Microsoft the Contest Machine
Last updated: 6/30/08
NOTE: If you've been forwarded a link to this blog post, it's likely that one of your colleagues thinks you should do your due diligence before thinking a contest is the fastest road to salvation. It isn't, here's why...
I'm starting a crusade at Microsoft to lower the number of contests we run a year. I know I'll be successful if the number of contests a year goes down. I fear I won't be. First, let me explain by showing how bad the problem is.
The Problem
Today is March 17th, and there are already *18* 30 contests that will finish this year. That's 1.6 contests a week. Mind you, this is only after a cursory search and it does not include Microsoft partner contests (large companies like HP, community sites like codeproject, etc).
2. Mix Show Off
5. Imagine Cup
7. WinTheWeb
8. PhizzPop
9. Office Developer Sleepless Contest
11. Next Gen PC Design Competition
12. Small Business Summit Technology Makeover
13. Microsoft Future Pro Photography Contest
14. PowerShell Mascot Naming Contest
15. Windows Live Contest (Zune A Day Giveaway)
16. https://www.donburnett.com/images/blog/MicrosoftisLIVEinEducation_68BE/MashupContest.png
19. Web Harmony
20. Dobbs Challenge (sponsored by Microsoft)
22. SQL Server Heroes 2008 Contest
23. INETA European Silverlight Challenge (sponsored by Microsoft)
25. Windows Media Center: Ultimate Install Contest
26. TechEd SEA: Blog and Win contest
27. AT&T and Microsoft Indy Mobile Game contest
28. Creator's Club Game On Sweepstakes
29. Microsoft Expression Contest
30. Microsoft Dynamics NAV Contest
31. Silverlight Control Builder Contest
32. INETA Silverlight Challenge
33. Software + Services - Bring it all together
34. PDC 2008 Show Off (and yes I am a "host" for this)
That's a *lot* of contests. Most customers don't even know that half of these exist. In fact, I only knew of about five of them, but still teams are out there building another contest page and even having contests for the same customer base overlapping each other. I wonder if you could live off of just entering Microsoft contests...it's never the same teams evaluating your project, idea, etc... Perhaps I should run a contest to have one person try to win every Microsoft contest :)
Why do teams run contests?
While this depends on the group, here are some of the most common reasons
- Community - Grow a community or get the community involved in a technology -
- Content - Get samples built using a technology
- Testing - Test a pre-alpha application
- Usage - Drive usage of a new product or service
Contests to Drive Community
Running a contest to grow a community is probably one of the worst reasons why to run a contest. First off, a contest pits users against users rather than having users work with each other. It forces what are considered equal members to be subdivided based on their skills or aptitude. It's not the healthiest thing to do and it certainly doesn't grow your community.
Contests to Drive Content
Another common reason is to have content built using your technology. There are a couple of reasons why this isn't a good idea, including:
- Not Shareable - You did not setup your contest so that the projects themselves would be open source projects and available to everyone
- Licensing - The content makes use of things like GPL code or components which may bar the Microsoft team from using it elsewhere
- Poorly written The content is built with poor programming standards
- Fragile bits - The content is built on an alpha or beta version and likely won't work with the final release
- Wrong audience - The audience for the technology is ISVs or VARs that really don't want to share code that they could use/resell
The goal is to drive some form of behavior and in many cases, it's decided by a product manager for a specific technology that really just wants applications and samples to be built using that technology and to drive PR.
Contests to Drive Testing
Things like a bug bash are probably some of the best suited for contests as it is really more of a reward for helping test an application. The only issues with testing is that depending on the software's complexity, the end result of a bug bash or extra testing end is likely to be documenting the issue in a "known issues" or "readme" document and punting the bug for the next service pack 6-8 months later. If the software is smaller, say like a starter kit or sample versus Windows 7, then most of the bugs found will actually be fixed by the time the sample ships.
Contests to Drive Usage
Usage can be a mixed bag, depending on the service. In many cases, it can drive the wrong behavior, like having hundreds of people sign up for a service that they will never use, just for a chance to win a sweepstakes. The important thing to remember is that you want to drive the right behavior with the right audience. In many cases, customers will jump through whatever hoops necessary to win your "prize" and things like free online services are the most likely to be abused.
Contest Realities
- Contests drive no PR to your product (with few exceptions). Look at Google and Apple, they smartly do very few contests, about 1-3 a year and they get great press out of them versus the 30-50 department contests we do.
- Customers are already inundated by existing contests, why will yours rise to the top?
- Contests may only have a limited number of entrants (20-30). That means your $5,000 in prizes and an additional $10,000 in demand generation, site design, legal fees, etc was a very ineffective spend for such a small audience. You could get more customers by handing out $100 bills at Best Buy
- Contests that do have lots of entries >1,000 require a massive level of effort to sift through and measure each entry. You better have planned for a lot of logistical time in reviewing each and every entry and how to deal with international entries that may have parts that aren't in English (assuming you're doing a worldwide contest). Judging is a huge, huge time sync and something that would be better used actually talking with or interacting with customers.
What else can I do besides contests?
This depends on your goal, but if your goal is
- Increase Community Participation - Don't spend money, spend your time engaging customers in your product/service or related community. This could mean 3rd party communities as well
- Increase Content
- Product Team: See if you can drive your product team to have an "App Week" where they build content that you would want
- RDs/MVPs/Student Partners: There are many existing "experts" that don't have a blue badge. In many cases, they're more than happy to help you identify and build content for your product/service, but price and quality do vary between groups
- Outsource: There are tons of content experts that exist in the world, ask your colleagues who a reputable vendor is for your product/technology and you'll likely find that the cost for a great use of your technology would cost the same, if not cheaper, than running a contest and without any of the associated issues
- Drive Usage
- Instead of "bribing" someone into using your product, spend money on making sure your target audience is aware of your offering (advertising/promotion), talk to influencers and see if they'd be willing to try and evaluate your product/service, find another organization to partner with that makes sense, or spend the money on improving the product through "unofficial" features that would make customers *want* to use your product.
- Get the Marketing Playbook
- Here's another 102 ideas of things you could be doing to achieve the same goal
Do you know of other contests?
Join the overdone-contest movement! If you find a Microsoft team doing a contest, comment here and we'll add it to the list.