My brief screed about, IMHO, EMC’s braindead policy of requiring analysts to sign an NDA to attend analyst briefings, touched a nerve.
I’m new to the analyst game. I don’t know anything about it. For all I know, almost every company insists that analysts sign NDAs before briefing them. Maybe financial analysts sign NDAs too. It was just my luck that the first and so far only analyst meeting I’ve attended was at NetApp, and they didn’t ask for an NDA.
Or maybe it’s that I don’t care about roadmaps
As another long-time analyst pointed out to me, most companies want to share their roadmaps with analysts and don’t want them published. From long experience I’ve come to believe that roadmaps are almost worth the paper they’re printed on. I agree with EMC and Apple that new product proof is in shipping, not hyping. So please, spare me your roadmap.
An essay question for analysts
Since I don’t know anything, I’m asking the people who do. Here’s the topic:
How does information you receive under NDA, that you are not supposed to share with clients, help you be a better analyst?
Some areas I’m interested in:
- Does the NDA change your behavior with clients? If so, how?
- How you evaluate the credibility of the information you receive under NDA?
- What, in your view, is the purpose of an NDA in a vendor-analyst relationship?
You can write your answer and paste it into the comment box below or send me an email. Please tell me who you are and if you want anonymity.
I’d like to hear from as many analysts as possible, and I’m very interested in what folks at Gartner – the 800 pound gorilla of IT analysts – think.
I’ve already heard from one analyst, my colleague at Data Mobility Group, Walter Purvis, who thinks I’m “dumb” for not signing the NDA. Former analyst Marc Farley also weighed in. Check out the comments to EMC has Ph.Ds? – Part I below.
The StorageMojo take
IT analysts play a useful role in a complex world. CIOs use them to assure CFOs that a vendor or strategy is OK. They lubricate dialogue between vendors and customers, providing an “impartial” outlet for each side to vent to. Analysts can be usefully brutal with clueless company managements, while providing a sounding board for those willing to engage in a dialogue. A group of smart, independent and well-informed people who talk to everyone and are beholden to none is a Good Thing. And nice work if you can get it.
There’s a dark side too. Weak executives who don’t want their world view challenged, or want analysts to spy on the competition. “Pay for play” white paper writers who happily regurgitate company talking points with a sentence or two of sorta, kinda, negative “analysis” in exchange for fat fees. Customers who want to ream their vendor with negative analyst commentary. It is all very human. Consumer beware.
Comments encouraged. How do you end-users view the analyst community? Clueless flacks? Deep wisdom? Useful fiction? Necessary evil? Phil, Prince of Insufficient Light? Please weigh in.
April’s Harpers Magazine included a lovely quote from an internal Microsoft document admitted as evidence in the Comes vs. Microsoft class action law suit. The “Generalized Evangelism Timeline” document includes many wonderful insights into Microsoft’s take on guerilla marketing and product evangelism, but the one that struck me as most appropriate to this post relates to creating a “stacked panel” to discuss Microsoft’s technology:
“The key to stacking a panel is being able to choose the moderator. Since you can’t expect our competitors to speak on our behalf, you have to get the moderator to agree to having only ‘independent software vendors’ on the panel. No one from Microsoft or any other formal backer of the competing technologies should be allowed … Get the press to cover the panel, and you’ve got a major win on your hands.
The best sources of pliable moderators are:
– Analysts: Analysts sell out – that’s their business model. But they are very concerned that they never look like they are selling out, so that makes them prickly to work with.
… ”
The document also mentions consultants as being the best bet as moderators, but it’s the views on analysts that I thought were most illuminating given this current thread.
BTW, I should point out that I think analysts serve a very useful purpose. Having worked in the trenches in IT, I can attest to the benefits CIOs and IT managers get from the insights analysts offer. But I think it’s a mistake to put analysts on a pedestal. They’re only human after all.
Robin, I think you ask some interesting questions here, especially the one about the credibility of information you receive under an NDA. There is that old joke about the difference between used car salesmen and computer salesmen that this remends me of. Supposedly, things you find out about under the cone of silence, should be accurate, but they also tend to be futuristic statements that never come to pass.
NDAs are usually abusive and unnecessary, so don’t give up too fast. Bloggers are changing the world, often because of the level of honesty you find in the blogosphere. NDAs are all about controlling what other people can say. If your market becomes attractive enough for EMC to do business with you on your terms instead of theirs, then you might just change the status quo.
Yesterday you were flogging EMC for “the lack of a vision, or goal, that EMC is aiming for… where is EMC investing its R&D dollars to create the products, services and architectures that customers will need in three years?”
Today, you say “please, spare me your roadmap.”
Do you or do you not want EMC to tell you where it’s going? Please clarify.
(You’re ludicrously wrong about EMC not having a vision, of course, but that’s a different matter.)
NDAs stink but they serve a useful purpose. After all Gartner couldn’t plot out their visionary axis of the Magic Quadrant if they just looked at public info. And we all know that without the Magic Quadrant, we would be just swimming in non-visionary poorly executed equipment and software – wait sorry but IE just crashed.
But seriously, the whole purpose of the analyst-company dance is to get good press that actually generates interest and then sales. So when they talk to you or any analyst, its all about how they are trying to manage their message. Back in the day when I was an analyst, the NDA’d information was great to have so that you could reseach and write at your pace. So when it finally was public, you were providing clarity and clear analysis rather than just another person bothering the publicity people.
Thats the real quid pro quo, the company is paying you off with information. You are more valuable because you have it. So you sign the NDA and mediate your speech but people come to you and pay you for the insights you have via the NDAs. You become another information channel.
The blog has changed that somewhat in that you don’t need a Gartner or IDC to have a platform to provide insights. So the channel that you use here may make signing NDAs counter productive.
Robin’s logic on this issue is absolutely correct …. what is the point in having a ‘roadmap’ and (as an analyst) not being able to discuss it.
As for the openly stated ‘vision’ at EMC … please see what you can make out from the article on
http://www.internetnews.com/xSP/article.php/3675886
This type of nonsense is not worth printing.
Hi, Robin, just correct some misstatements you made on your zdnet blog on RAID 1+0 here, since I don’t want to have another zdnet account. So no need to post this one here, but try to correct it on zdnet. I love both of your blogs, that is why I don’t want your readers to get mistaken information from you.
Anyway, RAID 1+ 0 or RAID 0 + 1’s fault tolerance comes from the 1 part, thus both of them can tolerate 1 disk loss in worst case. In your example for RAID 1+ 0, any one of disk 1′ or 2′ or 3′ dies, you lose your data, same for RAID 0 + 1. So it is not true to say RAID 1 + 0 can tolerate 3 disk failures at the same time. It cannot even tolerate 2 disk failures ( e.g., 1 and a ).
About performance of RAID 1+0 and RAID 0+1, if you implement in software, they are the same, which should be. ( We tested in my lab. ) The difference people see may come from RAID controllers, which I haven’t explored.
Cheers.