27 February 2023 – Stephen Grayston
Hide video Show video
(10.00 am)
Ms Kennedy: Good morning, chair.
Sir Wyn Williams: Good morning.
Stephen Grayston
STEPHEN ROBERT GRAYSTON (sworn).
Questioned by Ms Kennedy
Ms Kennedy: Could you state your full name, please.
Stephen Grayston: Stephen Robert Grayston.
Ms Kennedy: You should have a copy of your witness statement in front of you. For the transcript that’s WITN03920100. Have you got that statement there?
Stephen Grayston: Yes.
Ms Kennedy: If you turn to page 28, please, is that your signature there?
Stephen Grayston: It is, yes.
Ms Kennedy: It’s dated 14 September 2022; is that right?
Stephen Grayston: It is, yes.
Ms Kennedy: Is it true to the best of your knowledge and belief?
Stephen Grayston: Yes, it is.
Ms Kennedy: That witness statement is now in evidence. Everything I’m going to ask you is supplemental. Can I first thank you for coming to give evidence to the Inquiry today.
Your statement covers a variety of issues but today I’m going to focus on Phase 3 issues. But if we could start by talking about your background, you started working for the Post Office as part of the Royal Mail Group in January 1986; is that right?
Stephen Grayston: Correct, yes.
Ms Kennedy: You left the Post Office when the Royal Mail Group split in April 2012 –
Stephen Grayston: Yes.
Ms Kennedy: – is that right? And you have worked a variety of IT jobs since then?
Stephen Grayston: Yes.
Ms Kennedy: In 1995 you were appointed as a manager in the BA POCL programme; is that right?
Stephen Grayston: Correct, yes.
Ms Kennedy: What did that involve initially?
Stephen Grayston: The first role that I had was to support the evaluation of potential partners/suppliers for the scope of the programme and my particular role was to look at support service proposals.
Ms Kennedy: How did that change over time?
Stephen Grayston: So as the programme moved forward into its next phase, I moved into the implementation team and then did various work in the implementation team, probably most significantly was around – in office migration.
Ms Kennedy: I believe you were involved in the acceptance process; is that right?
Stephen Grayston: I was, yes.
Ms Kennedy: The Inquiry’s heard a great deal of evidence about the acceptance process so I’m not going to take you through all of that, but at the time Horizon was rolled out did you believe the system was robust?
Stephen Grayston: Yes. There were bugs, there were defects that had been captured and listed, but fundamentally my understanding was that the system was working and producing figures and outputs that were consistent with transactions and the inputs that the system was receiving.
Ms Kennedy: You mentioned that you were aware of bugs, errors and defects. What were some of the issues that you were aware of prior to rollout or prior to the acceptance?
Stephen Grayston: I think within the material that I’d been sent there was various reports coming out of testing on what the results of running various scripts, et cetera, were showing and, you know, as in any programme that I’ve been involved in, you know, the good part is that these things are being flushed out, that they’re being understood and they then need to be fixed.
So, you know, the follow-on from that is an evaluation as to the seriousness, either individually or as a consolidated group, of those defects that are arising as a result of testing. So there comes a point where you evaluate whether to move forward because what remains to be fixed is deemed to be not significant, or you hold and fix all the things that need to be fixed.
Ms Kennedy: Were you aware at that time about issues with EPOS, the electronic point of sale?
Stephen Grayston: Sorry, what – depends – what do you mean by issues?
Ms Kennedy: Well, problems, that that had been a persistent issue and that Post Office felt that that needed to be actioned by Pathway in order to correct it or to ensure the data integrity of the transactions that were being shown?
Stephen Grayston: From the material that I’ve been sent, I’ve seen reference to advice about rewriting EPOSS completely. That was not something I was aware of. That there were concerns over fundamental issues, no, I can’t – you know – no, because, as a – effectively, on the operational side, working with the regions, the IP areas, it was our job to make sure that we had something that was being implemented that was trustworthy.
Ms Kennedy: And it was trustworthy at the time of rollout in your mind?
Stephen Grayston: Yes, yes.
Ms Kennedy: If we could turn up our first document, please, POL00028441, I’m just going to take you to this.
This is a Christmas Horizon Research Report that was carried out in January 2000. Were you aware of this at the time?
Stephen Grayston: I believe I saw a copy of this, yes.
Ms Kennedy: What did you think of it at the time; do you remember?
Stephen Grayston: To some extent it wasn’t a surprise. If you have 60,000 users and then you have – you know, and that’s at the front end, front office, and then you have users in the back office, that they struggle to understand and use the system would be expected from some users, yes.
Ms Kennedy: If we could turn over on to the next page, please – sorry, the next page again, yes – we can see that this is appendix 2, which contains some verbatim comments from subpostmasters. Just to be clear, you saw the report and also this appendix at the time or …?
Stephen Grayston: I can’t be absolutely certain but I would have expected to see it, yes.
Ms Kennedy: If we could turn to page 15, please, and we can scroll down. The Inquiry’s been through this report before but just for your benefit there’s a section entitled “Not enough training on balancing” and we see there some of the comments:
“Training for accounting was very bad. Balancing took hours to sort out, and was kept up until midnight sometimes. Tried to call helpdesk but it was always almost engaged. But needed for time on balancing. The 1st day was all right, but the quality of the training was not good on the 2nd day.”
