It’s all about setting expectations clearly

I read this article in CNet today about how Betas are becoming longer and how the lines between a Beta and the final product are blurring.    


As you know, Microsoft Visual Studio took the lead in releasing CTPs (Community Technology Previews). Other products are starting to do this.  The goal behind this exercise is to provide customers access to our early builds and start a 2-way dialog that helps us deliver “the right product” for our customers.  I believe in the saying “Life is all about setting the right expectations and delivering on those or exceeding those”. Some people refer to this as UPOD – Under Promise and Over Deliver.  As long as you set the right expectations with customers in terms of what they can do and more importantly what they can’t do with our products, I would always err on the side of transparency.


For example, we are very clear to our customers about CTP drops, why we do them, what we expect customers to be able to do and not do in these CTPs. We also have the notion of a “go-live” license where for certain Betas, we work with a number of customers to deploy these Betas in production environments.  Once again, it has to be a decision made by the customer working in partnership with us.


By the way, it isn’t that we are super good today about being predictable with when we deliver our software.  Our goal, as part of transparency, is to ensure that we both share our roadmap and do appropriate change management as necessary.  


If we ever get to a point where we are not crystal clear on setting expectations with customers about what they can and can’t do with early bits of our products, that is a big dis-service to our customers and ultimately to ourselves.  I do hope that we and others don’t get confused between being transparent with customers/community and worrying about long beta cycles.



Comments (6)

  1. Actually I’d point to the Exchange early adopters program (started in 1992) as the leader in this field… We had representitives of our biggest email customers on-site during the entire lifetime of Exchange 4.0. They worked in offices side-by-side with the Exchange developers attended war team and directly contributed to the product’s features.

    But I’m biased 🙂

  2. S. Somasegar says:

    Yes – I do agree with Larry in that Exchange was the one who led the way with Early Adopter Program in a completely programmatic way at Microsoft. This is now a best practice that the product groups use today.

  3. What *really* pisses me off is the licensing status. I think that if I own all my own hardware, run my own lines and won’t be reselling my services, I should be able to go-live.

    I went to the Whidbey SDR in June of 93. yes, it *really* was June of 93 and then I was told the first lie of go-live around the first of the year. I won’t go through the trouble of cataloging all the lies told since then but suffice it to say I have no confidence in a tangible date. Hell, eWeek is reporting march 31 but that doesn’t gel with SQL Server team members saying IDW13 will be out in march (assuming a beta3 yukon/beta2 whidbey parity).

    It just chaps my ass though. I own my own server. I have my own lines. I write my own code. Why can’t I use it? I understand Joe’s Hosting Company not being able to put customers on it; but why can’t I use it?

    You guys aren’t pissing people off because of the extended cycle. We want you to get it right. You are pissing us off because you say one thing and then it changes and then you say something else and it changes again. The whole time we are getting strung along with no avenue.

    64bit is a perfect example. I bought 10 64bit machines almost a year ago for a very large project. The idea was to run 64bit netFx. Yeah right. Total waste of money. We ended up paying someone to port the code to the Java because a: it had generic support and I used them extensively in the Whidbey code and b: it had native support for 64bit.

    I just don’t get it. Use at your own risk on your own stuff seems fair. Especially when the product cycle starts to drag. At this point in time I have 4 competed applications that I would deploy tomorrow if I had a go-live license. Just thinking about how much money I’m losing by not deploying these apps makes me sick. Everytime I think about back porting these apps I get told a new lie on the go-live date. It will take me about 4 months to backport, the new lie is almost always two months away.

    Dunno. I’m a big fan of Microsoft, but to be blunt your go-live policy is bullshit.