Police Scotland and Accenture were at odds over ill-fated IT project i6 An audit into Accenture and Police Scotland's disastrous attempts to develop a unified IT system has found that the project collapsed because Accenture underestimated the programme's complexity and the resources needed to develop it, alongside a breakdown in the two parties' relationship. Published this morning, the report by …
Yes. And that'll be the same Accenture who employs the "consultants" I e-mailed yesterday. Here's is a line from that e-mail....
"It might be better for people – especially, it would seem, experts brought in from Accenture - to ask me specific questions rather than continually guessing."
The rest of the e-mail was along similar lines.
"Police Scotland was committed throughout to working with the supplier, Accenture, and the Scottish Police Authority, whilst maintaining the integrity of ongoing discussions during that process and commercial elements of the contract."
"There were challenges and issues on both sides, but we worked closely with Police Scotland ........ and we mutually agreed to end the project."
They're not at all 'at odds'. They both agree that they did the best they could under the circumstances. I'm sure they'll be dating each other again in the future.
They're not at all 'at odds'. They both agree that they did the best they could under the circumstances. I'm sure they'll be dating each other again in the future.
I don't think so. An alternative theory could be that they tried the usual Change Control racket to add charges despite the fixed price and got unexpected pushback which meant they had no way to recover what they underbid in the first place to win the contract.
Colour me impressed if someone managed to outscam them :)
They only lost a few ten's of millions of pounds in the process and it was resolved in a few years.
Given the history of government and public sector contracts, that is progress as normally they would replace the project teams three or four times before admitting that nobody really wanted to do this in the first place...
It's how the public sector works. You want to do ABCDEF, but there's no way you can get the budget for it. So you do ABC, and hope to tack on DEF as change requests later. A couple of years later, once the project has actually started, you are shocked, shocked I tell you, to discover that you also need DEF. The budget simply must be produced.
In the old days when government was flush with cash, that usually worked out ok. These days it doesn't.
If Police Scotland had been honest up-front about their IT requirements, the entire project would never have been signed off, and they'd be doing things on paper. That may be no bad thing.
"If Police Scotland had been honest up-front about their IT requirements..."
It sounds to me like Police Scotland were honest and up-front about their requirements. Accenture spent months assessing the requirements. The disagreement was whether Accenture's solution would meet the requirements, not that the requirements were changing, because changing requirements are where consultancies make their money.
The fact that Accenture gave a full refund plus compensation suggests desperation to avoid a court case and a desire to protect their reputation, whatever that might be worth...
"So they signed a fixed-price contract, but allowed the client to make changes? And both sides were surprised that it didn't work out? They never learn."
From the settlement I'd guess it wasn't the Police making changes to what was required, otherwise the requirements would differ from the agreed contract and there'd be no reason for Accenture to not only pay back what they'd already been paid, but also more than that again in compensation. That's why people like that don't mind requirements changing... when things don't work out you're no longer in breach of contract since the customer changed what they wanted.
I imagine the requirements given initially stayed the same, but they underestimated what would be required to fullfil them and perhaps failed to properly find what was needed. If the Police say that every officer needs x, you base your costs on a guess there are 1,000 officers and it turns out there are 5,000 officers, it's not the requirements that have changed.
From the settlement I'd guess it wasn't the Police making changes to what was required
Fom the article:
"As the report acknowledges, the scope and the complexity of the solution for i6 increased significantly during the project. This was driven by the client ... it was not possible to agree the necessary changes and we mutually agreed to end the project."
As I said, the requirements didn't change. To also quote the article, "Accenture underestimated the programme's complexity and the resources needed to develop it"
What's missing from the Accenture quote you requoted are the reasons why the scope and complexity of Accenture's solution increased. "This was driven by the client" could easily mean Police Scotland knew exactly what was required and were holding Accenture to deliver it as agreed.
If the Police say that every officer needs x, you base your costs on a guess there are 1,000 officers and it turns out there are 5,000, it's not the requirements that have changed.
the first question anyone competent should ask is, and how many officers, not just guess, then get that in the requirements. They should also ask about projected staffing increases over the next few years, at least over the run of the project.
My guess would be that they intentionally under quoted to win the contract, and were going to use change requests to bump the costs back up into profitability, but then found out that there weren't any/enough change requests to actually get the price up to where they could deliver without making a loss.