Further down:
“They didn’t inform us very much on cash accounts.”
So there’s quite a lot of feedback, I’m not going to take you through it all, but it sets out that people were quite frustrated at the amount of time that was being spent on training on balancing. Would you accept that?
Stephen Grayston: Yes.
Ms Kennedy: Is that something – you mentioned a moment ago that you would expect a certain level of difficulty or people to find things difficult to a certain degree. Is this in line with what you would have expected?
Stephen Grayston: I think my expectation is a generalisation in terms of the change curve. You know, when you ask people who have worked in a certain way for a long time to change the way that they are working then some will struggle and some will adopt the change very easily.
I think, though, you know, in this particular point, what was being referenced is that the training itself wasn’t good enough. So irrespective of the general point of people struggling with adopting to the requirement for change, the training itself should be adequate to allow people to operate the system.
Ms Kennedy: If we turn to page 19, please, and scrolling down we can see one section entitled “I’m not computer literate”.
So a moment ago you were talking about people who were used to working in a certain way. This shows the level that some people were at in terms of their base level of computing, doesn’t it?
Stephen Grayston: Yes, yes.
Ms Kennedy: We can see that some people describe it as:
“It was frightening. We were thrown in at the deep end and it was very unsettling. It was particularly difficult for those who had no previous experience with computers. They did not take account of our needs.”
And:
“I am not a computer person, I was put with people who had used them and with people who worked in head post office. I did not need half the information given it was a waste of time when there was other things I needed –total confusion in the end.”
So the Post Office were expecting people to go from not using computers at all in their day-to-day work to being across quite a complex system; would you accept that?
Stephen Grayston: Yes. I – you know – yes, yes.
Ms Kennedy: It’s fair to say, isn’t it, that some postmasters at the beginning struggled to use the system?
Stephen Grayston: Yes.
Ms Kennedy: Would you accept that if someone finds a system particularly difficult to use, they are more likely to make errors when inputting the data that’s held on that system? Would that be right?
Stephen Grayston: I think that’s fair. You know, it takes longer. People who don’t have the understanding – you know, even if there’s a helpdesk, there’s the call out for support, but there is the risk of error, yes. Yes.
Ms Kennedy: If we could take that document down, please, and turn to NFSP00000513, please.
This is a report of the National Executive Council of the NFSP in March 2001. Now, you wouldn’t have been at that meeting but there’s just one point that I wanted to take you to.
If we could turn to page 15, please, and scrolling down, please.
So this is a Mr Peberdy, who the Inquiry has already heard from. It’s him making a report to the meeting. He says:
“Mr Peberdy reported that these problems are still being highlighted [this is polling problems] and just recently had been circulated and reported on the problems in organising meetings with the business but now monthly meetings had been scheduled and there had been a meeting on the 26th February 2001 from which could be seen from the Action Points, there were 28 items that required action, some of them the business still had to come back to them on.
“Amongst it one was to set up the two day meeting, a separate meeting on Losses and Gains Policy, a separate group to bring in the Horizon problems. There had been stories about the problems that had been created by Horizon, shortages, Horizon was not doing things, the problem with losses having to be made good immediately, and all the things about Suspense Accounts. He reported that he wanted a group to examine this. He had been led to understand that there was £10 million in Suspense Accounts now as opposed to about £2 million 18 months ago. Another feature of the system was that it highlighted everything.”
So looking at that, it appears that after Horizon was introduced the money held in a suspense account went from 2 million to 10 million; is that right?
Stephen Grayston: In terms of what I see on the screen and what Mr Peberdy reported, yes, I mean, I heard this – I can’t remember the context in which I heard the same point but the point about Suspense Account and the amount going from 2 million to 10 million was something that came up in a conversation somewhere. I recognise that, yes.
Ms Kennedy: You recognised that, that – a conversation around this time, March 2001?
Stephen Grayston: I can’t say with any certainty of a specific date but, you know, that would be, I guess, roughly, yes, it would be appropriate.
Ms Kennedy: That conversation, I appreciate you’ve said you don’t really remember but was it something that members of your team or you were particularly concerned about at the time?
Stephen Grayston: Yes, because that is a fivefold increase, and as I understood it there was analysis going on. I mean, I think there was – and I think Mr Peberdy states that, you know, everything was being flushed out and I think that was the context in which I heard, you know, the increase from 2 million to 10 million, that everything was being flushed out.
But that didn’t explain, doesn’t explain, you know, what is going on. So I understood that there was some analysis on this going on.
Ms Kennedy: What do you mean by everything being flushed out?
Stephen Grayston: Well, because the Horizon System had been implemented and therefore data was flowing through the system – it wasn’t manual – the information that was being received in finance through what was being reported as in suspense was much more visible, whereas before, for whatever reason, it wasn’t that number.
Ms Kennedy: What does that increase from 2 million to 10 million suggest to you?
Stephen Grayston: Well, it suggests that there are errors that are being posted to suspense that need to be looked at. You know, what – my first question is: what has caused the posting to suspense of a fivefold increase? You know, it doesn’t seem to make a lot of sense.
Ms Kennedy: What do you think the reason was?
