£90m.. its not going cost us anything..
"Its the tax payers money, not ours!", said the Civil Servant.
MI5 has reportedly abandoned a planned £90m upgrade to an intelligence database after the delayed IT project failed to meet its requirements. The record management system was supposed to be up to speed in time last year to tackle the threat of a terrorist attack on the London Olympics. Designed to collect intelligence data and …
One of the most common complaints you hear during project reviews is "but we can't stop now, look at all the money we've spent!" Sometimes the smarter thing to do (and often the braver in terms of career prospects) is to upset some people by identifying a failure as a failure, pulling the plug, and stopping further money going to waste. Yes, £90m may have "gone to waste" (ignoring that hardware and software plus the experience gained could be re-used for another project or the second stab at the same project), but that's a lot better than letting a project meander through years of revisions and possibly billions of pounds before someone finally admits defeat. Whilst the headline will be (predictably) "90m wasted" that's probably not a completely realistic viewpoint.
The question is were Deloitte incompetent in identifying the requirement, or did they just build what they were told to build and now have been unfairly tarred as the cause of failure?
You have summed up one of the definitions of the "sunk cost fallacy" (sometimes called the sunk capital fallacy). Wilipedia link
I came across it when I was running a scientific enterprise for a banker. He had bought a laboratory that lost (a lot of) money. I had just got it to the point where it was breaking even and stood a reasonable chance of making a small profit. He killed it - When I asked why he told me about sunk capital. The point was that even though we had spent the money, it was unlikely that any future effort we put in was unlikely to make a reasonable return. It was better to write it off and spend time and money on something else, rather than to continue just because he had spent a lot of money (and we had spent a lot of time and effort) on it.
<-- I raise a beer glass to one of the best bits of business advice that I was given...
"Sometimes the smarter thing to do (and often the braver in terms of career prospects) is to upset some people by identifying a failure as a failure, pulling the plug, and stopping further money going to waste."
Perhaps. A braver and more useful thing still would have been to have identified that the system was a POS before reaching the fat end of £100m.
Given that this never got anywhere near operation, it would seem likely that the major part of costs would be labour, and on that basis we can guess that say £60m went on people's time. Assuming a generous £800 average day rate that's about THREE HUNDRED AND SIXY MAN-YEARS. If it takes that much effort to identify that something isn't going to work then all involved should be herded into a prison ship which should then be set on fire and sunk in deep water. I have, incidentally, recently worked on a government "managed" project involving one of PWC's competitors, and they have have toiled ceaselessly to build their time sheets towards the target pay out, undistracted by such piffling concerns like whether the system works, whether it is secure, reliable, or effective. The output is pure 5hit, delivered late, and barely functional, they have ignored all those who know better.
Management consultants: Useless f***ers.
He had a similar systems upgrade issue - his key systems were outdated and unable to provide the plans for world domination that he needed.
Unfortunately (or fortunately, depending on your point of view), he used <insert which ever consulting firm you dislike the most> and he has been unable to do any evil for the last 30-odd years.
If memory serves - evil supervillians always seem to be very good at large project management.
You never saw the underground layer half covered in scaffolding because a contractor was late or a plot to blow up Fort Knox stopped because an IT system had been abandoned.
Perhaps it's the shark+laser beam based negotiating tactics - but government IT negotiating with CapGemini / IBM / Whatever-Anderson-Is-called-Today should probably be issued with a white cat and a nehru jacket
I'd offer to help, but I wouldn't want to work for any organisation that wouldn't have me.
I'd like to say it's a true story, that someone going to work on the project said to me "Well I'm in programme management, so it doesn't really matter what the programme is." but that would be a lie.
Meanwhile up in Imphal Barracks, all the people who could actually implement something like this, aren't able to get their CV through the door, because you can only have DV clearance if you've already got it, or are the daughter of a friend of the counter terrorism head of the met, or a special close friend of a defence minister.
Mongo DB is web scale.
Additionally, they need a few secretaries to act as business analysts, and some "Performance testers" (because surprisingly, performance problems aren't self evident.)
Then what they need is a good old principal project manager at their supplier, who really knows how to trick the client into changing the requirements, because more requirements means more people on the day rate.
What could possibly go wrong? I mean, screw using your ability to listen in on worldwide communications to find people who don't screw up. Instead, take the view that big consultancies are big because they're good, not because they screw more money out of customers.
As you are quoting the Independent's story you missed a cracker of a quote -
"Among the added IT expertise brought in were a team of expensive consultants from Deloitte."
Yuck yuck yuck (in best Muttly fashion). At some point UK governmnt might just decide to invest in their own experts - at least fuck ups like this will be cheaper!
This post has been deleted by its author
The UK's Security Service drafted in consultants from Deloitte to assist in delivering the project, which has become the latest in a long line of failed multi-million pound UK government IT frauds.
There, fixed that for you, John. Haven't you yet twigged that that is the problem today. Far too many beating about the bush and being all coy and mealymouthed and deferential about serial incompetence in incumbent incompetents.
I wonder if A. Parker will be such a corker and continue working oversight in the traditional cock-up manner?
I would really, really like to get Deloitte's take on this. What they say happened, and why.
Was it a case of shifting-sands requirements? Did the MI5 project bosses even have a clue as to what they wanted? That's the killer with these applications that are not well defined - everyone can agree they want a new intelligence sharing database, but HOW it is to do that, and for WHOM, using what OPERATING PROCESS MODEL, and offering what SECURITY of what DOCUMENTS...ahhh, that is hard. And if it is not a well-defined, common business process (say keeping the books, or operating a call centre) then there is little "best practice" to crib from.
My guess is that MI5 leadership probably couldn't get their heads around any of those questions, or kept changing their minds in circles. I was recently working with a UK governmental agency that had a similar problem - because apparently once you take the profit motive away from an organisation, it is difficult for them to actually know what it is they are trying to DO, and optimise for it. Turf wars abound, and determining priorities and KPIs is near impossible - or at least getting leadership to agree on them. Again, because without profitability as a key metric, everything else can be argued qualitatively to death. "We need to do this as our top priority, but we also can't not do this other which is equally important...and in opposition." I left that project saying that we didn't need a systems design, we needed the top executives locked in a room for a few days to actually agree on what the organisation wanted to be and how it should be run, because everything we were going to design was bollocks without those decisions first. They might get to that conclusion in a few months...maybe.
I highly suspect that similar leadership/organisational issues are at the root cause of many UK government project failures...
My guess, although I confess to absolutely knowing nothing about it, is that someone else has managed to lay off shit, because Deloitte have taken the flak, because they were sent in as trouble shooters..
Of course that's the problem. Troubleshooters are born, like innovators. One thing they aren't is herd together like cats, er, cows. So it's not Deloitte totally to blame, but they're not blame free either.
Although I completely made all that up, knowing nothing about it, as I confessed earlier.
Uncertain requirements analysis constantly shifting x feature creep
"We can't get a straight answer out of them, lets go to plan B, "income protection," charge them for every change.
The only full figure given for the Govt Interception Modernization Programme was £12Bn. Does anyone think that would have been any better managed?
Does anyone think the £500m the Snoopers Charter wants to dish out to ISP's is the whole budget?
Biting the hand that feeds IT © 1998–2021