POSTING ON SHELL BLOG
(UPDATED 16 JANUARY 2011 WITH A RELATED COMMENT)
IT4me on Nov 1st, 2009 at 8:26 am
GSAP/GPMR: THE MARCH OF THE DINOSAURS
With COST-CUTTING now everywhere, where are the cuts in IT spending ? The dinosaur in the room of Downstream IT has long been GSAP. Why after 5 years of heroic spending for such pitiful results has GSAP not been stopped ?
One reason is undoubtedly the Star Wars effect. But another factor may be that GSAP was a pet project of the current CIO Alan Matula during his Downstream days. Mr Matula has a fondness for IT spectaculars, vast schemes in which everything changes at once and nothing ever has any track record. Mr Matula is also something of a Dick Fuld, a man with whom disagreement is unproductive. So GSAP blunders on.
Ingenious ways are found to bury costs. One simple trick is pushing work into the future. Another is to manoeuvre difficult bits into the Connected Application Portfolio, the family of systems that GSAP was supposed to replace, but which is actually still growing. But the most flagrant example must surely be the GPMR system. GPMR was launched to provide a NEW layer of Management Information except it wasnt NEW ! GSAP was originally going to replace all the MI with fossilised reports, dramatically sweeping away an entire portfolio of Legacy systems (which were mostly working just fine, and something of an embarrassment to the strategy). The GSAP MI plan failed comprehensively, so GPMR is a disguised 2nd attempt. Lessons learned? No. A new approach? No. Another dinosaur? Seemingly.
If you must build Jurassic, then MI cries out for a Velociraptor: fast and adaptable ! Instead, we have the full T-rex, built around SAP BW. GPMRs one innovation was a new package from Business Objects, recently acquired by SAP so a small opportunity to simplify supply. Brilliant ! Trebles all round ! BO XI was duly crowned as the new MI standard, just in time for the first project to discover that it didnt work with BW. Nobody had checked, and there was no track record. As GPMR approaches its 2nd anniversary, hundreds of millions have been spent with little to show for it.
Is there any escape from this Strategic Disneyland ? We desperately need the CIOs self-serving empire pruning back, and the business to re-assert control. Neither seems likely. Perhaps the estimable Paddy Briggs is right: this a good time to start selling Downstream !
RELATED POSTING: guest1 on Nov 1st, 2009 at 11:20 am
Hello IT4me, I recognise all you state! Since Herkstroter decreed in 1994 that Shell would go completely over to SAP, he had no idea how it would pan out. (Perhaps he had shares in SAP? As a banker he had no scrupules. He even tried to push Shell out of Petroleum Engineering because we have Schlumberger. Only a minor coincidence his son worked for Schlumberger? I nearly start to digress in his extremely poor role in ING so will stop here). And nobody in Shell is willing to make an estimate of the costs todate or the benefits todate. I bet the costs are a magnitude higher than the benefits. And costs are tangible, I would not be surprised if Shell had spent more than $10 billion on SAP related nonsense alone over the past 15 years but perhaps IT4me can be more specific?), benefits much less so. It all appears very similar to what is happening in the Dutch politics. Some politicians want to know the cost/benefit of all the immigrants in this country. Other politicians want to be silent over this at all costs. The silent majority who has a lot of common sense simply knows what it is. Just like in Shell. Your great posting would even be better without some of the specific jargon.
RELATED POSTING: T4me on Nov 1st, 2009 at 1:30 pm
Guest1, my apologies for the jargon ! I think the Herkstroter decision is interesting because 1994 is 15 years ago, a long time in the ephemeral world of IT. The ravenous SAP BW didnt even emerge until 1998. So you might suppose that any decision taken back then has limited validity today, but the SAP habit seems unshakeable !
You also touch on a serious weakness of these big lock-in deals, the suspicion that senior people might hold shares in suppliers like SAP and Microsoft. I have no idea if this happens, but the scope for abuse is considerable. And, by the way, I would guess your USD 10 Billion is not far off !
RELATED POSTING: guest1 on Nov 1st, 2009 at 2:46 pm
And like I told senior managers (much to their annoyance): SAP is like concrete, extremely flexible when it is in the mixing phase, zero flexibility after it has set. Companies like Aera adjusted their organisation to suit SAP, not the other way around. Not necessarily bad, but in the long term it wastes good money. The 10 billion dollar I estimated is money to SAP and the various consultants. The in house effort at least doubles that amount but is easily hidden.