Stephen Grayston: I don’t know. I honestly don’t know. I mean, the analysis was going on and speculation on my part was not going to help because I wasn’t close to the detail, neither was I expected to be, you know, involved in the analysis or close to the detail. But there were other people with the right level of knowledge that you would expect were looking at this and examining it.
Ms Kennedy: What were they saying about this? What was their theory?
Stephen Grayston: I didn’t see – I didn’t see any follow-up. You know, as I say, the context in which I heard this, you know, I’m comfortable in saying that I recognise that comment, but I didn’t see any follow-up that said, you know, “This is the result of our analysis, you don’t need to worry”, et cetera, et cetera. But that Mr Peberdy had seen it, you know, there were others that understood it, and I expect there would have been a report somewhere stating what the cause was and what was happening.
Ms Kennedy: But even though you were aware of this fact, you didn’t check to see what the outcome of that was?
Stephen Grayston: No, I didn’t, no.
Ms Kennedy: Does this not seem like quite an important point?
Stephen Grayston: It seems like an important point, yes. I mean, it – looking at it now, yes, it seems like something that should have been clearly understood and articulated back through the programme, so that the programme could then articulate what was going on back to the relevant business stakeholders.
What I can’t – having said that, what I can’t say to you is that there may well have been a strand that had done the analysis and had reported back to various stakeholders. You know, in my world it disappeared.
Ms Kennedy: Moving forward in time then to the IMPACT programme, you were change management lead on IMPACT; is that right?
Stephen Grayston: Correct, yes.
Ms Kennedy: What did that involve?
Stephen Grayston: So it involves understanding, effectively understanding the nature of the change. So what was the business intention? What was being changed by business unit, whether it’s finance or operations or audit? So what was changing process, what was changing system, and, having done the gap analysis between what people did today and what they would be expected to do tomorrow, to work on training and processes to enable people to move from one state to the future state. In the area that the changes were – had the highest level of impact, for example, in areas in finance, organisational design would come into it because you may have people exiting the organisation and new skills and new people coming into the organisation, and, depending on the scope of business change – and I can’t quite remember but there’s also the point about communication. So, you know, communicating to stakeholders as well.
Ms Kennedy: From what you’ve said it doesn’t sound like you were involved in the primary decision making in terms of what it would actually involve. Your role was to do with carrying out the changes that other people had decided upon; is that fair?
Stephen Grayston: Yes, I mean, the documents that I was sent were, you know, clearly laid out in terms of business strategy and what – the reasoning behind the IMPACT programme, yes.
Ms Kennedy: You set out what you understood the purposes of the IMPACT programme to be in your witness statement.
If we could pull that up at WITN03920100, and if we turn to page 20, please.
Looking at paragraph 52, scrolling down, it says:
“I believe that the IMPACT Programme was driven by the need to simplify and update many backend legacy systems to improve efficiency, accuracy, and lower operational costs. At the front end, in offices, the Programme also introduced the capability for Smart Card transactions and changed the Suspense Account process from manual to an automated process. The Releases also introduced various other changes to the Horizon System that were related to either products or service improvements.”
So is that how you understood the purpose of the project, this programme?
Stephen Grayston: That’s my reflection now. I might have been able to give you a more detailed statement a few years ago but, yes.
Ms Kennedy: Did you hear the evidence of Mr Philip Boardman?
Stephen Grayston: I think I did. It was only a few days ago, wasn’t it?
Ms Kennedy: Yes. He told the Inquiry that part of the simplification process that IMPACT envisaged was so that debt would be more visible. Do you agree with that?
Stephen Grayston: Debt would be more visible … In the sense of the suspense account? In what context was …?
Ms Kennedy: Was simplifying things so that it’s more obvious what debt is owing by either the subpostmasters or by clients.
Stephen Grayston: Yes. I mean, the – you know, in my understanding, you know, part of the reasoning was to ensure that data was generated accurately at the counter, that it was harvested into the finance systems accurately, and then passed to clients accurately and in a timely manner.
I think in the legacy world – and, you know, I’m not an expert on the legacy systems by any stretch – but there were timing discrepancies that would arise. So, you know, one of the things about simplification and the use of, you know, the new systems was to increase speed, accuracy – yes.
Ms Kennedy: Yes. Summarising that, I suppose, is do you agree that part of the reason for the programme was that the Post Office felt that cash was going missing?
Stephen Grayston: Yes, yes.
Ms Kennedy: Was that a big driver?
Stephen Grayston: It wasn’t – it was definitely a driver. I mean, I think there was reference to, you know, remittances, for example, into branches, you know, that were – where leakage or loss was being experienced. So, you know, if, as a process, you can automate remittances and tighten up that process, then you’re reducing the risk of loss or leakage. So, yes, it was definitely a driver of the programme.
Ms Kennedy: If we could turn up POL00038870, this is the accounting and cash management programme conceptual design, and if we scroll down, we can see your name is not on the list as programme manager or a design authority, but you would have seen this document at the time?
Stephen Grayston: At the time of the conceptual design – sorry, can you just give me the date?
Ms Kennedy: Yes. So if we scroll over, I think this is the date – if we scroll over on to page 5, we can see that the document history is September 2003, if we scroll down to the bottom.
