button.  

    But with the ever-changing Ribbon, this won’t work as well… because a single area is used for different groups of commands at different times.  This requires a lot more looking at the screen to figure out where things are.  

    The Ribbon doesn’t seem to change consistently, either.  Sometimes I have to click on a tab to get to the command I need; sometimes that tab is already selected.  So even if I remember where the group I need is, it isn’t necessarily there *yet*.  I have to look at the Ribbon tabs and maybe click on one of them first.  More looking at the screen, more decisions, more mouse movements and clicks… instead of simply "point at the command I want and click."

    ( Ask BMW owners how they like the "I-drive" system as a replacement for one-button-per-function HVAC and radio controls. )

    To make things even worse, the same command will appear in *different places* on different Ribbon tabs!

    All this would result in a massive drop in my productivity.  

    Maybe the Ribbon does make it easier to find out aobut previously-unknown functionality; I wouldn’t know.  But it most certainly makes some of the functionality that I use a lot, take a lot longer to get to.  I spend more time dealing with the UI, instead of creating my content.  

    So… PLEASE let me have my multiple, positionable toolbars back.  Multible QATs, if you will.

    And then, let me hide the Ribbon completely.  I don’t want it.

    Finally, I suggest this:

    Very few people customize their toolbars??

    Maybe toolbar customization is one of those features that few people used because… wait for it… they didn’t know it was there!  

    In other words, the usage stats are skewed by the very problem the usage stats are trying to research! (Lies, damned lies, and…)

    Ah well.  As a certain well known character in TV science fiction once said, it seems that "The avalanche has already started — it is too late for the pebbles to vote." 🙁

  60. P. Casado says:

    is it possible to change the "BLUE" background in the new UI (2007 beta) for another color, like the older Office "12" Pre-Release/Beta? from blue to silver?

  61. Patrick Schmid says:

    Jamie: I actually believe that PPT is one of the programs best suited for the Ribbon. I believe so, because if you make good use of the contextual tabs and mainly work with them, you’ll  actually have most of the commands you need at any time just one click away. So you basically get the selection of your 100 buttons that you could use on any particular object you selected one-click away. The annoying exception are text &  paragraph formatting tools, as the contextual tabs feature WordArt also this prominently. You can remedy this situation to some degree by using the MiniBar. If you want the MiniBar to appear for sure, select an object by right-clicking it and then the MiniBar appears on top of the right-click menu.

  62. Don Knowles says:

    I just discovered this blog and I’m glad I’m close to retiring. I’m a power user of Excel, but not a developer. I work for a very large aerospace company, and if the IT staff had any input to Microsoft I’m in trouble. Our IT staff is concerned with managing distributions, they have no idea of what the users want; their main concern is making their life easier, not mine. Limiting the user interface is criminal.

  63. Joe Infinity says:

    " I am well aware of the implications of being unable to truly customize an MS group. I have come to just accept it by now though. "

    Patrick: Do you work for Microsoft? Why do you accept this? I have not heard a good explanation why this shouldn’t be possible, and it is a serious problem for me. Why shouldn’t we be able to add buttons to the end of an MS group? What do you recommend doing, if we want to add a single button to the Home tab? Create a new group for it? Eurgh.

  64. Patrick Schmid says:

    Joe: I don’t work for Microsoft. I have come to accept it as I just got tired of complaining over and over again. I have been at this since November, and eventually gave up and resigned to accept what’s possible and what not.

    And yes, you add a new group with the one button to the Home tab. Naturally, your one button needs to belong on the Home tab and not just be there, because it is a prominent location.

  65. CustomMan says:

    You reap what you sow.

    If you turned off the "I want the product to be better" feature and now its not, well, you had your chance.

  66. Joe Infinity says:

    Patrick: FYI I have an extra case option ideally that i’d like to add to the change case menu. I have a styles menu I want to add to the styles group, i have bookmark tools i want to add to the Editing group. I guess i have to add all these in a group on the Home page named after the name of my program, but that’s not descriptive of the buttons functionality. I also have other single buttons I want to add to groups on other tabs. Have you had an explicit response from MS telling you they will not enable editing of groups? I only found out about this recently, so sorry if your been bored of this since November!

    I suppose I can hide existing groups, rebuild a copy of them with my addition. I know that would mean my new groups won’t collapse. Would you recommend / not recommend this?

  67. Patrick Schmid says:

    Joe: I’d go with the single group named after your add-in on the Home tab. This provides the best identification that the functionality is provided by your add-in and not MS (meaning that when users have a problem with it, they actually go to you and not MS). More importantly, replicating and hiding MS groups gets you in trouble with other add-ins which might rely on that functionality (and introduces some non-deterministic issues)

    For a lot more on this topic and reasons why MS groups should almost never be touched at all, read my RibbonX Style guide post: http://pschmid.net/blog/2006/06/09/20

    I am guilty of actually having written an add-in that hides the Styles group on the Home tab. For an explanation of why I think I can get away with that particular add-in, see this post: http://pschmid.net/blog/2006/06/16/23

    For the latest version of the add-in and its source code, see http://pschmid.net/blog/2006/07/02/27

    MS has been quite firm on keeping their groups off-limits. All requests to that matter have been denied. If MS actually were to open their groups up to RibbonX developers, it would be a surprise as big as if my birthday and Christmas fell on the same day. As my birthday is in September, you can figure out the likelihood of this actually happening for yourself.

  68. Joe Infinity says:

    Patrick: Thanks for your reply. Your advice about people coming to me when they have a problem with my button is very good. Plus, if people like what my buttons do, I don’t want MS to get the credit for it!

    Still not sure what to do with my lone Table button. Groups with one button look awful when the group name gets an elipsis. I’d still like to replace the Insert Comment button with a split button for extra comment options (the menu could have a heading "[MY PROGRAM] Comments") It’s late, but I’ll look at your links tomorrow. Thanks.

  69. Joe Infinity says:

    Patrick: Thanks for your reply. Your advice about people coming to me when they have a problem with my button is very good. Plus, if people like what my buttons do, I don’t want MS to get the credit for it!

    Still not sure what to do with my lone Table button. Groups with one button look awful when the group name gets an elipsis. I’d still like to replace the Insert Comment button with a split button for extra comment options (the menu could have a heading "[MY PROGRAM] Comments") It’s late, but I’ll look at your links tomorrow. Thanks.

  70. Patrick Schmid says:

    Joe: Make one large button in a group and give the group a different name than the button (giving the same name works too). Keep both names short, so that the group really only takes the width of one large button. None of the names need to include your program name, if you make a split-/menu-/dynamic menu button. For those, you can start the menu off with a labeled seperator that gives your program name.