"Accenture paid £24.65m back to the Scottish Police Authority the following July, refunding the £11.1m that had been paid to that date, as well as a £13.6m settlement."
Hats off to the lawyers and procurement personnel at Police Scotland who appear to have drawn up a better contract than many of their colleagues (in most public sector IT disasters the vendor seems to keep the money and get compensation, here it's the other way round).
So the contracted result would have to work for about 1 1/4 years perfectly to pay for itself.
Instead it didn't work at all.
Hats off to Police Scotland (or should that be Polis Scotia as First Minister Sturgeon would like them rebranded ?)
I suspect that merging the 8 (?) forces in Scotland will be like the back offices of the various UK councils. It will need to be done gradually, with the consent of the various IT departments IE bottom up, not top down.
Oops.
That should of course have been 2 1/4 years to break even.
I'll take a wild stab and say it'll be a lot easier to merge some of these forces IT systems together then you'll have a 2-3 clusters left whose structures and way of doing things will be so different that merging them will be a serious PITA.
OTOH how "real time" do you need your data to be? "instant?", Every few minutes? Hourly? Nightly?
Apparently clumsy (but fully automated) linking existing, working systems may beat a new centralized super duper (but overdue and highly buggy) SoA package.
Just follow the money. If the police had kept changing their minds about what they wanted, you can bet Accenture would be happy to oblige, and demand payment in full if it all went fubar.
The fact that they refunded AND compensated the police would suggest the audit is correct.
If only every public* sector IT project was as tightly governed.
* And a few private sector projects I've witnessed...
There's ideas being mooted of merging some of the remaining health boards and/or potentially parts of councils too. I can see the merit in some of it, but as always with IT there's a lot of contracts which need to expire etc for it to start happening without a huge amount set aside for buying out/penalty clauses.
What I don't get is why England can't do something similar, if anything everything there is becoming more fragmented year on year.
Lots of other companies were approached and avoided this like the plague, saying it couldn't be done in the way the Police wanted it done, and that the task was huge.
They may have refunded a sum of monies, and ICT are keen to stress we didn't lose out of it... what about the i6 Department that had over forty cops, up to Super rank working in it? What about the hundreds of thousands of pounds on their wages? Plus the cost of re-fitting Anderston police station to make it the i6 hub.
The whole thing was a joke, and lots of Superintendents were sucked in with hospitality and top class treatment.
1: The 8 Legacy Forces of Scotland -are- merged into one Force. We use in the main, the same systems which are generally accessed via the Intranet, but there are several different systems still in use. Primarily, these are the systems for recording crimes and the investigations into those crimes. It's not a huge issue as this is not information that is ever really required to go Force wide.
2: i6 cost a lot more than the quoted figure. The i6 department was around 40 cops, up to Chief Superintendent rank. Over three years, that's over one million pounds on wages. That's not coming back. We also lost 40 cops on the street to this department, although some were light duties and not operationally deployable
There were also two sites in Greater Glasgow that were refitted for i6 at great expense.
3: i6 caused the biggest problem in that none of our current systems were updated, because i6 would bring the full solution. As such, we made do with crappy systems that were old and not fit for purpose. We're now scrabbling around trying to find solutions to this future gap, and it's not easy. Some very important enquiries are done on MS Word for fucksake, meant to be accessible to all to update and track the investigation. A nightmare. There's one key system that was designed in house and the guy died / retired years ago and no-one quite knows how to update it!
4: There was almost no mission creep from the Police. We knew beforehand what we wanted to replace or integrate, the systems existed in different styles around the Force, we had everything roadmapped. Apart from some small extra tweaks, nothing big changed. Accenture simply couldn't cope and massively under-bid hoping to scam us.
5: Lots of other companies were approached with this, and they all said it's too big for them. That's coming from some blue chip IT firms.
6: The reason Accenture paid up, was that because when relationships soured (which they did) and we reached a state of no return, and things failed, the plug was pulled and Accenture started being dicks. The 40 cops who had been working alongside them were then re-tasked from developing i6, to conducting the investigation, gathering evidence and putting together the court case against them for charges relating to breaches of contractual agreements.
That, is why they paid up. :)
I often wonder whether some of these integration projects should be done in-house. The problem with many insultancies is they do not understand the client's workflow and the nature of the routine tasks. In this what really needs to integrate and can be safely ignored for later. Internal staff often have a better feel for this.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