Stephen Grayston: Yes, I mean, I think at that point I would have expected to see it, yes.
Ms Kennedy: If we could turn to page 14 of that document, please, and scrolling down, please, we can see at 3.2.2 the “Key Priorities” in this context, and it says:
“2 fundamental changes have made Post Office Limited’s funding position a critical business survival issue:
“- The business is trading at a loss
“- The migration of benefits to ACT will be accompanied by the loss of pre-funding by government departments of the necessary cash in the network.
“The business now has to borrow funds to fund its trading losses and to fund working capital needed in branches. Such borrowing is limited in availability and its cost add to the trading loss. From April 2003 DTI [Department of Trade] will provide a loan and I will require a robust statement of cash holding as security.”
So at that time the Post Office was trading at a loss and in a pretty dire financial situation; is that right?
Stephen Grayston: Yes, as far as I was aware, yes. Yes.
Ms Kennedy: Could you help us with, at the time was that something that was troubling people or worrying people? Was that something that people felt had to be actioned quickly?
Stephen Grayston: Well, it had been troubling. If I could just focus on the second point, you know, in 1995 the point about ACT was already recognised and the threat that benefits payment by ACT represented. So for a number of years, through Horizon, IMPACT and then the Post Office Card Account programme, this threat to Post Office and the financial position had first of all been recognised but then had materialised.
Ms Kennedy: So there was a need to bring in cash; is that right?
Stephen Grayston: There was a need because the payment of pensions and allowances was the significant product or service that was offered by post offices, without being able to replace – if that business was lost, without being able to replace it, then the Post Office’s position would become worse financially, yes.
Ms Kennedy: So it needed cash?
Stephen Grayston: Yes.
Ms Kennedy: Part of IMPACT was automating the part of the accounting process that had previously been conducted in Chesterfield, the error reconciliation; is that right? There were a number – 300 people in Chesterfield who were carrying out checking processes?
Stephen Grayston: Yes, it was a – like a big paper factory, yes. Yes.
Ms Kennedy: So part of what was envisaged was the reduction of those costs and move to automation; is that right?
Stephen Grayston: Correct and that’s – when I talked about organisational redesign, that would have been, yes, one of the areas.
Ms Kennedy: Most of the people who were based at Chesterfield doing that job of checking, they would have been removed essentially after IMPACT or cut down severely?
Stephen Grayston: Yes.
Ms Kennedy: Would you accept that part of the IMPACT programme envisaged the shift of responsibility from that team to identify errors, to the subpostmaster to identify the errors in the branch?
Stephen Grayston: I think there’s perhaps two parts to it. I think part one would have been that the introduction of the systems should have exposed errors quickly, which would have resulted in automated error notices being generated back to offices more quickly. But the onus would be on the people in the Post Office, the office manager or subpostmaster, yes, to understand how an error had occurred if their account was not balancing.
Ms Kennedy: Or identify the error before it’s put in, because they are the people who are putting in – manually – the processes and handling it on a day-to-day basis, isn’t that right?
Stephen Grayston: They are certainly handling transactions on a day-to-day basis, yes.
Ms Kennedy: So they would be the ones, in the first instance, who are responsible for identifying those errors; correct?
Stephen Grayston: Yes.
Ms Kennedy: At this stage was the reliability of Horizon taken as a given?
Stephen Grayston: The fundamental Horizon System and its ability to accurately represent figures through transactions and represent those into the back end systems, yes. There were still individual issues or defects that, you know, needed to be fixed, you know, as a matter of – in my experience as a matter of course that you will inevitably always find some defect even if you’ve gone through extensive testing. But something will always turn up. It’s the seriousness of what turns up that needs to be assessed and that’s why we had NBSC and the HSHD.
Ms Kennedy: So the way of double checking it – so you, in the first instance, have the subpostmaster, but then the secondary role is those helplines, the NBSC and the HSH; is that what you’re saying?
Stephen Grayston: Yes. I mean, you know, they are absolutely fundamental – absolutely fundamental – from early in the programme, that people are contacting, you know, the Helpdesk, creating the view of the data that then has professional people analysing what the story is behind what is being reported.
Ms Kennedy: But that’s always going to be limited, isn’t it, because the people on the helpline aren’t in the branch with the person on the ground, isn’t it? They are going to have to go off what they can see on the system, potentially, and what the subpostmaster tells them; isn’t that right?
Stephen Grayston: Correct. But if there is a recurring theme in calls coming in – you know, users – people express themselves in different ways, and if there is an art in it, it’s to understand and articulate into the Helpdesk system what the problem is and what the proposed rectification is. But what you would expect is that with recurring issues there is action taken, even if it’s not a system issue. So it may be that, you know, training itself or a note needs to go out to branches to say, you know, “We have received concerns from subpostmasters over this type of transaction. Please be aware, you know, to take this particular action.”
So it doesn’t necessarily always have to be system-driven but the analysis is critical, yes.
Ms Kennedy: But at this time did you think back to what we’ve been discussing, about the value of money in the suspense account, and think, “Oh, I wonder if someone bottomed out”, and why that money had gone from 2 million to 10 million?
Stephen Grayston: No, I didn’t.
Ms Kennedy: Do you think that that would have been something, with the benefit of hindsight, you should have done?
