Version usage

At the Midwest Regional Users Group meeting this week an interesting statistic was shared with us at the E1 town hall meeting.
Here is the current breakdown of EnterpriseOne use by version.

<ul type="square"> [*]XE......24% [*]8.0.....11% [*]8.9.....19% [*]8.10...19% [*]8.11...24% [*]8.12.....3% [/list]

Also Oracle's current time line (with appropriate safe-harbor statement) shows 9.0 and tools 8.97 in 2008 and 9.x and tools 8.9x in 2010.

That is very interesting. As we work on our upgrade to 8.11SP1, and run across what would seem like a very common issue I often wonder how many people are really on this release!
Thanks for sharing.

I think those figures are completely bogus - maybe slanted by shipped CD's ? There is no way that only 24% of users are "currently" on Xe - that figure must be higher.

Secondly, I don't know ANYONE who is on 8.9 and liking it - almost everyone that started on the 8.9 path very quickly updated to 8.10 based on the issues there. I also think the same will occur with the 8.11 users, and cannot believe that there are as many 8.11 users as there are Xe users !

I wonder how many users are running B7332 still. I hear of quite a few smaller customers looking to update.

I'd love to know where Borgacle is getting their info from - cos its obviously wrong. Even discounting those users no longer paying for maintenance isn't enough to swing these figures.

Now. If you were stating that this was the breakdown of support calls going to GSS by version ? Those numbers would make sense !

I totally agree on all points. These are probably call stats. That's the only way that 8.9 can show such a high count ;-)

There would also be a fair number of ERP8 users - I know around here a few sites moved XE->ERP8 even relatively recently.

I think it would be be virtually impossible for Oracle to know for certain which version of E1 their customer is running, considering their support policy.

If a customer pays maintenance, they are entitled to run any version they choose provided it is covered by the support policy. If a customer doesn't call support and rarely downloads an ESU for a particular release, they would be hard pressed to know exactly what is out there in the weeds.

For instance, under the same support contract, my employer has an absolutely massive 8.9 installation and a subsidiary which is running a new(ish) and extremely tiny 8.11 SP1 installation. The 8.9 install is so massive, they haven't even bothered retrofitting custom mods which would be required if they applied certain recommended ESU's (for instance, the "baseline" for Sales Order maanagement which shipped in late 2004). Rather, certain SAR's were cherry picked when the risk analysis of testing the application and all point solutions from end to end was in the "red zone".

If we were to add a separate instance of 8.12, they'd only know about it based on ESU "fix current" downloads. If we were to buy a company with an existing B733x release and did nothing to maintain it, they'd likely know nothing about it either.

All things considered, Oracle can certainly come close to a reasonable number (possibly within a margin of error of +/- 10%) based on the combination of support calls, ESU downloads (which are tied to the specific release) and courtesy calls paid by the sales and/or services rep. I would say they have more direct evidence of this than any number of independent consultants would have, based on the estimated >5000 paying support customers (a figure I heard in early 2004.)

Keep your hair on Charles - I'm sure they WOULD have a better idea of what is out there - but they're not telling anyone ! They ALWAYS slant their version statistics to favor newer versions - they want customers to feel that by not upgrading they're being "left behind". They did this when I worked for them, and in 2001 they claimed that 80% of customers were running Xe. That figure was PROVEN to be bogus back then since it came out later that the figure reflected the % of CD's that had been requested by companies.

I think we'll just have to see - its not a scientific method of course, but I have put up a poll asking what version people are running. Of course, someone could easily corrupt the results - but I'm hoping we get some sort of cross-reference of the community. I know I have a list of 1,400 customers and a list of versions that they run in the US, and I know that from THAT list, 65% of all OneWorld customers are certainly NOT running 8.9 and above.

Now - maybe the results are the % versions of customers on MAINTENANCE ? I mean B7332 and earlier isn't listed, and I know several companies on "frozen" versions prior to Xe - and that list doesn't take into consideration any of those customers. A simple search on JDE List will easily show B7332 and B7331 version customers are still active.

Finally, I also know of companies that ordered 8.9 or above, started an upgrade and then ran straight into a wall - whether technical or functional. Those "upgrades" would also be "false positives" in Oracles little database !

I'm trying to underline the fact that Oracle/Peoplesoft/JDE or whoever is in charge of OneWorld has actually VERY little idea of how Oneworld is being used at customers. Thats obvious with their blanket requirement for users to stop using the Win32 client with 8.11 and above.

So, Charles, are you stating that you feel the list is 100% accurate just because someone from Oracle tells you ? I think even YOU might have SOME reservations....

Sir Quark,

My hair is always on and firmly cemented to my head using the proper amount of toupee adhesive.

This is what I actually stated:

[ QUOTE ]
I think it would be be virtually impossible for Oracle to know for certain which version of E1 their customer is running, considering their support policy.


[/ QUOTE ]

Where do you derive that I must feel the list is 100% accurate? I didn't hear this from Oracle - I read it on JDEList.

...and finally:

[ QUOTE ]
All things considered, Oracle can certainly come close to a reasonable number (possibly within a margin of error of +/- 10%) based on the combination of support calls, ESU downloads (which are tied to the specific release) and courtesy calls paid by the sales and/or services rep. I would say they have more direct evidence of this than any number of independent consultants would have, based on the estimated >5000 paying support customers (a figure I heard in early 2004.)


[/ QUOTE ]

So, a +/- 10% margin of error is what I stated, not 100% accuracy. I was being liberal with my guesstimation of 10%, it may be closer to 5 or 6. If each percentage were off, and by 5% each, this would bring us to no more than 70% accuracy, no? I'm no math major, but seems reasonable to me.

My opinion is that these numbers would be customers who are on maintenance. If a customer chooses to drop maintenance, they aren't really a customer any more, are they? They contribute nothing to Oracle's bottom line. Oracle would be unlikely to pursue those customers as a reference point for other paying customers, but more likely to pursue them for lost maintenance dollars. It is a capitalist system, after all.