The i6 system, which was contracted to Accenture to design, build and implement, for the sum of £46.11 million, failed because of poor design, over-ambition, limited understanding of user needs, vested interests, terrible contract management, bad relationships and poor buyer capability.
The whole process has resulted in wasted resources, wasted money, wasted time - setting Scotland’s police forces back a number of years.
The results of the failed project in an Audit Scotland report should be used as one of many examples of why big government IT projects simply don’t work and aren’t fit for purpose. It’s a story I’ve written many times before, and should be used as evidence against those that would like to see government go back to its old ways of contracting.
Whilst many say that the change delivered in Whitehall and across the broader public sector has now been embedded and is irreversible (with a focus on agile delivery, digital technologies, and designing around the user need), I am beginning to see examples, and hear stories, of why that’s unfortunately not the case.
Old habits die hard, and there are commercial interests at play. Let’s also not forget that big government IT is perceived as ‘less risky’ for many buyers in government, and there are vested commercial interests that like the idea of continued outsourcing. Equally, we have seen outsourcing agreements being extended and a number of supplier opportunities that sound awfully like projects of the past.
Throw Brexit into the mix, which is taking up a lot of resources and focus - there is a possibility that the continued focus on delivering digital change in government could be sidelined.
Problems from the start
The i6 system had been intended to hep Scotland’s Police record, manage and analyse information. And after a long market analysis, where it assessed over 100 suppliers, it ended up contracting with Accenture, based on its previous experience with police systems.
In fact, Accenture and Police Scotland believed that Accenture would be able to implement the system based on a previous system it designed in Spain - with a few bespoke add-ons built in to ‘top-up’ for Scotland’s requirements.
However, surprise surprise, soon after the ink had dried on the dotted line, disagreements began to take place between Accenture and Police Scotland about the original requirements and scope of the project. The report states:
The i6 programme team believed that the functionality of Accenture’s solution did not meet the requirements it had agreed in the contract. Accenture maintained that Police Scotland had not specified a detailed description of business requirements. This issue had not emerged during months of pre-award dialogue.
Accenture also believed that it had set out clearly what its solution would do and maintained that Police Scotland, as part of procurement process, had accepted its qualified solution. Accenture stated that, to meet Police Scotland’s interpretation of requirements, it would require more time and money.
Reading this, I want to both laugh and weep. It’s so textbook, it’s almost satire.
So many IT projects in days gone by have faced a similar situation, using the waterfall approach to design and build. The requirements of the system are scoped out entirely from the start, everyone agrees this is what is needed…and then there is shock and horror when further user research shows that this isn’t the case and changes need to be made. But going back a step and making changes requires more money, obviously.
And this is how public sector IT contracts have worked for decades. The buyer says what it wants (without really knowing what it wants). The supplier agrees to supply that (without really knowing what the buyer needs). The buyer then asks the supplier to supply something, which the supply then says it didn’t agree to. The supplier then asks for more money to make the changes. The buyer says no. The supplier doesn’t make the changes. The system is then not fit for purpose. And the contract gets cancelled, with the buyer being blamed for poor contract management.
This is the story of i6. Albeit - this time, it seems Police Scotland had a sturdy enough contract to recoup some of the costs down the line when it was clear that the system was going to be a failure (£24.65 million). The report states:
The two organisations agreed that continuing the high level design phase would allow them to quantify the gap between Accenture’s solution and Police Scotland’s requirements. By end of this phase more gaps had been identified. Accenture estimated these would cost an additional £1 million to fill, which they agreed to fund.
Police Scotland's i6 programme team and the i6 programme board repeatedly expressed their frustration to Accenture about the disagreement, particularly when they had followed good procurement practice and spent a considerable amount of time discussing system requirements. As the contractual dispute continued, relationships between the organisations were strained and trust was limited.
Waterfall not fit for purpose
There is a growing train of thought in central government that the public sector should re-introduce some of its old-school waterfall practices. Reading the report on i6, you’d be surprised that this was even a a consideration.
By February 2015, the programme as seven months behind the original plan and Accenture was developing far more from scratch than it had originally anticipated. There was plenty of back and forth between the buyer and supplier, with varying levels of optimism and doubt over the years. However, the report states:
Accenture maintained that Police Scotland was extending the programme’s scope and this required a greater degree of bespoke development. Police Scotland maintained that there was no extension to the scope beyond changes agreed through the change control process.
The Auditors put a lot of the blame on the use of waterfall as an approach. The report reads:
The waterfall approach contributed to the fact that the Police Scotland only discovered the true extent of problems with the system when it was delivered for testing. Although Accenture had provided Police Scotland with demonstrations of the developing i6 system, it was after a period of testing that the i6 programme team reported to the programme board in August 2015 that there were:
- critical errors in the technical coding
- higher-than-projected levels of flaws that Accenture was not able to resolve
as quickly as expected
- serious concerns raised about the criminal justice module, which did not
comply with the Integrated Scottish Criminal Justice Information System
- errors in the search and audit modules
- problems around the limited functionality in the administration module
And the kicker? Despite the poor user feedback, Accenture had already received payment for successfully delivering this element.
By December 2015, Accenture reported that the work still required on i6 would take an additional 30 months. This would mean go live would be delayed until 2018 and the cost would be many millions more than the original contract price. Police Scotland rejected the proposal.
The contract was terminated and the project abandoned, with Accenture having to repay the SPA £24.65 (which often doesn’t happy in situations such as this, due to poor contracting on the part of the public sector buyer). The report states:
There was a need to modernise police ICT systems six years ago when the procurement of i6 began. That need has not been met. Police officers and staff continue to struggle with out-of-date, inefficient and poorly integrated systems.
This also hinders how Police Scotland interacts and shares information and intelligence with the other parts of the justice system. There is an urgent need to determine what the next steps should be, and to carry out an honest assessment of how to procure, develop and deliver the much-needed police IT system.
A timely reminder. So much of this is stuff we have seen time and time again. And yes, there are things that are wrong with GDS and the current approach in Whitehall. But when I read stuff like this, it makes me wonder, is it as bad as it was before? And the answer is: absolutely not.