Stephen Grayston: With the benefit of hindsight, yes.
Ms Kennedy: If we could turn to the next document, POL00038878, please. So this is the another document to do with conceptual design. This is “Branch Trading Reporting, Management and Control and Transaction Management. Conceptual Design”. Again, would this have been the kind of thing that you would have seen at the time?
Stephen Grayston: Yes. I mean, there would have been a lot of documents being circulated talking about design, and I can see from the contribution that different areas of the business were obviously contributing to that view.
Ms Kennedy: Can you explain what you mean by that.
Stephen Grayston: Well, I think, from – from my recollection, I mean, in there it looks like there’s audit, there’s obviously finance, investigations team.
Ms Kennedy: Could you just – I think you are going through the names of contributors. Could you just tell us the name of the person and the field that they are speaking to.
Stephen Grayston: Tony, Tony Utting, I think would have been representing investigation or auditing in that area. Ann Clarke was an expert in the processes within Chesterfield. Karen Hillsden I think had been involved in the conceptual design, and Gareth Jenkins obviously was there from ICL Pathway.
Ms Kennedy: Did you know Gareth Jenkins?
Stephen Grayston: No, I didn’t, but I’ve seen his name a few times on various documents, yes.
Ms Kennedy: Did you know him by reputation at the time?
Stephen Grayston: No. You know, I know he was an architect or the senior architect. You know, the – my interface with the architect team primarily would have been Torstein, and I think it was Torstein that probably had the most conversation with Gareth.
Ms Kennedy: That’s Torstein Godeseth?
Stephen Grayston: Yes, yes.
Ms Kennedy: If we could turn to page 13 of this document, please, and we scroll down, we can see again recorded as some of the key – or the “Key Priorities” of the IMPACT programme, which state:
“Make the identification of debt easier
“Reduce the amount of reconciliation required
“Increase the amount of debt recovered
“Put the emphasis on clients and customers to validate the data
“Simplify branch processes by reducing the amount of paper
“Centralise/consolidate agents debt
“Enable matching of cash at branches with settlement with client.”
Those are consistent with some of the things we have been talking about, aren’t they?
Stephen Grayston: Yes.
Ms Kennedy: If we look further down at “Business Drivers/Issues”, it states:
“Re-focus on Debt Recovery (financial recovery of money), target 95%.”
Do you know what that would have been referring to?
Stephen Grayston: Well, I – my assessment of that is that where losses had occurred, then it was the recovery of the monies associated with those losses or discrepancies, and primarily I guess that would have been focused on the branch.
Ms Kennedy: The subpostmasters or branch staff?
Stephen Grayston: Yeah, yes.
Ms Kennedy: When it says in the second bullet point, “Only 10% of discrepancies are actually debt”, what would that have meant?
Stephen Grayston: My interpretation of that is that – I think I mentioned timing discrepancies previously. I think that, you know, one of the challenges with the legacy systems was to remove what looked like debt, it wasn’t actually debt it was just the timing of cut-offs in systems when data was provided to other systems and that was subsequently resolved.
Ms Kennedy: So it’s a timing issue rather than – can you just explain that again.
Stephen Grayston: I can explain my understanding. So my understanding is that if at the point that data is sent to, for example, a client that data from the front office, under the legacy world, may not have reached the central system, so there may be that money had been taken in but wasn’t – the data wasn’t represented back to the client in a timely manner, and that might represent debt in certain circumstances.
Ms Kennedy: In the majority of circumstances or …?
Stephen Grayston: Well, I mean, it says only 10 per cent of the discrepancies are actually debt. So, you know, as I say, my interpretation of that point, as I’m sitting here today, is I can relate it to timing. 90 per cent seems a high number but I didn’t work in the back end in Chesterfield in finance, so, you know, that could well be accurate, yes.
Ms Kennedy: If we look at the bottom of Business Drivers/Issues”, it says:
“Accounting and settlement on our data, not clients.”
Stephen Grayston: Yes.
Ms Kennedy: So when it says “our data” that basically means the Horizon data; is that right?
Stephen Grayston: Yes and, you know, it’s a challenge that I’ve come up against, you know, in other programmes where settlement on client data versus the data that you have in-house leads to lots of questions, yes.
Ms Kennedy: So from this the Horizon data is becoming all the more important, isn’t it?
Stephen Grayston: Absolutely.
Ms Kennedy: It’s the start and end of the matter –
Stephen Grayston: Yes.
Ms Kennedy: – as regards settlement with clients; is that right?
Stephen Grayston: Absolutely. If you want your clients to settle on your data, then your data has to be good.
Ms Kennedy: So all of this is predicated on the idea that, to use your words, the Horizon data is good?
Stephen Grayston: Yes.
Ms Kennedy: Because without that, none of this works?
Stephen Grayston: Yes. Yes, it raises too many questions.
Ms Kennedy: If we could turn to page 15 of that document, please.
Sir Wyn Williams: Before we do that, could I just understand the word “client”.
Mr Grayston, do you understand client to include subpostmasters or are we talking about third parties whose products are being sold in post offices?
Stephen Grayston: Third parties, sir.
Sir Wyn Williams: Right. So where we see in this list “Accounting and settlement on our data, not clients”, you would agree that does not refer to subpostmasters?
