Tuesday, December 13, 2005
Enterprise Architecture and Analyst Relations
When large enterprises started purchasing industry analyst research several years ago, they really weren't interested in understanding whether information was accurate or even credible. They simply were overwhelmed with the myriad of choices that new technologies such as Client / Server and the Internet brought about. They looked to the analyst community for direction. Any direction!
Nowadays, analysts are not be used to guide high level "strategy" but surgically to assist in providing strategic competitive advantage to the enterprise. Analysts in order to provide value to the enterprise, first need to deeply understand the problems that enterprises are facing. Many analyst firms only have an eye on the problems enterprise architects face through the eyes of software vendors. While it is interesting to be able to have a piece of software that can solve every possible problem on the planet, reality dictates that that approach simply isn't sustainable.
It takes a lot of effort to maintain a relationship with a software vendor and having hundreds of them in a large enterprise simply isn't sustainable. Enterprises nowadays are attempting to consolidate the amount of vendors they do business with in order to bring sanity to the process. This serves as an introduction to how to work with the analyst community:
- Don't allow analysts to tell you about vendors during a briefing. Have them describe the problem-space and potential solutions in a vendor agnostic manner. You should enumerate all of the existing vendors within your portfolio and instead seek guidance on how they could be steered into extending an existing product to meet your needs instead of acquiring a new one.
- Give analysts tough problems to solve. By providing analysts with the opportunity to tackle tough problems and letting them know where their contributions have paid off provides an opportunity for a two-way dialog. Way too many enterprises establish briefings with analysts and don't provide them with any homework. They may simply allow them to defer answering questions within the context of a half-hour call and only provide summary level information.
- Not requiring industry analysts to list open source products alongside commercial offerings. The biggest opportunity for cost savings within an enterprise comes from open source software. If the analyst hasn't mentioned a single open source project in context of their report or briefing, question them immediately. It is important for the enterprise to be informed on all opportunities for cost savings not just the ones the analyst "feels" you should hear about.
- The best way for an analyst to understand the struggles of an enterprise is via a face-to-face conversation. It should be mandatory that every single analyst meet face-to-face with their clients at least once a year (and no, conferences do not count). Some analyst firms will do this under the guise of a consulting agreement but this misses the entire point of a two-way dialog. This should be part of the normal course of operations for an industry analyst
- Using only large analyst firms. While the breadth of the portfolio may be vast, the thing that is crucial is that enterprises also have the ability to go deep. Usually depth is best handled by boutique consulting firms. For example, if you talk with Jason Bloomberg of ZapThink, he can go deeper into an SOA conversation than the big boys. Buy research from large and small.
- Choose an analyst firm that gets quoted for being controversial. If the analyst firm you are looking at only says glowing things about the biggest vendors, walk in the opposite direction. You want to have someone on your side with a unique perspective, not the repeat after me crowd. Unique perspectives will usually bring about an insight not obtainable in any other forum. Another reason I like the folks at RedMonk
- Tough love is crucial to the value proposition. The old adage of the customer is always right is always 100% wrong. Never trust any industry analyst who doesn't provide brutally honest candid feedback on your strategies and ideas. I have been busy bouncing my thoughts of Fred Cohen and Dan Blum of the Burton Group security practice and they have been doing a great job of being honest. These folk understand that we are paying for insight, not compliments...
- Beware of the analyst who doesn't blog. Engaging in a dialog with the community can provide more insight than used closed approaches. Likewise, blogging isn't meant to be one-way. Beware even more wary of analyst firms who don't enable trackbacks...
My list is by no means complete and I will even disclaim there are several situations in which I will disclaim my own recommendations but for the most part they hold true. One alternative perspective I have is that sometimes enterprises don't really want unbiased industry analysis. Sometimes lack of bias can help us more quickly reach consensus and deliver valuable work.
Anyway, I leave you with several interesting blog entries in this space:
- Customer research continues to grow
- Blogs impact on analyst relations
- Analyst fingerprints on billions of HP's revenues
- Managing an analyst relations crisis
- Innovation and R&D
- The Changing Influence Game
- About the business myths of IT analysts
- The technology of cooperation.
- Solving Classic Problems Using SOA
- Measuring Innovation
- Thoughts on OpenTTO.org
- Vendors Are from Mars, Customers Are from Venus
- The Customer Has the Answers
- Ensuring Resilient Collaboration
- Why can't we all just get along?
Links to this post: