Sent: 1 October 2005
To: Steve (– my old boss, mentor and a CIO in GE)
Subject: The Untold Story - a great case study
This one is to read during a coffee break or on the train. There's one thing that I've always wanted to tell you about. I think of it as the darkest hour of my career. On the one hand, I've wanted to tell the story, in order to come to terms with it internally. On the other hand, I was afraid that competent people like yourself simply would not believe that this situation could be true. I promise you that everything that I tell you is entirely true, without any attempt by myself to alter the facts in my favour. (I have changed most of the names.)
First, here is some background information.
Background Fact 1:
In my old consulting firm in the US, I knew a Senior Consultant called Jim-Bob. He had learned a little about Oracle (not as much as he thought) and then quit to make lots of money as an independent. In this role, he had built a virtual company, 'Jim-Bob Consulting', with a deep reputation for hype and exaggeration, always presenting itself as bigger and more solid than it really was – a company with a fantastic website that had no office address or telephone number, only anonymous e-mail addresses like info@ and sales@. Jim-Bob would recklessly win work and then hire any contractors he could find to create the apparently seamless, experienced, expert 'Jim-Bob Consulting' teams that he presented.
Background Fact 2:
When I was running an Oracle practice in the US, I had to let someone go, Bill, who had been hired before I started. I really liked Bill, but he had no Oracle experience at all, could not be deployed for chargeable work, and should never have been hired for this role. It was typical of the hiring hysteria of the late 90s that if you had somehow been touched by Oracle or SAP, you could be hired as an ERP consultant. This was not Bill's fault; he was hired by people who were ignorant and reckless, but I could not get chargeable Oracle work for him, as he had no implementation experience at all.
Back to the story:
Zebra is a UK-based IT services and outsourcing company. One of its operating strategies is to provide outsourcing services by acquiring entire departments of people from the client company, when it takes over that function. Zebra had acquired the Finance and Accounting function of UKTTint, as well as a group of Oracle support 'experts' tied to Finance. UKTTint had been using Oracle Financials in various divisions, but not in a coordinated way. I joined Zebra in December 2003, reluctantly leaving GE when I could not find another suitable London-based role, at my level of seniority and experience (I didn't want to move back to the US just yet).
Within a short time of my joining, Zebra announced the closure of its Asia Pacific, mainland European and US offices, leaving only the UK and India. My role at Zebra was to be the Programme Director of the Apollo project, which was a big programme to put all of UKTTint globally onto a single common Oracle 11i platform, with a new common Chart of Accounts. I was replacing the original Programme Director (a contractor), who left because of ill health (apparently).
The new common Chart of Accounts had many segments (I think it was 10+). This had already been decided, so was not open to review. In addition to a huge CoA, the solution also included a Data Warehouse, and some reporting tools, all from different vendors and requiring integration. (No comment here on a big CoA and DW).
The pilot site of Belgium had already gone 'live' on the new system in November 2003. My task was to ensure the successful roll-out of the system to all the other businesses, keeping to the roll-out schedule. Both the schedule and the price were fixed. I was shown the roll-out schedule.
It struck me as incredibly aggressive, but I did not say that it was impossible. At that point I knew nothing about the reality of the situation.
However, later I learned that each local implementation involved further custom development, making the roll-out schedule simply irrational (even without the other issues to come).
I then learned that the Belgium 'pilot' site, although two months old, was using about 50 of the development/support team on constant fire-fighting – urgent fixes needed just to keep the system working.
The Zebra arrangement to acquire UKTTint's Accounting department included the head of that function, Peter, who was now on the Board of Zebra. Peter was managing the relationship with UKTTint for Zebra, but it was immediately obvious that he was looking-after UKTTint's interests, not Zebra's – making commitments that were impossible and commercially meaningless.
It was Peter who told me that he wanted me to ensure that Apollo was delivered on time. Peter was very close to the Oracle team that Zebra had acquired from UKTTint, always looking after them and protecting their interests. The head of that team, Casey, was considered an Oracle expert and was generally never challenged. He had overseen the design and development of the new system, the 10+ segment chart, the DW, the reporting tools etc.
I learned that Casey had met someone in the US at OAUG (Oracle Users Conference) who ran 'Dynamic Global Consulting', and this person had convinced Casey that 'Dynamic Global Consulting' could solve his problems. (I had never heard of 'Dynamic Global Consulting', although apparently it was based in the same city as I had been, in the US). A large team of experts from 'Dynamic Global Consulting' had been brought in to work on the Apollo development and testing. (They were all Americans and this entailed travel, hotel and living expenses for Central London). Apparently they were experts in the specific technologies being used and had some fantastic proprietary testing tools that would really help.