Stephen Grayston: Correct.
Sir Wyn Williams: Okay.
Ms Kennedy: If we could turn over the page, please, to 15, and scroll down, and down again, looking at paragraph 12 – just down a bit further, thank you – it states:
“By the end of a monthly trading period, branches should be required to make good discrepancies between Horizon generated cash and stock positions and the actual physical position determined by branch office staff. To help facilitate this, existing Horizon facilities that permit branch staff to post cash discrepancies to a cash suspense account will be removed. Remaining branch suspense accounts should only be used following prior to authorisation via Post Office central processes and will be restricted to use by branch staff with Horizon manager/supervisor roles.”
Is that in accordance with your understanding of what was to happen?
Stephen Grayston: Yes.
Ms Kennedy: It goes on to – I mean, essentially what it is saying here is the suspense account is going to be removed, effectively, which is where subpostmasters previously posted discrepancies; isn’t it right?
Stephen Grayston: Yes, the ability – yes, I mean, the ability to post to suspense lay with a subpostmaster, or the Crown Office branch manager should they choose. Under the changes, that facility was no longer going to be there. It was being closed down.
Ms Kennedy: At the time this programme was being developed, was there a perception that subpostmasters were using the suspense account to hide money that they couldn’t account for or had stolen?
Stephen Grayston: Yes.
Ms Kennedy: How prevalent – unpacking that a bit, was it generally the perception that subpostmasters were using it to hide amounts they’d stolen?
Stephen Grayston: In my experience – because prior to joining the programme I had been with Royal Mail Group investigations – there were instances where subpostmasters wished to use an amount of money for other purposes, not – not with the intention of theft or permanently deprive, but wanted to or needed to use it for other purposes. So it was a facility or an opportunity, should someone so wish, to undertake something short-term using Post Office cash.
There were instances, I believe, where it involved theft, and, you know, I’m sure there’s a lot of analysis within Post Office on the types of cases, the numbers of cases, the amounts involved that, you know, were regularly discussed at a post office management level.
Ms Kennedy: You mentioned using the money for short-term purposes.
Stephen Grayston: Yes.
Ms Kennedy: That’s not something that they were allowed to do, though, was it, to use that money in the suspense account for short-term purposes?
Stephen Grayston: No, no.
Ms Kennedy: So that’s something equally Post Office would want to clamp down on and didn’t want to continue?
Stephen Grayston: Yes, but it’s – I was distinguishing between somebody who perhaps was – you know, or was stealing and somebody who was – was – been in need of an amount of money but it was not with the intention of keeping that money.
Ms Kennedy: But in both cases Post Office didn’t want them to be – well, they certainly didn’t want them to be stealing but they also didn’t want them to use that money for those purposes either –
Stephen Grayston: Well, it was Post Office money not the private business side money, yes.
Ms Kennedy: Over the page, the document goes on to explain that suspense account can be cleared in several different ways, and that includes cash or transaction, the subpostmaster paying for – out of their salary or credit card.
I mean, in the IMPACT programme there was no provision here to challenge the sum owing on Horizon itself, was there?
Stephen Grayston: I think when the Horizon produced a position then the – you know, my understanding was that there was an opportunity to challenge but it wasn’t, you know, through the system necessarily, it would have been through your retail line manager, maybe a call to the Horizon System Helpdesk saying that, you know, “This has happened, I don’t know why”. But that was the process about making good was – was what was agreed, yes.
Ms Kennedy: So there was nothing on the system itself. What you’ve just described involves phoning the helpline but not on the system itself, you wouldn’t dispute?
Stephen Grayston: Yes.
Ms Kennedy: If we could turn to page 18 of that document, please, and scrolling down and looking at “Legal & Regulatory”, it says:
“It will be verified that branch processes and reporting changes meet legal and regulatory financial reporting constraints (eg auditors) to ensure that there is sufficient information from the new system to support regulatory reporting, litigation and criminal prosecution.”
Was the ability to prosecute subpostmasters under the criminal justice system a key driver or a key factor in the IMPACT programme?
Stephen Grayston: I think with any system, if you looked back to Ecco or Ecco+, which was in Crown offices, that the same statement would, I expect, apply, i.e. that the system produces data which is trustworthy to the extent that it can be used to support, if necessary, a criminal prosecution, yes.
Ms Kennedy: In your mind at this time, how important was it within the Post Office to have the ability to prosecute subpostmasters?
Stephen Grayston: I actually think at this point in time, if anything, it was diminishing. You know, the – the prosecution of an individual, you know, that Post Office went through was not, you know, a cheap – it was expensive and – but on the other side, you know, it was the deterrent effect as well.
Ms Kennedy: So the deterrent effect was still important even if you felt that prosecutions themselves were becoming less important; is that what you’re saying?
Stephen Grayston: Well, resolving in some appropriate way was absolutely important. If a situation was so significant and serious that prosecution was merited, then, you know, prosecution was appropriate. But, yeah – so it is important, though, that if that is the step that you take, that the data on which you are basing your decision is robust, is accurate.
Ms Kennedy: If we could turn to page 70 of this document, please. In fact if we could go back over the page to page 69 and scrolling down, just so you can see the context of what I am asking you about. This is in the context of “Discrepancy Management”, and it mentions:
“1. Receive Automated Message
“2. Handle Transaction Corrections.”