  71. Jamie Hanrahan says:

    > If you turned off the "I want the product to

    > be better" feature and now its not, well,

    > you had your chance.

    One, given the current concern (in some cases paranoia) over spyware and privacy and phone-home behavior and so on, it is completely silly of Microsoft to expect that a representative sample of users will leave that feature turned on.  I don’t know what protocols or whatever are used, but unless it’s pure port 80 stuff a lot of corporate firewalls will block such activity (and some even then).  Further, it seems obvious to me that "power users," those who choose the non-default installs of Office for example, will be among those likely to be aware of "phone home" issues and therefore more likely to turn it off…

    Two, this is not just a matter of a feature here and a tweak there.  If a lot of us had known earlier that MS was going to throw the existing UI completely away and replace it with this "Ribbon" concept, that the Ribbon would not be easily customizable, and in some respects would be utterly NOT customizable, and would arguably result in reduced productivity, there would have been much more hue and cry…

    …and apparently it wouldn’t have done a bit of good, because the people who make these decisions at Microsoft don’t care what we think.

    Three – this is not a democratic state we’re talking about here, where the minority sometimes has to accept the tyranny of the majority because you some things just can’t be done in more than one way. It’s a computer program.  There’s utterly no technical reason why it can’t include customization.  There’s no reason why it shouldn’t be able to slurp up my existing ppt11.pcb file and comply.  It’s just that the wants and needs of the people who might want this are not important enough.

    I somehow doubt that a few more users not turning off the phone-home behavior would have affected this.  

  72. Ed says:

    I don’t know that I would be considered a "power user", but in a world of blind people, the one-eyed man is king.  I know in my corporate environment, (a) any "phone home" option is definitely disabled, and (b) I’m one of a very small minority who do any customizing, though others seem to like the macros I pass on.

    I’ve got some macros that rely on creating a custom menu command, and now this will break?  I’ve got to go through any macro or add-in that relies on this an re-code or lose functionality?  Why do you think we go through the trouble to customize?  To add functionality that otherwise isn’t there!  And while we may do it only on our machines, these are often tools to help many people.  Will they scream at MS when they don’t work?

    Seriously, if the problem is bad customizations performed by ignorant users, then make customization more difficult to access.  But don’t completely take it away, or confine it to a tiny corner.  Give the power to those who are willing to take the time to learn how to implement it.  (Granted, that may mean _I_ get left out for a while, but I’ll come up to speed eventually!)

  73. Manish says:

    I lament the loss of customizability but I think the ribbon is a great idea. I agree that removing floating abilities from the ribbon is also a great idea. I will encourage all "office-for-dummies" based office users to upgrade to 2007 simply because I am fed up of searching for their "lost" toolbars.

    The ribbon provides much needed visibility and improved organization for the common user.  A previous post had complained about losing his keyboard shortcuts. Well actually they are all there and then some. I love the way pressing alt brings up all the letters on the screen under each of the tabs.  In some ways things haven’t changed a lot.  They just look different.

    In response to a previous post the buttons on the QAT are not any smaller than they used to be in 2003. They just look smaller in comparison to the new bigger buttons and icons

    I work in a research institution which has a lot of smart inquisitive people who like to experiment. Surprisingly though only a very small minority of these people come under the realm of power users. Most people are highly technophobic, I am sure have never done any customization and are easily befuddled by “accidental customization”.

    I have a feeling that the minibar is the next thing that is going to annoy common people.

    Here are my suggestions:

    1. Add an option to make the buttons on the QAT a little bigger when docked under the ribbon.  Sort of like the changes that you can make to the quick launch bar in XP.

    2. Have the option to  have upto two  rows of buttons in the QAT when docked below the ribbon.

  74. Manish says:

    I lament the loss of customizability but I think the ribbon is a great idea. I agree that removing floating abilities from the ribbon is also a great idea. I will encourage all "office-for-dummies" based office users to upgrade to 2007 simply because I am fed up of searching for their "lost" toolbars.

    The ribbon provides much needed visibility and improved organization for the common user.  A previous post had complained about losing his keyboard shortcuts. Well actually they are all there and then some. I love the way pressing alt brings up all the letters on the screen under each of the tabs.  In some ways things haven’t changed a lot.  They just look different.

    In response to a previous post the buttons on the QAT are not any smaller than they used to be in 2003. They just look smaller in comparison to the new bigger buttons and icons

    I work in a research institution which has a lot of smart inquisitive people who like to experiment. Surprisingly though only a very small minority of these people come under the realm of power users. Most people are highly technophobic, I am sure have never done any customization and are easily befuddled by “accidental customization”.

    I have a feeling that the minibar is the next thing that is going to annoy common people.

    Here are my suggestions:

    1. Add an option to make the buttons on the QAT a little bigger when docked under the ribbon.  Sort of like the changes that you can make to the quick launch bar in XP.

    2. Have the option to  have upto two  rows of buttons in the QAT when docked below the ribbon.

  75. Kevin Wright says:

    I’m glad to hear that floating things (like toolbars) are going to be appearing less often.  What I would REALLY like to see is a non-floating find/replace dialog.  I am forever moving this dialog out of the way so that I can view the document below it.  Take a look at how Firefox has the find dialog as a bottom-docked bar that pops up when needed.  

  76. Keith Green says:


    Thanks for the explanation.  I have requested via blog, frowney face, and now again that you re-consider your position on this particular subject.  

    Granted, I am certain that I will not offer arguments that are new, nor none more eloquent that those offered already on this site.  

    However, in the unfortunate case that you may be more interested in the NUMBER of arguments, and less in the LOGIC of arguments, I shall add my voice once again.

    1) Having been a user for over 10 years of Previous versions of office, (although not 2003) I have never been monitored(to my knowledge) nor have any of the thousands of users in our company.  

    This fact does not indicate my desire for MS to dis-allow essential features of a Microsoft product, it is simply evidence that I do (and my company does) not want another company to monitor my usage without clarification of what data will be collected and why.

    With this clear in my mind, and assuming that there are many 10 year users who would feel the same or even more strongly, it does not appear that you could have a clean study of your user base with the CEIP.

    2)  You state… "Is additional customization against the design philosophy of the Ribbon? No, provided that we could add it in such a way that it added no additional complexity for the vast majority of people who aren’t interested in it."

    I am so glad to hear you say this…. Simple solution….. If you are concerned that the vast majority user will ACCIDENTALLY customize…. Default a checkbox under "options" to "Allow Customization to Ribbon – No", and let those of us who would like to work efficiently CHECK "YES" and get busy customizing.

    3)  If the customizers are a small percentage of your survey, doesn’t it still represent the "power users?"  

    Assuming that a power user is the person who uses the app for the most business, has taken the most time to learn the app, and has figured out that it can be made more efficient by customization, it does not make sense to ignore the power users requests because the average user did not know how to customize.

