Wednesday, 3 April 2013

Failing to Learn From History.....

All the whispering about the new Universal Credit software system leaves me with a distinct sense of deja vu. The same warnings about government IT projects not having a good track record, the insistence from on high that things are "on time and on target," the same assumption that melding several disparate IT systems into one seamless entity is a fairly simple task.

Back in the early days of the Labour government, the piloting of the new Education Maintenance Allowance was carried out by a number of local authorities, using a range of assessment and delivery methods, including vouchers, bus passes, etc and a cash payment system. My local authority opted for the latter and developed software which utilised a system which is widely used in schools to create an integrated assessment and payment module. 

A couple of years later, the government decided to pilot another new payment for learners as part of the Skills Strategy (remember when governments actually had strategies?). The Adult Learning Grant, which was intended to help those over the age of 19 who wanted to return to full time education to improve their qualifications and thus their employability, and the DfE* again asked my local authority to run the pilot using similar software to that which had worked for EMA. Creating a new ALG module was a fairly simple addition to the system and required only minor tweaking to take account of the different eligibility rules for new new scheme. A team was set up to run the pilot in a few areas across the country (not locally) to determine whether a central assessment and payment body could process applications from differing geographical locations. The first year, very small in scale, worked well and the following year it was extended to take in more regions.

When the time came to roll the EMA programme out nationally the new provider (a well-known government outsourcing specialist) developed its own software system to process applications. There were the inevitable teething troubles, with students failing to be paid, or their applications not showing up on the system, etc, and we gloated slightly as we watched the private sector tripping itself up through its refusal to listen to what it had been told might be pitfalls before they started. Eventually, things calmed down and students started getting paid on time.

In the meantime, we had also undertaken a couple more pilot schemes, all of which had different entitlements and eligibility criteria. Again, new modules were created within the main software suite, but all were free-standing and custom-built to serve a specific purpose. Year on year (and you'll note, re: UC, not just a few months of piloting) these schemes were developed and implemented, trying things out, keeping what worked and abandoning what didn't (online application was one of the elements abandoned) until the government could be sure that everything was working well.

And then came  the 'big idea'.......

"Let's roll out all these grants and EMA nationally, through one contract and one central computerised system." they said. Of course, this meant that only the big players could tender for the work, as EMA was a huge deal. Thus, the local authority which had a proven track record on development and delivery was effectively prevented from tendering to run the 'small' schemes as a contract separate from EMA.

The national 'learner support' contract was unexpectedly awarded not to the company that had already been running EMA for 3 years but to another BPO company, so the computer system they had devised and improved on was now also surplus to requirements and a new, integrated system had to be built (another outsourcing giant had already dropped out at the bidding phase when they realised they would have to build their own software rather than inherit an existing one). 

Those of use who had been brought across (against our will) to the new private contractor were told that we were "the experts" and they would need our expertise to get the new system right. It won't come as much of a surprise to learn that our warnings that certain things wouldn't work (based on having tried and rejected them at the pilot stage) were largely ignored and that this, coupled with a few genuine technical problems, resulted in a disastrous first year, with backlogs and missed payments, (rightly) unhappy students and a call centre overwhelmed with enquiries and complaints. The new integrated system was so bug-gy that eventually, ALG was returned to its old system for processing and payment.

As a result of this debacle, the initial national contractor lost the contract and the former EMA contractor was brought in to rescue it. We were all TUPE'd again (although the original BPO's senior management all left shortly thereafter) and another attempt was made to build an integrated assessment and payments system. 

"Close, but no cigar", as the saying goes. It was better, possibly because the 'new' contractor did at least have some prior knowledge of what could go wrong, but the integration of all schemes never did come off in the time I was there. All the schemes had different eligibility criteria, so the algorithms required for determining entitlement were fiendishly complicated. Some were paid weekly in arrears, others were paid termly, or part-termly, in advance, some were dependent on weekly attendance being confirmed, others weren't. The attempts to bring across historical data (vital for fraud prevention) proved almost impossible. And none of this involved real-time data!

When I left, by mutual agreement (we agreed that they didn't want a manager who knew a scheme inside out and backwards), the problems still weren't fully resolved. And I'm not sure they ever really were, right up until the time the coalition pulled the plug on EMA and ALG as being 'unaffordable.' 

I don't think I've compromised any commercial confidentiality here. Anyone who knows about education will know who was who, and there's nothing here that hasn't already been covered in the media. I just worry that if one of the country's biggest (and least incompetent) outsourcing 'big guns' struggled to create an integrated system for something as relatively straightforward as student finance, how on earth (given the alarm bells already ringing) is UC going to be introduced without causing mayhem for millions of Britons on the lowest incomes and the most precarious financial situations?

*I've used 'DfE' here for simplicity's sake - the department underwent several name changes during this period and I can't remember what it was at this point.