We can see there the “Receive Automated Message” section.
If we go over the page, I wanted to ask you about handling transaction corrections. So you can see there the description says:
“This is the mechanism for Processing the Transaction Correction by the branch.”
It says:
“Trigger: User Initiated
“Automation: There will be a button for Transaction Correction Management within the menu hierarchy which is only accessible by users with the appropriate role. This will provide the user with a list of the unprocessed Transaction Corrections, displayed in date/time order. Having selected the Transaction Correction to process, the system will display text making clear what will happen when they select any of the options presented. For each Transaction Correction the user will have up to three options - Each option, when selected, will perform an identified set of transactions, defined within the Transaction Correction (which may include an option to Do Nothing - requesting further investigation).
“Should the Transaction Corrections fail validation, then an error is displayed to the user with a request to contact NBSC. The Transaction Correction will be marked as complete, but no change will have been made to the local system.”
What type of situation does this envisage or how would this work?
Stephen Grayston: To be honest, I’m not quite sure. I’d have to take that away and have a long hard look at that.
Ms Kennedy: That’s fine, thank you.
Turning then to our next document, if we could turn up POL00038909, please. We can see here “IMPACT Programme S80 Migration Strategy”. Could you explain what is the document is and how it came about.
Stephen Grayston: The – well, a migration strategy would define how you move from what you have or where you are to where you want to be and, in that sense, you know, I’d need to see the rest of the document as to what the scope was.
Ms Kennedy: We know it refers to the S80.
Stephen Grayston: Yes.
Ms Kennedy: If we could just look at your witness statement – and that’s WITN03920100, please – and if we can look at page 23 and looking at paragraph 57 it says:
“Within the scope of the S80 Release changes were introduced that moved office accounting away from weekly Cash Account production to Trading Periods and also introduced an automated process to manage Unclaimed Payments and Uncharged Receipts that existed as the office level Suspense Account. Up until the S80 Release errors made by office in transacting business had been dealt with through a paper process that required office managers to post details (enter details) of the Error Notices into the Suspense Account; S80 introduced an automated posting process.”
So can you explain and clarify further what the S80 did?
Stephen Grayston: In the sense of this particular point, my understanding, if I’m remembering it correctly, was that error correction was a manual process. We talked before about the factory and all the people working on pieces of paper. Well, those people working on those pieces of paper would turn up errors and that would generate a paper error notice, which then would need to be posted back to the branch that made the error.
Now, if things were working well, the branch – because this may be sometime later – the branch would already have recognised in the accounting period that an error had been made, so when the error notice came in it was a contra-entry in suspense to the error that had already been recognised, if everything was –
Ms Kennedy: If everything was working properly?
Stephen Grayston: – going great. The S80, or IMPACT, introduced an automated process. So on the basis that data was being generated into the systems and, at the back end, the ability for those systems to process the data, that any discrepancies could then be posted automatically – recognised automatically and posted automatically, is my very simple, simple way of understanding it.
Ms Kennedy: So the S80 was an important release for making that fundamental – I mean, it’s quite a fundamental change, isn’t it?
Stephen Grayston: It is a fundamental change, yes.
Ms Kennedy: If we could turn back to that document that we were on before, which is POL00038909, please, if we turn to page 6, we can see the date of this document which is – the updated draft for discussion is 21 June 2004, and this is for discussion in the Design Authority. What is that?
Stephen Grayston: The Design Authority were effectively the people that had analysed and thought through conceptually what IMPACT was about, and then it had been broken down into constituent parts and the Design Authority, or my interpretation of a Design Authority’s job, is to protect the design. As you may appreciate, the world is not standing still as this programme is taking place, so there are always new changes, maybe product changes, new products, challenges to the design coming in, and it is the job of the Design Authority to – that effectively owns the requirements to make sure that the design remains consistent and gives a view on CR, change requests.
Ms Kennedy: If we could look at page 30 of that document, and scrolling down slightly, the “Roles and Responsibilities” section, it says:
“The responsibility for leading the detailed migration analysis lies with the Impact Business Change team - primarily Steve Grayston (Business Change Manager), Ann Clark (Back End), Ben Gildersleve (Counter), and Mark Kirton (Implementation).”
So that was your business change team; is that right?
Stephen Grayston: I think it was wider than that but, given that the highest level of IMPACT was back end, so Ann Clarke, and at counter, Ben, yes.
Ms Kennedy: You would work with these people to carry through the changes that had been designed; is that right?
Stephen Grayston: Yes. I mean, the – fundamentally, you know, the conceptual design needs to be understood. For example, you know, counter, if you took counter, the front office, you need to understand what is changing. So what is expected, what needs to be done, in terms of process, eventually, so that you can define the right level of procedural documentation and the right level of training, and that behind that there is the right level of understanding in the support desk to support the people when this change is going through.
You know, there’s also, as part of that, an evaluation of what is needed at the point of migration from what happens today to what needs to happen tomorrow.
Ms Kennedy: If we could turn to page 20 of this document, please. Scrolling down it says:
“Preparation to Implement POL_FS.”