    4)  You elude to "future versions".  Please don’t make us wait for another upgrade to get customizaton back…. I am on my knees begging.

    5)  Finally, …."You can change between the three schemes from the first page of the Options dialog box. People always want more color choices…"

    Did your survey somehow reveal that this feature is worthy of your programming resources above customization?  What was the percentage of users that changed the schemes in office 2003? (I assume it is 0%, since it was not a feature??) How did it improve the USABILITY of the app?  How will you prevent the vast majority of users from accidentally changing their scheme?  Oh and one more… be cautious when someone uses "always" in an argument.

    PLEASE PLEASE PLEASE let us customize Ribbon without having to learn to program in ribbonx!

    Keith Green

  77. Jamie Hanrahan says:

    > Assuming that a power user is the person who

    > uses the app for the most business, has

    > taken the most time to learn the app, and

    > has figured out that it can be made more

    > efficient by customization, it does not make

    > sense to ignore the power users requests

    > because the average user did not know how to

    > customize.

    Excellent point, Keith.  Let’s expand it a bit:  It isn’t clear exactly what MS is counting but it seems to be something called "sessions."  I assume that this means "startups of any major program in the Office suite."  

    What if they instead had counted keystrokes? Or documents saved?  Or slides created?  What would the "percentage of work done with customized toolbars" have looked like then?

    "Sessions" seems to me to be a remarkably poor way to determine the usage of a feature so obviously skewed toward power users.  Your power users are likely the ones who will be creating the most content per "session."  

  78. Jamie Hanarahan says:

    I wasn’t going to post again until others replied, but since it’s been over a week… here is a "ha ha, I was only serious" prediction for the future of Office 2007:

    Large numbers of people with heavily customized toolbars in Office 2003 and earlier will simply not move to Office 2007.  It’s just too much of a productivity killer for all but the "Office for Dummies" crowd. So would be learning RibbonX.  So we’ll just stick with 2003.  

    Of course since these users were such a small percentage of Office users, Microsoft will not notice.

    Later, Microsoft gathers O.2007 usage statistics and finds that the percentage of users with customized ribbons and QATs is even smaller than the percentage of toolbar customizers in previous versions.

    "Ah ha," they will say.  "We were right! Customization is not that important, and furthermore we must have gotten the ribbon design near perfect!"

    I think someone really needs to read _How to Lie With Statistics_, in particular the part about "the sample with the built-in bias."

  79. Tim Teichman says:

    Personally, I loathe the ribbon.  I can never find anything, and worse, I can’t customize it.  To me, the groupings don’t make any sense, and the condescending UI is infuriating (for example, try to close the app with the close box and only the active workbook closes. Then you have to do it again.)

    I develop add-ins and custom workbooks, have reviewed the posts on how to customize the ribbon with code for hours, and can’t for the life of me figure it out.  It appears that what was once a simple VB command in a code module in VBA has turned into a completely unworkable situation for someone distributing custom workbooks that need an interface to launch code.  What could Microsoft possibly be thinking?

  80. Chuck E. says:

    All I want to be able to do is paste text without formatting, and without having to click "Paste Special" (whether on the ribbon or not) to get there. That’s a pan-Office request. Having "Paste as Text" on the right-click menu would be nirvana to me.

  81. jeffreywsmith says:


    I’ve been reading a number of your blogs on the new UI and trying to understand both Microsoft’s perspective as well as those of other users who have commented on same. As I see it, there seems to be a great divide between some of the UI team’s direction and decisions thus far and many in the Microsoft user and partner community who are concerned with both: a) the claims that the new UI makes on their screen real estate; and b) the reduction in the built-in customization abilities and simultaneous limitation of the new object model that is exposed to us with the tools that are included in the Office 2007 suite.  I think the issues here are really important so I’ve taken the time to do some analysis – I hope you’ll (all) bear with me while I try to explain why I think some additional consideration needs to be given to these issues.

    Some of what I have to say may not be welcome because if it’s correct, fixing it might cause some delay in the release schedule. But if the Beta feedback is truly going to inform the design as we’ve been told, isn’t even a delay worthwhile if it helps to both: a) ensure a more successful introduction of Office 2007 and all the good things it brings); and b) respect the rights of your customer base to control their screen area and their investment in your prior technologies to optimize their productivity?

    Since I have built (and expect to bring to market within the next 1-2 weeks) an alternate user interface for Excel precisely because I wanted a more user-friendly and productive Excel environment, I can understand why Microsoft might take this opportunity to address some of the same issues themselves.  From this perspective, I can also appreciate that the process of building a UI necessarily involves a series of compromises to be made.  But here’s the rub:  Customization is one of the primary means we have to productively manage the delicate balance between our document workspace and the room allocated to our tools, all seen through a prism of our own individual (and differing) visual and mental acuities as well as the hardware we can afford to manage all these variables.

    Using a military analogy, users are the “Generals”, the Toolbars are their “weapons” and the usable screen space is their “battlefield”.  Now, we Generals need SPACE in our battlefield to maneuver our "troops" (to develop our data, produce our analyses, render our presentations, communicate same with others, and all of this, part and parcel, is a vital means of understanding our operations, ordering and controlling our objectives in order to achieve our missions).  And in order to do that productively, we need to be able to SEE our work without having to constantly scroll the screen which adds to our “casualties” (ala the relationship between excessive mouse and keyboard clicks and carpel tunnel syndrome) and greatly detracts from productivity (and especially so in more complex documents) because of: a) the time and clicks it takes to scroll; b) the spatial distraction scrolling causes; and c) because we have to remember what is in other non-visible areas of the document that impacts what we can see.  So in the "battle" for screen space, there is a constant struggle (and need) for balance between the space allocated to our tools, and the remainder left in which to do our work.  The end result of this whole design process becomes the UI in which we have to contend to productively accomplish our missions.


    That we (users and partners) have embraced Microsoft Office as a platform of choice to help us achieve this balance is a testament to how much we value the flexibility – and stability – Office has offered us these last 10 years.  However, as others have mentioned here and in related blogs, I believe the CEIP data used to drive the new UI design is flawed, and summarizing all those reasons here in one place and offering a related thought or two (as well as a possible way to remedy same), here’s why I think that:

    1. First, as I understand it, the CEIP data has only been collected from Office 2003 customers. Jensen has stated that there are 450 million paid Office customers of all versions. He’s also said that there are about 100 million (22.2%) such Office 2003 customers in the CEIP database (but for all the reasons listed at 2., 3., and 4. below, that’s not all of their 2003 customers). To my knowledge, Microsoft has not made public how many Office users there are of each version and what information is publicly available is slim to none. At this site:

    (http://www.fontstuff.com/comment/comment03.htm#survey2 )

    …  on August 1, 2006, only 16% of voting Office users are supposedly using Office 2003 … although as the site makes clear, this is not a valid, random statistical sample. Regardless, even using the 22% represented above, Microsoft believes (as detailed in 6. below) that the CEIP data probably overstates the screen resolution of their installed customer base because of the expected higher incidence of early adopters.  So why should the UI design be based upon only the screen resolution of approximately 22% of their customer base, when their screen resolution is thought to be higher than that of their customer base at large?

