Keeping on with the Coasian hell series, what about Capita? Like everyone else I was watching the shares melt down of a morning and read their Operation KITCHEN SINK trading statement. This quote stuck out at me:
We have reallocated our previously centralised business development capability to the divisions, bringing it closer to the markets we serve. This reduces complexity and increases accountability, in relation to the bidding, implementation and management of contracts, as we seek to improve sales performance.
To put it another way: they let the sales force be in charge of the business. This is something I remember my old boss saying; if you let sales run the business, what they want to do is sell, and they never understand that you make money on the deals you don’t do.
Of course any business must sell. The problem is that they’ll promise customers the moon on a stick, or equivalently, drop prices to a level you won’t be able to deliver. In Capita’s case, they would then chuck the job over the wall to the relevant operating division, and run off to seek new nadirs of debauched depravity, &, &, sorry, average contract value.
This is a classic business problem and not that interesting in itself. Also, Capita at least had the decency to cheat itself and its shareholders rather than stiffing thousands of blameless tradesmen over months’ worth of bills. But it does sound quite a lot like the business model Carillion and the prime contractors operated: the bit that mattered was the central contract-hunting speciality, and delivering on the contracts took a back seat. Part of the problem was internalised rather than put out to contract.
Capita claimed it had more to offer than just being a soft layer between public sector customers and small construction tradespeople. It’s an IT consulting shop, after all – it’s meant to know about computers, a useful speciality in itself. For example, they successfully implemented the London congestion charge, a big, dramatically new, complicated, and politically exposed project that worked on day one. They failed on a lot of other stuff, of course.
Importantly, though, they also took on the operational responsibility for a hugely diverse variety of public service projects. Rather than a design, build, transfer, and operate contract, they preferred a full business process outsourcing project where they would employ and manage everyone involved. A whole weird variety of administrative functions were redefined as technology projects, precisely because this meant they became Capita prospects. This took them a long way from IT consulting and software development towards being a Carillion-like general-purpose prime contractor. They went further down this road by acquiring competitors for their contracts and piling up goodwill on their balance sheet. The most damning evidence of this is that they now acknowledge they need to spend heavily on internal IT, or in other words, on the stuff they are meant to be good at.
Years ago, during the eventually successful campaign against national ID cards, blogging legend Chris Lightfoot posted some old policy papers about the National Identity system that ran during the Second World War and a few years after. The paper-based process was discussed in such detail, he said, that it amounted to debating the SQL statements involved in the House of Commons. I think he thought this would be a good thing. Snark aside, the elite felt it worthwhile to understand the process.
Somehow, this intense attention to systems design has been lost from the government and from the political debate. (Today, you might get rich because you do computers but you’ll be channelled into the role of an eccentric billionaire.) Instead of it, we blame “computers” and then, if that doesn’t get people to go away, the outsourcing contractor. So many problems that are spoken of as “computer systems” problems are systems problems, that just happen to be implemented using computers rather than paper or giant rocks or whatever.
In important ways, this is the service that Capita provided and still provides: the ability to blame problems on computers and computer people, while ignoring the physical reality of policy.