I have no patience for analysts. I have little respect for them, and even less for those who pay them big bucks for their “expertise”. Not all analysts mind you, I have not had to deal with them in lots of industries, and maybe some in some industries have some opinions of some value. The ones I have had to deal with extensively (in numerous capacities with several different companies) are the IT analysts.
You’d be hard pressed to find an article in a prominent IT publication that does not quote one of the major analyst firms spouting forth on .NET versus J2EE or WiFi or security or open source or standards or whatever else is the issue of the day. Similarly, you’d be hard pressed to find a CIO of a major organization who does not in some way rely on reports, conclusions, and recommendations provided by these analysts. After all, the IT press and CIO’s can hardly be expected to be experts in everything IT related, and so a hefty check and nicely printed whitepaper or briefing is exactly what is needed. Right?
Well, at the risk of generalizing, maybe not. I have read lots of analysts reports, I have spent way too much time with analysts briefing them on various topics, and I have had too many conversations with decision makers who approve all sorts of spending based on analysts reports and recommendations. And I have yet to meet a single analyst who actually knew, or had experience with, what he or she was talking about. Few have ever held IT jobs, and yet they make all sorts of predictions about the IT job market, decisions that actually impact that same market. Few have ever written a line of code, and yet they have no problem discussing languages, coding practices, and methodologies. Few have ever actually touched a server, and yet making infrastructure recommendations is the norm. Few even know what the acronyms they use really mean, but they use them anyway.
How do I know this? Analysts in the IT sector are not unlike analysts in the financial sector. I always find it amusing when publicly traded companies post numbers (good, bad, whatever) only to be followed the next day by analysts changing ratings to buy or sell. Does anyone actually need an analyst for after-the-fact ratings changes? Now, if the financial analyst changed ratings based on research and market analysis, before financials were disclosed, yep, that would be of value to investors, but that is not the way it usually works. And IT analysts are no different. Most seem to just go with the flow, spouting conventional wisdom, getting paid to regurgitate opinions they were just briefed on. I know, because I have done this. I know because I have spent time with analysts, I have briefed them, and have seen the subsequent published data. I have been presented with findings that were my own comments prettied up. Lots of spending and decision making is affected by analyst data, and so the major players make sure to feed the analysts with all the right information so as to ensure that the reports say exactly what they want. Some analyst reports are even funded by the vary companies whose products and technologies are being analyzed. Companies have to do this, because their competitors are doing the same. And as soon as one analyst makes an insightful comment, others are quick to agree. Opposing analysis is not the norm, regurgitation and lemming-like behavior is. And the scary thing is that billions of dollars are influenced by analysts, the futures of products and even companies has been impacted by analysts pontifications.
As a rule, I avoid referring to charts and stats provided by, well, you know who the major players are. I have no faith in them, and have a hard time accepting the fact that these do indeed impact decision making, and a harder time respecting those who rely so heavily on analyst data. When analyst data is presented in a meeting, regardless of whether it is proving may point or otherwise, I cringe. As I said, no faith, at all.
The problem is that I don’t know of a better option. The fact of the matter is that IT is complicated and grows increasingly so. Decision making has gotten complex and will continue to do so. Relying on outside experts make sense I guess, so long as they are actually experts, and in my experience few (if any) analysts actually are.
There has to be an alternative, but there does not seem to be one yet. It’s a tough one, and it is only going to get worse.

11 thoughts

  1. FWIW, I’ve been both and prefer to manage code than people;)
    Ben, thoughtful analysis. (Sorry, couldn’t resist:)

  2. Ben,
    I’m curious… was there a specific report that you read that inspired this? What was the ‘straw that broke the camels back?’

  3. Ben, I can’t but agree with you 100%.
    From my own experience, analysts suck!
    So many times, they come around for one job or the other and you always end up doing it yourself because they make you gather the data, analyze it, have an opinion, etc. Ok, they go and print the reports in their offices.
    Anything else?
    What I have learnt to do now is to work with my guys and brainstorm and talk to industry partners. At least, if we all resign and go home, we could be analysts too!
    Anything else?
    What I have learnt to do now is to work with my guys and brainstorm and talk to industry partners. At least, if we all resign and go home, we could be analysts too!

  4. what’s even worse is organizations allocate annual budget to get those reports that their CIO’s & CTO’s never ever read (If you pull a report at the end of the year, you will be shocked to see the number of hits to be in the low single digit). They even make decisions to pay for it by cutting-down on software licenses that could have made the development team more productive.

  5. We are lacking an example of what it is that makes you dislike or diagree with the analists.
    Of course I can understand your frustation, as a "guru" having to explain things to people that can not stand in your shadow technically.
    On the other hand its another ball-game on decision-level (You of all people know this, so why…). Maybe analysts say too many things they know people want to hear. On the other hand I cannot blaim an analist trying to promote .NET to a decision-maker. He is not wrong. He is telling him to play it safe. Even if it makes my ColdFusion hart crying.
    Is it not the industry itself that has the blame? Our lack of standards, hype-promoting, etc. Of course analists can not see the light, but can we see it ourselves? No wonder decision makers are eager to try anything to make something out of this mess. And listening to not-IT people is not bad, as long as we use these outside opinios to improve ourselves. We need it.

  6. Here’s a viable alternative:
    Get one of those 8-balls filled with water with the little window on the bottom and use it to make your decisions. You’d most likely get similar or better results than dealing with the analysts.
    I can see it now: "Ben, should we go with J2EE to manage our enterprise?" … "Let me shake up the 8 ball … here it comes – ‘All signs point to yes’. Yes, let’s go with J2EE." … "Very well, Ben. SANDRA (shouting at secretary), GET ME THE PEOPLE FROM SUN ON THE PHONE!".
    The lessons we learn! 😉


  7. Straw that broke the camels back? Huh, what made you think that there was one? 🙂
    Actually, it was what you’d expect. A large organization, very successful and comfortable with a technology and solution, suddenly considering sweeping changes because of comments made by an analyst.
    Fun stuff.

  8. Ben,
    Thanks for the insight. I recently changed jobs, and one contributing factor was a shift in development platform based on whatever analyst opinion was highlighted in various magazines on a given month.
    I hope that this will change as my (younger) generation of developers slowly gains seniority, creating a IT management culture that has had hands-on experience, and knows better than to listen to those who haven’t.

  9. I did a job once that involved working with data provided by Gartner Group. As our effort progressed, we found mistake after mistake in the calculations by Gartner Group. Not to mention figures pulled out of thin air.
    And to think, the Gartner Group was paid a thousand dollars an hour for that data.
    Guys, lets face it, we are in the wrong business.

Leave a Reply