    2. Many Office 2003 users have opted out of CEIP because of Privacy concerns and/or because they didn’t understand what data would be collected or how it would be used.

    3. Many others were prevented from participating in CEIP because of corporate firewalls and/or security policies.

    4. Still other Office 2003 users (particularly in some corporate settings) could not participate because they have no Internet access.

    5. In qualifying the CEIP conclusion that: “In fewer than 2% of sessions, the program was running with customized command bars.”, you’ve stated above: “Buttons added by add-ins and templates are also not included”. So, effectively excluded from the data is, I expect, what is one of the most popular and productive means of customization.

    6. You also stated above: “(One metric that causes us to believe that the Office 2003 data slightly skews towards power users even today: Every month, the average screen resolution of people using Office 2003 decreases as more of the core installed base adopts it.)”  Since the Office CEIP data has only been collected from your 2003 customer base, doesn’t this declining average screen resolution suggest that your “core installed base” hasn’t been proportionately reflected in the CEIP data?  That the average screen resolution of those users that drove the UI design was overstated when compared to that of your “core installed base”? That even more of your customers are going to resent the demands of the new UI for screen space, once they are more fully aware of those demands? That this resentment will cause a less successful debut of Office 2007 in the marketplace?

    7. You stated in your Summary here: “As with everything in the Office 2007 user interface redesign, we informed the design both by analyzing the usage data and then by adding a sprinkling of anecdotal feedback.” Don’t all of the above flaws with that data suggest that something more than “a sprinkling of anecdotal feedback” is required to truly measure both the average screen resolutions and UI-customization characteristics of your “core installed base"?

    As far as what the “average screen resolution” really is, I’ve followed your blogs on the "Mythbusters: The Office 12 New UI" and in particular, the "The Ribbon is huge!" discussion on this page:

    http://blogs.msdn.com/jensenh/archive/2005/09/15/467956.aspx )  

    … as well as "The Size of Things"


    Given how important  customization is to the UI, I think it’s important to make sure these numbers are well founded. I’d expect that the numbers used in your earlier blogs have probably changed some since those blogs were written and the product has since evolved through the various Beta cycles.  At any rate, as compared to the present Beta 2 release with all updates applied, I’ve been doing some “pixel-counting” to try to determine what the Office 2007 UI leaves as Screen area for Excel spreadsheets. I’ll share what I’ve found across the 5 most popular* screen resolutions in hopes that more specific data may advance these discussions:

    Test conditions for both versions were (as were used in Jensen’s earlier studies):

    • Quick Launch Bar – Off

    • Task Bar limited to one row

    • Display Setting: Windows XP (unmodified) default Theme

    • Excel Window Maximized

    • Worksheets were formatted with a 10 point font (Note that this produced a Row Height of 17 pixels in both 2007 and 2003 – apparently, Jensen’s use of the older Office 97 produced a Row Height of 18 pixels in that version which allowed fewer visible rows**).

    * = Since I couldn’t find any published Microsoft data on Office users screen resolutions, I determined the “most popular” resolutions and some other statistics by consulting thecounter.com web site at


    (I used the June 2006 survey as it was the most recent when I started this analysis)

    ** = I used Excel 2003 to compare with Excel 2007 because it equalized this variable (in fact, Office 2000, 2002 and 2003 all have a Row Height of 17 pixels with a Font Size of 10), and because it seemed appropriate to use the most recent version rather than one that was 10 years old.

    The survey data at thecounter.com’s is captured from the web-surfing profiles of all users that allow their machine characteristics to be detected when visiting monitored sites.  So, while their data is gathered from more than just Office (or Excel) users, with nearly 95 million known user screen resolutions in their June 2006 dataset, I would think that such a large sample would approximate real world usage of Office users at large, too (and may well be superior to the CEIP data since it came from only Office 2003 users).  Interpolating June 2006’s thecounter.com data, I’ve also computed the present "theoretical (mean) average user’s screen resolution to be 1088×816, and I’ll use that “average user” to sum up some of these comparisons.

    It’s appropriate to measure more than just the Ribbon vs. the Toolbars it replaces in earlier versions because that’s not the only thing that’s changing in 2007 … the Formula Bar is also bigger and so is the area below the last row.  Taking all this into account, and focusing on the Vertical dimension because most documents are vertically-oriented and because that’s the one that concerns most people, I found that Office 2003 allows 2 10/17 (2.59) more rows to be visible than did Office 2007 at each of the 5 screen resolutions.  As a percentage of the vertical screen area available in 2007, though, this 2.59 row advantage varies across resolutions with those at lower resolutions feeling the most “pain”.  For an 800 x 600 screen resolution, the 2.59 additional rows is 12.8% of the rows available in 2007. For 1024 x 768, it’s 8.6%; 1152 x 764, it’s 7.2%; 1280 x 1024, it’s 5.7% and 1600 x 1200, it’s 4.7%. Conversely, the same numbers represent a loss of 11.3%, 7.9%, 6.7%, 5.4% and 4.4%, respectively from the rows that were available in 2003. The “average user” has 8.75 % more visible rows in 2003 and loses 8.00% going to 2007.

    If we take into account the additional 26 pixels (1.53 rows) required by deploying the Quick Access Toolbar (QAT) more productively below the Ribbon (where users’ favorite tools can be reached with less Vertical mouse travel and more tools can be stored), the combined impact of the 2007 UI is worse still (an additional Toolbar was added to the 2003 Excel these to equalize the comparison with the QAT below the Ribbon in 2007).  In terms of lost rows in 2007, the total 4.12 rows (2.59 + 1.53) is 19.3%, 13.2%, 11.2%, 8.9% and 7.3%, again respectively, running from the 800 x 600 resolution to the 1600 x 1200 and computed as a % of the rows available in 2003. The “average user” would lose 13.4% of their visible rows going to 2007.

    Another way of looking at this is to determine how many pixels are devoted to the Vertical UI (Non-Document Area).  In 2007, that number, 221 (or 28.8% of the maximized Excel Window), was a 24.9%  increase from the 177 vertical pixels (23.0%) allocated in 2003.  If the same comparison is made with the QAT productively deployed below the Ribbon (see above), for 2007, that number, 247 (or 32.2% of the maximized Excel Window), was a 21.7%  increase from the 203 vertical pixels (26.4%) allocated in 2003.

    Is a 24.9% (or 21.7%) increase in the space required by the UI a big number?  Is the “average user” going to miss being able to see 13.4% of his rows?

    I’d suggest that these numbers are big enough and the impact wide enough people (77.1% of the known resolutions of June 2006 users in thecounter.com survey were using 1024 x 768 resolution or less), that ensuring the utmost is done to maximize customization abilities ought to be very important to Microsoft.

    I believe there are some viable ways to remedy the flaws in the data (and perhaps to fix the underlying problem, too), so I’ll offer a thought or two about what might be done:

    First, Microsoft has to consider that the CEIP data might be wrong, that it might not accurately reflect either the average screen resolution or the degree of customization employed by the full spectrum of their “core installed base”.

    Then, ask, what can be done to fix it?  In this respect, I would suggest that the aforementioned  “sprinkling of anecdotal feedback” be supplemented by a more rigorous study designed to deal with all the aforementioned flaws of the CEIP data. I would suggest a random, statistically valid survey, perhaps augmented by further site surveys.  Using the tools found at:


    … such a sample size might look something like this:

    – Population of “paid Office customers” : 450,000,000

    – Acceptable margin of error : 1%   (we all want this to be right …)

    – Confidence level : 99%   (we all want to be confident this is right …)

    – If the “Response distribution” (expected results) = 2%, the sample size = 1,301

    – If the “Response distribution” (expected results) = 10%, the sample size = 5,972

    – If the “Response distribution” (expected results) = 20%, the sample size = 10,616

    – If the “Response distribution” (expected results) = 50%, the sample size = 16,587

    Obviously, the expected results drives the sample size more than any of the other factors but even the largest sample size above is only one out of every 27,130 paid Office users.  One of the larger numbers here is probably appropriate since: a) the site’s discussion of “Response distribution” indicates that: “If you don’t know (the expected result), use 50%” as it provides “the most conservative choice”; and b) the CEIP’s data supporting the conclusion that “fewer than 2%” of user sessions are “customized” is so disputed.

    Chief amongst the questions to be asked in such a survey ought to be:

    1) What is the users screen resolution?

    2) Has the UI been customized, particularly including those customizations that the CEIP data is sorely lacking?  Here, I’d suggest the survey at least measure those additional customizations provided by add-ins and templates since (I believe) they are one of the chief means of customizing the UI, and perhaps consider even the customization of toolbars by relocating them vertically since that appears to be a bone of contention in these discussions, too.  I’d venture a guess that the answers to these questions might be different across the various applications in the Office suite, so that might be an additional (and appropriate) variable to consider, and may have further implications for sample size (and UI design).

    With this more rigorous information in hand, ask what can be done to optionally marry the already developed classic UI of Office versions past to Office 2007 in ways that don’t invalidate the investments in customizations, education and training and existing technologies that so many of your customers have made.  Not to oversimplify, but perhaps all that’s necessary is a means to allow users to choose to switch to the previous UI, while meeting your above-mentioned standard for allowing additional customization when: “we could add it in such a way that it added no additional complexity for the vast majority of people who aren’t interested in it.”

    I think it’s important to understand why so many are so vocal about these issues – beyond the military analogy of the screen area concerns discussed above and how that will impact individual users, there are other significant costs to be considered, too.  As it is presently constituted, the new UI presents severe challenges to not only many CEIP-excluded users, but to the power users and third party developers at large who have heavily invested in learning the technologies to both become optimally productive personally and to build tools and applications that enable all kinds of users and all stripes of organizations to maximize their efficient use of the Office productivity suite.  Their collective efforts have been built on the relatively stable foundation of the VBA language and customization interface that has characterized the Office suite since Office 97.  The “one-size-fits-all” 2007 UI turns unknown amounts of those efforts on end, at best, requiring significant re-tooling and re-education costs, and at worst, invalidating uncounted applications and user productivity shortcuts.

    Jensen has stated above that we can “Use RibbonX XML to customize Ribbon content through a COM Add-in” but even that doesn’t replace all the capabilities we’ve had in previous versions and as others have pointed out, RibbonX is a vastly more complicated language, requiring expensive software tools to use, thus effectively removing the full-set of UI controls from both the built-in UI toolset and the reach of many users who don’t have the skill sets or funds to utilize RibbonX.  As compared to the VBA and customization tools built-in to the Office suite since Excel 97, this will severely limit the number of available programmers on hand to make user-requested changes to the UI and severely restrict the ability of power users to make their UI’s more productive at will, in all the ways that make sense. As with all such matters of supply and demand, if you restrict the supply of resources (qualified programmers, with the skill sets and tools) available to meet customer demand, the prices for such resources will rise accordingly).  Organizations used to being able to maximize productivity of their employees by customizing their UI are going to see their costs rise substantially (and that’s on top of both the higher costs for Office 2007 itself and the corollary costs of buying bigger monitors and/or higher resolution video cards for those users whose present hardware can’t productively support the demands of the new UI.  We who are protesting fear a loss in our productivity, the invalidating of our code and/or customizations, and, in an age of an increasingly demanding global economy whose “productivity wars” require ever more personal sacrifice, a squandering of our resources, both personal and corporate that this new UI will require. I hope the powers that be will take another look at these issues.

  82. Patrick Schmid says:

    I have made an alpha version of my add-in to customize the ribbon available. See more details in this blog post: http://pschmid.net/blog/2006/08/01/35

  83. Anonymous says:

    Jeffrey Smith:

    Very good and hard job!

    With the last improvements done in the Ribbon, it is much better than in beta 2 but I still think that it would be considered an evolution for the menu bar not for both menu and tool bars.

    With the Ribbon the great majority of the user will not need toolbars anymore, but many, so many that MS think and like you job stands up, will miss toolbars and mainly their customization feature. Then I think the toolbars could continue like 2003 but all hidden as default.

  84. jeffreywsmith says:


    Thanks for the praise. I’m guessing your comments are related to "DataSmiths Evolution", the "alternate user interface for Excel" that I mentioned in my comments above (for anyone interested, a sneak preview is available by clicking on the URL link in my signature). It’s interesting to see that some of the innovations we have included in DataSmiths Evolution are also showing up in Office 2007.  For example, similar to Office 2007, we provide:

    a) Toolbars that expand with the user’s screen resolution so that we provide an "environmentally aware" user interface that takes advantage of (and even optimize) the user’s individual hardware resources to deploy still more useful single-click tools. A similar methodology is used to deploy a few selected (but choice) tools that are specific to later Excel versions so that as users upgrade they’ll find the best of the new version’s tools front and center (this all happens automatically – the software determines the operating environment and automatically loads the correct set of tools). Yet since the "core tools" are relatively constant across the three versions presently supported [Excel 2000 – 2003], we are able to deploy a relatively consistent interface, seamlessly across all three versions while expanding the core tool set for those with the hardware to support it;

    b) Our own version of the "Quick Access Toolbar" is also an optional part of DataSmiths Evolution.