RELATED POSTING: “hard worker” on Nov 1st, 2009 at 10:57 am Has been announced that GSAP will not now be rolled out in US, China, India. Lack of money has forced this decision
RELATED POSTING: IT4me on Nov 1st, 2009 at 1:37 pm
Hard Worker, thats news to me ? As you probably know, US Lubes (US52) and Canada (CA04) have been live since November 2008, and I thought that China and India were still on the chart. Happy to be corrected. Are you talking about cancellations or just deferrals ?
RELATED POSTING: Jim Smitheman on Nov 2nd, 2009 at 10:35 pm
I have waited 5 years to see this post! I used to manage the legacy MI applications that GSAP and the BW warehouse were supposed to replace. Whoever you are Hardworker you are absolutely right that they were rubished by the IT heirarchy because they did not fit the wall to wall SAP strategy. It was made clear to me and my team that our role was to keep the legacy applications running for a couple of years until BW was up and running. The team quite rightly felt there was no future for them in such a role and they reluctantly abandoned ship I have to say with some encouragement from me. For my part I retreated to a nice corner of Devon and took up Golf far more fullfilling. As a shareholder I am apalled at reading this post but it would be dishonest of me not to admit that I dont draw some satisfaction from being proven right in believing that GSAP would never be delivered on cost and in budget nor would the forecast economic benefits ever materialise. And as for the proposed BW application the only thing that surprised me was that the Management Team of the day swallowed what Matula was telling them. In my opinion a career built on Powerpoint skills but I dont blame him, rather I blame the weak managment that has continually let him get away with it!!
Given the difficulties that the Downsteam business is wrestling with declining margins and rising costs it appears strange that a project which is aimed at addressing both those variables is slowed down after 7 years. Its difficult not to draw the conclusion it is simply taking too long and not delivering. Or am I wrong?
Comment by a Shell insider… Nov 3rd, 2009
I dont know this guy from Devon, but he has the right idea. Golf in Devon is much better than screwing around with incompetent management
Musings from a highly knowledgeable source: Nov 3rd, 2009
Ever Decreasing Circles:
1 Congratulations on the realisation that selling Downstream is the right process… except that you are six years too late. This was mooted in various conversations back in 03 but this was and is seen as increasing heresy by the Downstream/Finance dominated CMD. Whoops, sorry, I mean Executive Committee. How could I confuse a committee of five business leaders with the modern replacement, the EC with err… five business leaders. Paretos rule is never more prevalent than the consideration that 80% of the costs in RDS in organisational terms are held by the part of the business that offers at best 20% of the revenues and latterly profits. The strategy then was correct back in 2003 but why sell at the top of a market when there is a big slump ahead a few years later? Bear in mind too that Downstream is the part of the business where the competition authorities take the keenest eye especially on retail and this is a kinda poison pill for any potential acquirer.
2 Some genealogical details. Alan Matula was the architect of Chemicals SAP, which kinda worked, and given the nature of the local flora (this is an analogy for the Dutch folks out there) a dandelion looks good amongst a sea of weeds. However and believe it or not I got this from your own site Mr Donovan the G-SAP curse comes from a long line of inbreeds, including the EP Blueprint system and especially Shell People. The notes from Wally vdWs court case showed Messrs Matula and Dorland presenting the benefits of extended SAP based architectures to the CMD back in 2003. If you really want to bandy figures, Im loathe to suggest that $15 billion might be high but back in 2002 the cost of Galaxy alone which became Shell People was already at $800 million declared. Actually, maybe $15 billion is starting to look light J
3 The best quote on G-SAP came from Rob Routs when he was audited back in late 04. He saw G-SAP as the enabler for the cost savings that Downstream-One would deliver. He also amusingly pointed out that he would be long-retired when the benefits of Downstream-One would be apparent. He was right on that one, albeit not exactly as one might have hoped.
4 The architects of the G-SAP plan for world domination were of course the same folks who had implemented the Engineering Box and other wonderful solutions earlier on a regional basis on the doomed ship that was SEOP. It had far less functionality than the systems it replaced echoing IT4mes comments but kept a goodly few consultants in place. Price Waterhouse Coopers I believe at that point. And why not then try and implement on a global scale a system that had failed regionally? If nothing else, the original failure would disappear quietly…
5 Back to the central characters. Dorland has risen to the rank of Matulas deputy on the back of… er… well a CEP of 1 2 whilst at Pernis and a track record of non-delivery unmatched in RDS. He admitted over lunch back in 2004 that he had been offered the job of G-SAP implementation manager but was happy to stay out of the mess and let Blauwhoff fail. Funny now that he is technically responsible for implementing it…
6 Lets look at his last big SAP programme, which came to birth whilst he was parked in Global Solutions. S-ERP was dreamed up by Dorlands pet consulting firm, Accenture. This is actually where a lot of the ERP/SAP nonsense comes from, although much to their chagrin Accenture have picked up comparatively few crumbs from the IT4Shell table. Anyway, S-ERP was Dorlands path to the top table and he spent two years fudging the numbers and seeking justification from an unconvinced and perceptive Tim Morrison. Eventually it got the green light albeit in modified form and the usual swarms of Shell folks jumped onto this lifeboat. Multiple VARs and audits were postponed in pursuit of the Shell goals of Honesty, Integrity, Teamwork (geddit Shell Honesty Integrity Teamwork?)
7 Its not clear how many bills RDS missed paying or how many late fees RDS incurred but there were several billions of unpaid invoices floating through the system when the first part of S-ERP went live in 2007. It is uncharacteristic of RDS to act so quickly but the whole thing was largely cut back as a programme almost immediately, especially once Dorland got Matulas old job and was free of that albatross though albeit he surely has a flock now with G-SAP?
8 One cannot obviously speak ill of the short but one amusing aspects of S-ERP is that it was designed specifically to reduce credit card fraud. This part of the programme went its own way largely but was for a goodly period under the control of the S-ERP programme manager. Hes no longer with RDS however I will let the more inventive minds out there work out what misdemeanour rumour has it he was responsible for but heres a clue it involved credit cards.
9 Back to the chase. RDS actually tried to achieve something spectacular with its IT strategy back in the 90s but the oil industry is dulls-ville with regard to these systems and that is why the top echelon in IT4S has long been either weak or Dutch, there is no real challenge and good people dont come here. When Mike Rose quit in 2005, there was no external candidate who would take it. The single common platform (GI-D) is still a great achievement but is wildly over-engineered and has cost an absolute fortune (the budget for the planning stage alone exceeded the CMD 501 proposal for the whole thing and Harry Roels lost his job as a result). GI-D, Blueprint, Shell People G-SAP and latterly S-ERP have delivered great profits and promotions all round the former for the pet consulting firms and the latter for a particular breed of executive.
10 Full circle. To be able to effectively operate the business, one needs real cost data and G-SAP, for example, would have delivered excellent benefits of it had ever been possible and was properly implemented. Without that level of data, MIS in Downstream remains a cottage industry employing thousands and the real cost of operating a headcount intensive business in the most bureaucratically inefficient hierarchy of the Top Five (recent quote from the FT) will never be seen. The guys on the SEG salaries for pumping gas will carry on happily in pursuit of those index linked pensions. It was ever thus companies get the IT systems they deserve.
Received by email 4 November 2009
John, please add this to the IT string GSAP/GPMR: THE MARCH OF THE DINOSAURS…
Matula has been mentioned a few times in this string of posts. He was the ‘brain’ (meant cynically) behind the complete outsourcing of IT business last year. That was another ‘spectacular’ from him and earned him promotion.
Yet again a very big promise and great underdelivery and as we all know, this is the key to succes in modern Shell. It now is such a mess that even the helpdesks do not function anymore. We have 3 partners that are quite disconnected and have no idea what the others are doing.
The business now needs to contract for a simple project more than 1 project manager (from competing service outfits) because no single IT service company has all technologies in house. Obviously recipe for disaster, they can always blame the other. And watch this space: Matula will muddle on for another 1-2 years, thereafter he will disappear to another company with a great CV and the next IT head honcho can go and sort out the mess. The outsourcing to AT&T, T-SYS and EDS is becoming the mother of all skeletons in the cupboard. And that cupboard is already very full as we can all read on the Donovan site!!
roger miller on Jan 15th, 2011 at 11:25 pm
I was pleased to note a page on GSAP and the foolishness of the project. I worked on GSAP and it is much more of a disaster waiting to happen to the extent it has not happened. The whole GSAP project management seem to be only interested in hiding risk. Internal promotions are handed out to those who can hide the risk. I was a consultant to the project but it pained me to see so much of deliberate waste. Advice after advice pointing out stupid designs only invited suppression and punishment.
I did not understand why speaking truth hurt so much at GSAP. Now I do. It was just flowing top down.
It was not technically feasible to roll it out in US, whatever cr*** management wants others to believe. This was evident even in initial phase of the project.