EXCLUSIVE: How not to commission a complex IT project

Going around the country trying to spread the word about the Post Office Scandal (whilst also hoping to sell copies of my book) has brought me into contact with a class of bright, often retired, professional people who are appalled at what happened.

Some even have direct experience of the story, usually from an interesting angle.

Every now and then we strike up a correspondence. The following piece is the result of a chat which started with John’s partner Pam getting in touch after she came to see one of the presentations I gave earlier this year.

John Murray (pictured) is a now-retired project manager whose most recent job was working for the Cabinet Office as part of its Corporate Services Improvement Programme. Before that he worked for Barclays Bank and before that… the Post Office. For fifteen years.

John has very kindly written what I think is an important piece about the disastrous origins of the Horizon IT system. I have already suggested (in the light of Dave McConnell’s evidence) that Terry Austin might be recalled to the statutory inquiry. I hope that in the light of what John Murray says below, Basil Shall is asked to give evidence. Here is what Mr Murray has to say:

Your first mistake is always your second mistake

Writing about Horizon, the temptation is to start with Alice in Wonderland… but which quote to use?

– “Why, sometimes I’ve believed as many as six impossible things before breakfast!”

– “It was all very well to say ‘Drink me,’ but the wise little Alice was not going to do that in a hurry. ‘No, I’ll look first,’ she said, ‘and see whether it’s marked ‘poison’ or not’.”

– “No, no” said the Queen. “Sentence first, verdict afterwards”

All of these work perfectly, I think. 

I was first involved with the precursor to Horizon – the BA/POCL [Benefits Agency/Post Office Counters Ltd] Programme – working in the Post Office as a link between Account Managers in our Business Centres and their clients on opportunities to automate their products in a future where all Post Office outlets had the appropriate technology. 

I also became involved in the requirements stage for the forthcoming Invitation to Tender (ITT). This was where the trip into Wonderland maybe started. The Programme Team ran the requirements sessions, which were in non-Application areas such as Security, on the basis that as this was going to be a Private Finance Initiative (PFI) procurement then you could ask for whatever you wanted as the supplier would have to provide it! 


Challenging this approach in these sessions was pretty fruitless as the Programme didn’t want to listen or engage. The problem, of course, was the obvious disconnect with the business case with the Benefits Agency which was based on rollout happening to eye-wateringly tight timescales. 

It was clear even at this stage that having, say, security requirements that the Pentagon would blanch at would impinge on this. 

It was around this time I was asked to produce the requirements to go in the ITT for the Post Office EPOS [Electronic Point of Sale – ie front end till position] Application, largely because of my work on previous automation programmes. 

Now, as the evidence to the Inquiry so far has shown, EPOS was a key factor in all of the future problems.

Bob n’ Baz

The issue with trying to write meaningful requirements was twofold: POCL had brought in a new Director of IT Strategy (I think that was his title), Basil Shall, who had convinced Bob People and Paul Rich, the key programme sponsors in the business, and through them the Board, that EPOS could be provided by a standard off-the-shelf solution such as you would find in any supermarket, rather than anything more bespoke. 

As Basil knew about as much about how Post Offices worked in practice as I do about the Hadron Collider it is difficult to overestimate how much this flawed ‘assumption’ led to the future problems. 

The second problem was that I was told that PFI meant the requirements had to be based on Outputs rather than any kind of technical specification, which meant that the chosen supplier would need great technical expertise to design from scratch something which could actually support how Post Offices worked, or POCL would need to radically redesign and standardise their processes in agreement with Clients to enable a more simple EPOS system to be implemented. 

Again, the disconnect with the business case was stark but any attempt to challenge the EPOS assumptions were ignored… to suggest the Emperor lacked a certain something in the sartorial department was seen as career-limiting.

Flights from Reality 

It’s also worth noting that Basil Shall sold the idea that the Benefits Encashment Service element could be provided by an off-the-shelf banking package, another flight from reality, as it failed to consider the business rules and exceptions that would need to be included in any specification. 

I continued to work on how to automate client transactions etc while the procurement continued as we weren’t allowed to speak or interact with the prospective suppliers but when we were finally allowed to meet them I went to a series of meetings with each of the final three shortlisted – Cardlink, IBM and ICL Pathway. 

I also attended meetings with all three specifically on EPOS together with John Meagher, from the Programme, and Basil Shall and his deputy. At this meeting Cardlink and IBM explained that there was no conceivable way that an off-the-shelf solution could be provided and they would need a considerable amount of input from POCL to design a bespoke solution. It is fair to say that by the end of these meetings, Basil Shall had very little to say. 

ICL Pathway, by contrast, were confident they would provide EPOS as per the limited requirements. This confirmed my impression from the sessions with the three that Cardlink and IBM had a chance of delivering a workable system, although not remotely to the programme timescales (as this was always a fantasy), but that ICL Pathway, where I’d had my first meeting with Terry Austin, who struck me as a DelBoy Trotter-type character, had no chance at all. 

Cardlink and IBM produced very professional presentations, and had a clear understanding of the difficulties and complexities involved. This contrasted sharply with ICL Pathway’s presentation, which what was, to my eyes, amateur hour.

When ICL Pathway were then chosen (which didn’t surprise me as I’d always suspected price would be the ultimate determinant and I was convinced the POCL Board hadn’t a clue about the complexity of what was going to be required), I went to join the Programme, working for John Meagher, to look after the POCL Business Products, as they were called, but that’s another story. 

I wasn’t surprised how things turned out on a technical level but I couldn’t have envisaged the level of wickedness that would happen. Having worked closely with Subpostmasters on designing and project managing the Capture package, which was sold as an aid to back-office accounting, I knew what decent people they were. Having worked at head office, I suppose the Three Wise Monkeys approach of Post Office senior management was to be expected.

My work on the Post Office Horizon IT inquiry is crowdfunded. If you’d like to contribute, please click on the widget you should be seeing to the right of this text (or below if you’re reading it on a mobile). To find out more before donating, please go to my tip jar web page. All contributors will be added to the ‘secret’ email newsletter, which offers irregular, and at times, irreverent insight into the machinations of the inquiry and the wider scandal.

Subscribe For Latest Blog Updates

Leave a Reply

Your email address will not be published. Required fields are marked *