    We’ve also introduced any number of other tools as well, and organize them in a consistent manner so users’ tools are not constantly moving around on them. To “prove our case”, so to speak, we’ve also built-in a "productivity tracker" that will track individual users’ own savings using this product, indicating about a 75% improvement in productivity with a 1024×768 screen resolution (the higher the resolution, the more tools we can deploy and the move savings can result).

    With respect to your comment: “With the Ribbon the great majority of the user will not need toolbars anymore”, I think it still remains to be seen how well the market as a whole will react to the new UI: Many users and partners have voiced dissatisfaction with some of the UI team’s decisions thus far because: a) it encroaches on some of the turf we’ve been able to use to productive advantage; and b) the tools they’ve used (to date in Office 2007 Beta 2) to replace this previous functionality are neither as capable or as accessible in delivering to users the flexibility they’ve enjoyed in past versions to order their UI in all the myriad ways that the (really) uncounted millions of Office customers past (to whom productivity is important) who have welcomed the previously stable UI as a means to make the most sense in the various (and differing) work immediately at hand.

    Too, to be determined, is how well potential Office 2007 customers will take to all the ways the new UI attempts to reveal the tools the users want at any given time, for any given work. Some times, it’s spot on, but these blogs are replete with Beta users’ reports of how many times it misses the mark (closing tabs prematurely when users weren’t yet finished, or perhaps failing to open others they might logically want next) & inconsistent displays or ordering of the same functions when accessed from different tabs.  The fact is, Office 2007 (as did Adaptive Menus and Toolbars before it), embodies a “moving UI”, requiring users to scan the displayed interface to find the tools they want, and they’re not always rendered in the same logical space. I’ve not yet immersed myself in Office 2007 (so I’m no expert here), but I’ve spent enough time there to see these inconsistencies myself and I expect it will trouble many others, too.  All the more reason, I think, for the UI team to study these issues further, to see if they can’t make the classic UI optionally available to all those who have good reasons to choose it.  Quite possibly, this is a “win-win” scenario for both “us” (users and partners) and Microsoft, for whom a large part of their fortunes these last 10 years have rested with the enthusiastic acceptance of their Office innovations in the marketplace.  To do otherwise may well be, as Steven Bullen has said at:


    … to have “thrown the baby out with the bath water”.

    I’ve taken a quick look at how our alternate UI holds up in Office 2007, and on first impression, I expect it could be worse. The good news is that much of what we built to provide a consistent, enhanced and user-friendly interface for Excel 2000 – 2003 still works in Excel 2007. The bad news is the way (as 2007 is currently constituted) that custom toolbars are relegated to the “Add-Ins” tab of Excel 2007, even if they are resident in stand-alone workbooks, or even a user’s “Personal.xls” file and not part of any “Add-In” application per se. If you’ve got a fully populated Toolbar, and open a workbook with custom menus for example (that you only want to see when this workbook is open and to close when the workbook closes), that custom menu gets tacked on to the back of the toolbars in the “Add-Ins” tab of the Ribbon, so users may not even know those additional menus are there (because they’re not necessarily visible).  Equally as bad, VBA code that serves in earlier versions to exactly specify the Row Index and Left positions of custom Toolbars or Menu Bars is not obeyed in Excel 2007. No errors are given (the code is just ignored), but these tools are ordered and positioned in accordance to the beat of a different drummer, it seems.  Perhaps there’s a way to take control of these issues without having to bite the bullet on RibbonX – we’ll see (I’d welcome anyone’s insights into this issue).  At any rate (in the spirit of the aforementioned military analogy), we’ll remain committed to helping our customers marshal their “troops”, to command their “weapons”, to become the best, most effective “generals” possible so that we all might recoup some measure of the increasingly scarce personal time that defines our quality of life.  I just hope the Office 2007 UI is still listening and remains devoted to preserving what’s been so good about the flexibility and customizability of Office versions past so that all of their customers with differing visual and mental acuities, different hardware resources (and budgets) and differing needs can find their requirements still met from the newest version of their productivity suite of choice.

  85. Patrick Schmid says:

    Jeffrey: Thanks for the hint to the sneak preview. When I read your last post, I was quite interested in what your alternate UI would look like. For those who don’t like searching around, screenshots: http://www.datasmiths.net/compare.html

    For me personally, that UI would be overkill and way too many buttons at once. However, I am not an Excel junkie and I know that Excel power users tend to have a lot of buttons accessible with one click. I guess for those, this is the kind of UI needed.

    I don’t think you’ll get around RibbonX. You probably want to make one custom tab that has only one group. That gives you essentially three rows of icons across the screen. Then I’d suggest to place that tab before the Home tab. Whether you want to use startFromScratch and how you are going to deal with contextual tabs (which automatically come into the foreground when an appropriate object is inserted) is a more tricky question though. To get started with RibbonX, I can highly recommend my own blog at http://pschmid.net

  86. jeffreywsmith says:


    Thanks for the pointers on what I might need to do to optimize our Toolbar presentation in an Office 2007 environment.  From what I’ve been reading so far about Office 2007’s restricted customization abilities (largely as explained in the knowledgeable comments you and others have made to some of Jensen’s blogs), I expect your advice is the direction I’d need to go to have the most control over the UI as it currently exists today (however, I’m hoping that the UI team is still paying attention to all of us who are suggesting their work isn’t done yet and that they’ll find the ways to deliver the tools we need to keep us from having to take the proverbial two steps back for the one forward).

    With respect to your comment: “For me personally, that UI would be overkill and way too many buttons at once”, I recognize that this is the first reaction of some people.  However, that perception ought to be informed by two important observations:

    1) First, the Toolbars are delivered in the same (fixed) 3 rows that characterize previous versions of Office (i.e., that’s with the Customization option to “Show Standard and Formatting toolbars on two rows” selected so as to avoid the terribly ineffective “rafted toolbars” that Microsoft has acknowledged as an unwanted “feature” ( http://blogs.msdn.com/jensenh/archive/2006/03/31/565877.aspx ) … because “The order of commands was no longer constant, scanning for functionality was inefficient, and predictability suffered as nothing could ever be guaranteed to be in the same place even from click to click.”).  It’s just that we’ve made maximum use of all the space those 3 fixed rows provide, while preserving all the document workspace that the default Toolbars did.