As soon as I started as the Programme Director on Apollo, I went to the office where the development team was based. I ran into Bill, from my former Oracle practice in the US. I could not believe my eyes. Bill told me that he was with 'Jim-Bob Consulting' (as a contractor), but it was now called 'Dynamic Global Consulting'. Bill was operating as a free-lancer and had been picked-up by 'Dynamic Global Consulting' for this project in London. There were a number of individuals from 'Dynamic Global Consulting' in the London-based Apollo project – they were virtually all contractors. 'Dynamic Global Consulting' was engaged on a T&M basis (hourly rate plus expenses). They weren't able to deliver much at the moment, as they had no system to test, no access and no terminals. They were sitting together in a conference room, looking at manuals.
The website for 'Dynamic Global Consulting' was the same as the one for 'Jim-Bob Consulting' – obviously with a new name – but making the same crass claims and still without any names or addresses or phone numbers – just those anonymous e-mail addresses. Any fool should have been able to see through this.
As I slowly learned more about the project, I uncovered things which were simply not credible.
- What had been built?
- What was going to be built?
- What was all this customization for?
- What were all these people (200 resources - many or mostly contractors) doing?
- What was 'Dynamic Global Consulting' doing (on a T&M basis)?
No-one knew. No-one could tell me.
There was no documentation for:
- Business requirements
- GAP analysis
- Functional specifications
- Technical specifications
- Set-up configuration
This system existed entirely in the minds of the ex-UKTTint Oracle team, who apparently had always done it this way. Not only that, but there was no philosophy of minimizing customization. It was the reverse. If a user even mentioned a requirement, this team would just go ahead and customize it, without even exploring standard functionality. They would put changes into Production with minimum formality. (I know that this sounds unbelievable, but it is true). Nothing was documented.
The idea that we would eventually transition Support to an India-based team was a complete fantasy.
The plan was that there was even more customization to be done, specific to each of the successive planned roll-outs. Change requests were still coming in so fast, they had a team of people just to track these, but actually implementing them all was impossible.
Another point to mention is that Zebra had laid-off virtually its entire pool of Oracle resources sometime before this project started, strategically choosing to focus on SAP instead. Zebra then had to engage a huge number of contractors to staff this project – a project it was not qualified to deliver and should not have become involved in.
Oracle Corporation was commissioned to inspect the Apollo system and their report was so damning, they said that they would not support Apollo because of all the illegal coding in it.
The 'go live' decision for the earlier Belgium pilot was completely flawed. It had been driven by political expediency and Peter's need to save face. The system should never have gone 'live' and now needed a big team to constantly fire-fight the endless problems.
Peter (I was told by an insider) had promised another release specifically dedicated only to fixing Belgium's issues, but this had been immediately forgotten in the rush to further customize the system for Germany and Finland. I was never told officially about this promise.
I tried to shake things up a bit, but Gladys (my boss on the programme) kept blocking my every move. She kept hovering over my shoulder the whole time; there was no sense in which I was being allowed to function as the 'Programme Director'.
Early on in my involvement, there was a meeting with UKTTint management about the 'lessons learned' from the pilot implementation. As Programme Director, it was obvious that I should be at this meeting. But, just as it was beginning, Gladys sent me out on a 'change control' errand which she insisted demanded my urgent and immediate attention. I missed the 'lessons learned' meeting completely. With hindsight, I see that situation completely differently now. Gladys did not want the new Programme Director to actually learn any facts about how the project had been conducted. Not being very political, I had no suspicions at the time.
My job apparently was to miraculously fix all the problems and deliver the project – keeping to the promised schedule.
One thing became absolutely clear to me. Even though my deliverable was to ensure that Apollo was rolled-out according to the schedule (and fixed price), this was not the right thing to do. The system built so-far not only could not be rolled-out according to the schedule, but should not be rolled-out at all. It was not fit for purpose.
Nobody in Zebra wanted to hear this, but it was the only thing that I could say with honesty and integrity. Of course, I could not convey any of this to UKTTint – they were the client – but they really needed to be told. It was a matter of integrity. This software would jeopardise any of their businesses into which it was implemented.
I must have seemed inept to the client, but I could not tell them the truth and I did not want to propagate the lie, so I said very little, conducting my analysis and hoping that we could correct this within Zebra.
I was feeling deeply concerned about Zebra's conduct in this, and very uneasy that I was now a part of Zebra. I felt that my GE-instilled integrity was being severely compromised.
