Original URL: https://www.theregister.com/2006/11/16/surveyor_speaks/

Oracle: a surveyor speaks

How can two similar surveys of the similar people produce different results?

By David Norfolk

Posted in Software, 16th November 2006 21:07 GMT

I'm no survey enthusiast (in fact, as readers must realise by now, I'm pretty cynical about survey usage in the PR industry) so I thought I'd ask Dale Vile, our resident survey guru, about some of the points I raised about a recent UKOUG (UK Oracle User Group) survey here.

How might the Oracle user sample in the UKOUG survey differ from yours?

Our sample is probably best thought of as being made up of IT pros involved in either the continued implementation or running of Oracle applications – not necessarily the primes. The other thing is that it will be skewed towards those who have an awareness of Fusion, primarily because it is online research and that was what we were asking about. So, as you may have read in the report, this has been positioned as a temperature check rather than definitive study.

And the Fusion question?

Our question was specifically about whether the Fusion App strategy would help with investment protection. This is not the same as asking about whether you are confident in future support. I haven’t seen the UKOUG survey yet, but if you put together the fact that half of them don’t know about Fusion, with you telling me that 3/4 are positive about recommending Oracle to a friend, say, that kinda suggests that some of them were commenting based on a degree of ignorance about Oracle’s strategy (you haven’t told me enough to know for certain though). I would be interested to know how many of the Siebel respondents you mention, for example, have a good knowledge of Fusion.

Personally, I don’t see how you can make a sensible judgement on the Oracle future unless you have a handle on Fusion.

Of course, when you're part of an acquisition, confidence can see-saw.

This is a guess, but I reckon the lifecycle of confidence goes something like this (e.g. for a PeopleSoft customer):

  1. Oracle: “We’re taking over”
  2. Customer: “Oh, shit”
  3. Oracle:”But we pledge future commitment to your product”
  4. Customer: “Oh yeah, heard that one before”
  5. Oracle: “No really, here’s our support commitment”
  6. Customer: “OK then, I’m happy”
  7. Oracle: “BTW, we’re going to pull everything together into this Fusion thing”
  8. Customer: “Oh shit”
  9. Oracle: “But we’re going to do it in the right way, let me tell you more …”
  10. Customer: “Ah, I see – that’s actually pretty good”

I guess this is the ten steps to informed happiness.

I reckon most of our respondents were between stages 7 and 10 (because of the self selection), where as at least half of the UKOUG respondents were between 2 and 6, obviously with quite a lot at the 6 stage. Maybe the difference is between happiness and fully informed happiness.

Of course, the only ones that really matter are the ones that are at stage 10 – our survey says 4 out of 5 that get here are “happy”.

Any other thoughts on this survey thang?

Well, I used to do a lot of rigorous surveying of the SAP user base. We found as part of this that if we interviewed those responsible for the day to day relationship with SAP or the SAP programme itself, we got a very positive skew on everything – i.e. by default, they give the vendor the benefit of the doubt and in extreme cases, almost go native. At the very least, they are the champion internally so don’t want to undermine that which their professional lives revolve around. If you interview architects, FDs, even CIOs, etc, you get a much more “warts and all” view, and these guys are not afraid to challenge what they are hearing from the vendor or seeing in terms of behaviour.

Now, who tends to represent the company at user groups for products like the UKOUG? Typically the champions with the less critical outlook....

Yerrrss, I remember people in IBM shops who almost sold their soul to IBM. Now let me tell you how good AD/Cycle really was....

Indeed - and, BTW, I stand by my overriding conclusion that Oracle is not being as clear as it could be in getting objective, complete and unspun information out there. ®