    2) Secondly, after the Toolbars are calibrated to match the user’s Screen Resolution and Excel version, the toolbars are fixed, so the tools don’t keep moving around.  That allows users to get better acquainted with all the tools at their disposal, so that they are fully in command of all the “weapons” in their “arsenal” (it’s not unlike an analogy of a carpenter’s tools being arrayed on a well-designed pegboard where all of the regularly-used tools are visible and easily accessible as opposed to being squirreled away in a set of drawers somewhere where one has to know that they not only have such a tool but remember where to find it).  As the following text from our page that you referenced above explains ( http://www.datasmiths.net/compare.html ), we’ve gone to considerable effort to provide our users with adaptive documentation that enables them to rapidly adapt to their new UI:

    “While the array of features might initially appear intimidating, this product includes: a) an integrated User Guide which orients users to the organization of their tools; b) "mouse-over" tool tips that appear when the cursor hovers over a given function; and c) our innovative "Adaptive Learning" tool which accelerates the learning curve by allowing users to focus on only those functions they have infrequently (or never) used while presenting expanded descriptions of their associated functionality (note, too, that while this product supports 21 configurations of Excel version and Screen Resolution, these Adaptive Learning aids "know" what the user’s current configuration is and present only those functions applicable to that environment). Combined, these learning tools will quickly bring users up to speed on their new User Interface.”

    Not to “wax on” too much, but I might also point out that our alternate UI is endowed with still more productive functionality than the primary Toolbar functions themselves provide, like:

    a) a Multiple Sheet Protection/Unprotection feature that allows either all Worksheets or only selected ones to be protected or unprotected at one go;

    b) a custom Print Control solution that allows users to design a custom print template applying some 28 different Print setting parameters, effectively creating their own custom Print specifications, including automatic footers with Date/Time and Page Number(s), File Name (and/or Worksheet Name or Path, all optionally on one, two or three lines in case any one of these elements is too long), Portrait or Landscape Orientation, custom Text in the Page Header or Footer, custom margins, and, for Excel XP & higher, the ability to specify the inclusion of a company logo in the Page Header. Users can also specify that this custom Print specification will use (for example) legal-size paper, while our built-in, alternate Print Setting (which was designed to maximize the print content per vertical page while providing a more minimalist automatic footer with frequently-used settings) uses letter-size.  Between the two custom Print solutions, users can build their custom Print template just once and then apply either to the current work at hand by either clicking the indicated Tool, or in the alternate, by invoking the Shift Key when clicking same; and

    c) as another learning aid as well as user convenience (and something I hope Microsoft might yet include in the 2007 UI before the Beta design is finished), we also provide a backwards-compatible ability to switch back to the previous default UI (in case there’s a priority project users might need to finish before they’ve had a chance to learn the new interface, or in the event that they are installing a third-party Add-In that expects to find the default Worksheet Menu Bar).

    There’s more, too, and all of this is in addition to the abovementioned 75% increase in productivity that most users will see (this statistic is understated because we weren’t able to measure how much these additional “bonus” features saved users).

    On balance, I guess the proposition we make is that we’ve built an impressive array of tools that users will come to know and love to have at hand and that spending an hour or two with our adaptive documentation is a small price to pay for a product that will save you 75% (or better) of the unnecessary keystrokes and mouse clicks in all your future Excel work.  But users don’t have to take our word for it, either.  Not only do we include the aforementioned “productivity tracker” that informs them of how many of those unnecessary keystrokes and mouse clicks we have saved them individually, we will also be hosting a free 30 day trial of DataSmiths Evolution on our site within the next 2 weeks, so anyone who likes can have a no-cost look for themselves.

    Thanks once again, for your advice, Patrick and I’ll be sure to take a look at your referenced blog on RibbonX and perhaps join in the alpha testing of your related Add-In once I know more fully what is required to participate.

  87. Patrick Schmid says:


    you are welcome on the pointers. I seriously doubt that we’ll see much change in the 2007 customization system, so what there is right now is what will mostly be in the final product. Jensen has already pointed out a few things that are going to change, but that’s merely fine tuning.

    You mention the hope that MS might provide a "classic UI". MS has made clear from the very beginning that this won’t happen. If you read Jensen’s "Why the UI" posts, you’ll see that one reason for the new UI was that they had no idea where to put all the 2007 features into the old UI. So they simply don’t even have an idea how a classic UI would look like. Debating classic UI support is a mute point. There is just no way it will happen.

    The productivity argument is actually a very interesting one, and one that has been argued a lot about with the new 2007 UI. Interestingly, it generally is made by Excel users. The argument comes down to Excel users performing a vast array of the same tasks over and over again. Therefore, how fast a command is accessible (e.g. one vs. two clicks, travel distance of the mouse) is used as the main measurement of productivity. Your UI maximizes that to the largest extent possible. The new Office UI on the other hand, doesn’t do the greatest when measured that way (e.g. no floating toolbars, not enough toolbars, etc.).

    The productivity measurement that Microsoft uses though is whether users can do things that they couldn’t figure out how to do with previous versions, or whether they can do them better and faster. For the most part, the new UI achieves that goal (as said before, it doesn’t work the greatest for Excel power users, some book authors are struggling with Word and charting is an Office-wide issue). For  specific examples, just read through Jensen’s blog.

    Personally, I like the new UI completely. Nowadays when I open any Office 2003 application, I frown and get annoyed that I actually have to deal with the old UI. I don’t consider myself to be a hardcore power user in any particular Office application. I rather think of myself as a power user of Office in general. Therefore, an interface with everything one click away just wouldn’t do it for me. I’d have to remember what each of the buttons did and where the button was for each application. The new ribbon UI on the other hand, makes it easy for me to find stuff with one look without needing to remember a whole lot.

    It took me a while to get there, but I definitely don’t want to go back. The ribbon UI isn’t perfect, but MS is still working on it. The new PPT Home Tab e.g. got rid of my main issue with PPT that has been bugging me since the first beta. For the things that they won’t get "right" (a very subjective measurement), I develop customization tools.

  88. jeffreywsmith says:


    Re: your comments: “I seriously doubt that we’ll see much change in the 2007 customization system, so what there is right now is what will mostly be in the final product.”

    You may well be right, Patrick, but I think that, for all the reasons listed in my first comment in this blog, the UI team’s reliance on the Office 2003 CEIP data has MIS-informed the new UI design.  If, because of a poorly designed and nowhere near representative “sample” of users’ needs, they don’t really *know* what the “average” screen resolution of their entire “core installed base” is, if they don’t *know* how much their *real* “core installed base” has customized the UI, then the result is likely to be a UI that fails to meet (at least optimally) the needs of their customer base.  I would hope that they can still reconsider this issue, because I think the consequences may be large, both for Microsoft, and for their users who have come to rely on a more flexible UI.