Now, I’ve been in difficult and stressful situations before, of course, in my career. The most significant situation which comes to mind was back in GE. It had been my idea, my vision, that our newly-created and small business unit, GESS (GE Software Services), could meet the Oracle consulting needs of the huge and demanding GE Power Systems. Once I had persuaded them of this, I was not allowed to walk away. They told me, in effect, ‘You think this is a good idea … now you deliver it!’ But they never gave me the two months’ lead time I had always said was essential for us to ramp up resources. Those days in Atlanta were extremely stressful, and the time when my first grey hairs appeared. In a very short timeframe, we built that business from about 25 to about 150 consultants, to meet the needs of the GE Power Systems project, and ultimately that programme was successful and had a very good reputation around GE. I became a Vice-President in GESS on the back of making that opportunity successful. But the difference was – although GE was a very demanding environment – people were professional, competent and honest. I was now in a situation where some people were not professional, not competent, and not honest – and I was struggling to believe this and come to terms with it.
I was also wracked with doubt sometimes. Perhaps these people are smarter than me and I'm just not seeing it right?
I learned through incidental conversation that a particular person (Steve L) from Zebra had been involved in the project at an earlier stage – a name I recognised and respected from my KPMG days in the early 90s – and this person had been raising serious concerns about the project approach and the quality rigour. He was removed from the project by the senior people. (As soon as I learned that Steve L had been removed from the project, alarm bells should have been ringing in my head.)
So, in Zebra, in true six sigma fashion, I produced a one page pitch to brief Zebra senior management on the utter seriousness of the situation, and the need for a complete reappraisal of the plan. UKTTint really needed to know this, but, of course, I could not convey it to the client – that would be the job of the senior-most Zebra execs. What I said about the over-complicated design, the reckless customization philosophy, the poor build quality, the insane roll-out schedule, the use of 'Dynamic Global Consulting' etc etc threatened Peter, Casey and my boss (on the programme) Gladys.
My mistake was that I did not present this directly to the very highest level in Zebra. In fact, I was blocked by Gladys from seeing the MD, Sven Eastman. ("You don't need to see Sven Eastman.") I presented it only to Gladys (who, along with Peter and Casey, was largely responsible for the mess). I was not looking to assign blame; I wanted to steer us towards reality and the need for a complete rethink about the plan and approach. I did not want to roll-out this 'solution' any further.
Gladys was a long-time employee of Zebra, who had reached the highest level below the Board, and had a reputation for ruthless perception management of her own career. When I sketched out the pitch to her on a piece of paper, and explained why Apollo could not possibly be delivered according to the schedule, she listened in complete silence and uttered not a word. Not a single word!
Just a few days later, I was removed from the project by her and Peter. I told the head of the ERP practice all of this. She listened sympathetically, but said nothing. Later it emerged that she was in the process of closing another position and she left the company shortly thereafter. (Before she left, she warned me: "Gladys is very powerful.")
There was no Oracle work being won; in any case it was unlikely that I would be assigned to it, as Gladys and Casey were closely entwined with the Oracle practice. I was let-go from Zebra in August. (Thank you, God!).
Three days later my Dad passed away –suddenly and completely unexpectedly – and two weeks later I received an offer to return to GE, in a London-based role.
It was a wonderful learning experience!
Sometimes in a situation there is no way to win within the parameters they fix for you; you just have to thine own self be true.
The funny thing is that in the period immediately before I started at Zebra, when I was meeting with them (Gladys and the former Programme Director, who was transitioning out) and they were telling me about the Programme I would be running, I had a strong feeling not to join (even though I had accepted). My inner voice was telling me to run away!
But, the salary was good, there was free petrol (gasoline) and it seemed like a company going places. (This was before they closed the Asia, Europe and US offices).
A year later I learned from an insider that – after my departure – the project continued its agonising struggle for another year, but there were no further 'go lives', just more Programme Director changes. Eventually, the plug was pulled completely and the resources were instructed to pack up and leave the client site.
- When the project was being presented to me and I had the intuition about it, I should have refused to take the position, if necessary reneging on my signed contract.
- I was naive and failed to understand the political manoeuvring that was going on, such as Gladys controlling me and keeping me out of meetings that would have given me more insight, sooner.
- Despite my years of experience in the corporate environment and the many difficult situations I’ve been in, I’m still stupid enough to think that everyone from the same company must be on the same team, and that really senior people can’t possible be incompetent.
- I should have gone over the heads of Gladys and Peter and taken my assessment directly to both the CEO and the Chairman of Zebra, preferably early on, but even after I was displaced from the position.
Remember: telling the client the truth about the situation and giving them a sense of reality was not an option for me, as an employee of Zebra.
I hope that you enjoyed this account and I promise you again that it is all true.
Author’s Note: The names have been changed, but the account is 100% true.
A couple of years later I was told by some ex-Zebra people that it wouldn't have made any difference if I had voiced my opinion to the highest ranks – they would have ignored me, as the truth was too uncomfortable.
A very senior UKTTint employee (who had been a colleague for many years earlier in our careers) told me that UKTTint was furious that Zebra had kept the truth from them for so long, and no longer trusted them to do this kind of work. He said I should have told UKTTint directly the truth about the project, but he appreciated that I could not do this because I worked for Zebra.