    And: “… you’ll see that one reason for the new UI was that they had no idea where to put all the 2007 features into the old UI.”

    I don’t know – is that even necessary? If they would adopt (as we have done in DataSmiths Evolution) the optional ability to switch back and forth from the new UI to the previous (already developed) UI, then users may be allowed the best of both worlds (with it being understood that some elements are only accessible in the new UI, but others, like the expanded grid of 1,048,576 rows by 16,384 columns would be equally accessible by both). My sense of the discussion thus far (and I don’t pretend to have covered it all) is that much of the discontent rests with the ways (or lack thereof) for previously used code and methods to alter the new UI. I guess my suggestion (and hope) is that, rather than trying to crack that harder nut, to just provide an optional switch that allows users to elect to invoke the classic UI when it suits them best.  I don’t mean to over simplify the prospect – I know that Microsoft works with different, more complicated tools in creating Excel than I (and others) do in creating and altering our UI.  But those more complicated tools are also more *capable* … and it’s difficult for me believe that if I can find a way to provide a backwards-compatible means to switch to the previous UI that users have invested in and built upon, that they can’t find a way to do the same.

    And: “So they simply don’t even have an idea how a classic UI would look like.”

    I’d suggest it could look approximately just like the classic UI looks now, with the same full abilities to optionally alter that classic UI when it suits all our myriad interests to do so.

    And: “The productivity argument is actually a very interesting one, and one that has been argued a lot about with the new 2007 UI. Interestingly, it generally is made by Excel users.”

    I expect that it’s true that this argument is particularly applicable to Excel users, because I’d venture that of all the applications’ users in the Office suite, Excel users, and the widely diverse nature of their collective works and all of the wonderful, productive ways in which we’ve been able to optimize the Excel UI to achieve our ends,  have likely produced the single most customized application in the Office suite.

    And: “I rather think of myself as a power user of Office in general. Therefore, an interface with everything one click away just wouldn’t do it for me. I’d have to remember what each of the buttons did and where the button was for each application.”

    I’d guess that most Office users are “generalists” in using the full spectrum of applications in the Office suite, but because most users also have specific, specialized and “work-centric” responsibilities in our jobs and careers, I’d also venture that most users also spend the majority of their “Office time” in just *one* application (e.g., for secretaries, journalists, and some echelons of management, that *one* application might be Microsoft Word or Outlook; for analysts, accountants, engineers, and other echelons of management, that *one* application might be Excel; for still others primarily involved in marketing and advertising, most of their time might be spent in PowerPoint or Publisher).  As 450 million users have said, Office has been wildly successful in the past as a general purpose, *productivity* suite, in providing *both*: a) a highly customizable UI that satisfies the needs of specialized users to alter their UI in order to productively meet their specialized needs; as well as b) a reasonably consistent, default UI that allows us to be plausibly proficient (if not optimally efficient) in those “other” applications in which our work only requires us to “dabble”.  

    As admirable and ambitious as it might be to try and enforce a “one-size-fits-all” UI across all applications, if the result is one that not only doesn’t *support* (with a full set of built-in tools) but doesn’t *allow* users to fully customize their UI to meet their specialized needs in their *one*, most specialized application of choice, then, I think, we’ll all experience a loss of productivity in those specialized areas where most users spend most of their time. And (for the reasons stated above) across all of the Office applications, nowhere will that be more evident, I think, than with Excel.

  89. Patrick Schmid says:


    don’t make yourself any false hopes. Neither a radically different customization system nor a classic UI will happen in Office 2007. If any of these two will make it ever into Office, it will be in the version after Office 2007. Microsoft has made this very clear.

  90. jeffreywsmith says:


    I do take your points and am neither waiting for nor overly optimistic about any near term "relief" coming from Microsoft on these issues.  I guess I’ve taken the time to comment here because: a) I think there’s better ways to design UI’s then the ones they’ve taken here; and b) if the Beta cycle is supposed to elicit responses from their user and partner community about the directions they are taking, I wanted to suggest ways the UI design process (and its result) could be improved and why I thought it was important to do so.  While it may well be that they are too far down this road and too far behind schedule to consider this input now, I don’t know what technical challenges they might face in building an optional switch to connect Office 2007 to the already developed classic UI – heck, it might even be easy, and maybe, with all the continuing discussion and expressed discontent, they’ll listen and see what might be done here (even before release to market).  

    If not, well, at least I’ve tried and perhaps (if they are still listening), they might realize that their CEIP project (as it is presently constructed) had no basis driving the UI design because it was too unrepresentative of the market they serve, and they’ll consider better alternatives on the next go.  I have to think they would have spent less money and gotten much, much better data if they had resorted to the random sample approach I mentioned in my first comment in this blog.  Certainly the sample sizes were within their means and the methodology much more inclusive than the narrow look they’ve taken with CEIP.  

    Well, I think it was you that said in another blog comment that you had “flogged that horse enough” and I guess I’m there, too.  They’ll either listen and respond if they can … or they’re past that now and we’ll all just have to see how the market responds.  Since there’s only been about 3 million downloads of Beta 2 so far, it almost goes without saying that the other 99.3% of the 450 million users in the Office community haven’t “voted” yet and most of them aren’t even yet aware of the discussion.  I expect that “input” (when all our “votes” are counted in the marketplace) is the *one source* that informs Microsoft best.  If they’ve failed to anticipate their user’s needs as well as they should have, then maybe some of the dissenting comments offered in these blogs will have more cachet as they try to discern where they went wrong and how best to fix it.

    In the meantime, I’m glad for the efforts of people like yourself that are building “bridges” to the new technology and helping the rest of us develop a road map through this unfamiliar territory … it may (eventually) become the only way to cope (at least once the life cycle of earlier Office versions expire and no longer receive Microsoft support against the hazards of viruses and such).

    Best regards.

  91. splib says:

    Frank Eskridge (and maybe others)

    i’d guess that people who avoid customizing for portablity reasons, also tend to use keyboard commands, rather than "drilling" thru menus with the mouse?

  92. Julian says:

    I get the ideas here in relation to most of office – however I am a heavy user of Access – which is different. Access is essentially a development tools and REALLY NEEDS a capability to customise an develop the ribbon and for such  changes to be contextual. Most of the time you don’t want users having access to ANY standard options.

    Comments please…

  93. Patrick Schmid says:

    Julian, Access actually has the most flexible customization system of all the ribbon apps from a developer point of view. After all, in Access, you can specify a ribbon via VBA, which you can’t in the others.

  94. Wenn wir über Erweiterbarkeit des Ribbon UI reden, kommen schnell auch die Power User zu Wort, welche

  95. I missed this awhile back I think. Jensen Harris has a great post on customizing the Ribbon.

Skip to main content