And it says:
“The following activities are required …”
And lists a number of activities in terms of hardware and software implementation.
Scrolling down, it says:
“In POL-FS activities must be undertaken to load the start of the financial year opening balances from CBDB …”
What does that mean?
Stephen Grayston: Counters business database.
Ms Kennedy: “… into POL-FS.”
And POL-FS is?
Stephen Grayston: That’s SAP, I believe.
Ms Kennedy: “This is in addition to any identified previous year closing balances and movements that need to be put into POL-FS to create the correct starting position.
“There is also an activity to address the position of the suspense accounts both centrally and locally particularly as the current ‘unknown items’ option will no longer be available to the branch. An exercise to cleanse suspense accounts in advance of implementing POL-FS is envisaged.”
So this is the process of cleansing the suspense accounts to move forward with the plan; is that right?
Stephen Grayston: Yes. I think “cleanse” – my understanding in terms of the use of this term is it was envisaged that operations team, so the line management operationally, and the subpostmasters would be encouraged to deal with items in suspense. Because items were sitting in suspense, I believe, sometimes for an extended period of time.
Ms Kennedy: That document can come down, thank you.
Do you think the suspense account was removed because the Post Office desperately needed the money in the suspense accounts?
Stephen Grayston: No, I don’t believe that was a primary driver for closing the suspense account. To me, it was an appropriate action to take if you were running true end-to-end processing. You didn’t need or you shouldn’t need the ability to manually post into an office’s accounting position. So I don’t believe it was a primary driver.
Ms Kennedy: Not a primary driver but do you think it was a factor?
Stephen Grayston: Well, I think, you know, if it was envisaged – and I can’t say I’d have saw it anywhere, that it was envisaged that, as a result of IMPACT being implemented, that there would be a, you know, significant inflow of funds, I – you know, possibly in somebody’s mind somewhere that might have been a factor. But I can’t say I saw that.
Ms Kennedy: I want to ask you some questions about feedback from subpostmasters. You talk in your witness statement about feedback being obtained.
Stephen Grayston: Yes.
Ms Kennedy: If we could turn up WITN0392100, please, and if we could turn to page 23. Looking that bottom of that page, at paragraph 60, it says:
“Whilst I am unable to reference specific notes, or documents, I can confirm that user feedback was important to the IMPACT Programme team and that feedback would have been taken on board and acted upon where appropriate. The feedback would have included comment on User Interface such as screen workflow, colours, positioning on screen, understanding of language used in instructions. There would also have been feedback gleaned from users interacting with the testing team with the aim of reducing the risk of errors. Whilst I cannot provide any specific example I am sure that not all user feedback was accepted; for example if a user disagreed with a fundamental aspect of the concept, the business design, I believe that the overall business benefit to POL would have been the over-riding necessity.”
Could you explain a bit more about what you mean by that.
Stephen Grayston: Yes. The high level design and the conceptual design of what Post Office was attempting to achieve, was setting out to achieve, was signed off and agreed, and agreed between Post Office management and I believe with relevant stakeholder groups.
Inevitably you get people who will actually disagree and challenge the fundamental conceptual design. And, you know, that’s – here is one example. It happened to me in other programmes.
But, you know, what I take from that is that it’s about explaining the benefits of the programme overall because, in isolation, somebody may be sitting there being asked to do something different and not understanding or realising the benefit to the organisation as an overall factor.
So, you know, that’s where people would express their views, but that feedback would not necessarily be taken on board. However, what should be taken on board is that if there is a fundamental lack of understanding why this is being done, what the benefit is overall to the organisation, then, you know, business change should reinforce the reasons behind why the change is happening.
Ms Kennedy: So if a subpostmaster said, “I don’t agree with the fact that suspense account is going to be removed”, that’s not something that would have been taken on board, because it’s fundamental to the programme itself and the design of it; is that right?
Stephen Grayston: It is. But I would expect, out of, you know, courtesy and the appropriate professionalism that, you know, a rounded response would be given to the person who’d raised the point.
Ms Kennedy: But it couldn’t be changed? The IMPACT programme was what it was fundamentally, and feedback could be sought on more peripheral or user-based things such as the interface; is that right?
Stephen Grayston: Yes.
Ms Kennedy: If we could turn up POL00038986, please. This is the IMPACT programme implementation plan for the S80 release. Can you help us with what this document is.
Stephen Grayston: Well, I would expect it to include all the details of how S80 would have been implemented, as it says, at a high level. I’m not sure what the detail is after that.
Ms Kennedy: The difference between an implementation plan and a migration plan?
Stephen Grayston: Yes. Well, migration is part of the overall implementation.
Ms Kennedy: Okay. But they are two distinct things; you would expect to have separate plans for them, would you?
Stephen Grayston: Yes, I would expect the overall implementation plan to highlight the migration perspective and then, as you drill down into detail, that you get a migration plan and processes, et cetera, as you go into further levels of detail.
Ms Kennedy: We can see here that you are a reviewer of this document.
Stephen Grayston: Yes.
Ms Kennedy: So as a reviewer, does that mean that you would have had input into it or you would have had a look at it at the time before it was finalised? How would that have worked?
Stephen Grayston: Yes. As a reviewer, yes, I was expected and required to provide feedback from a business change perspective and, you know, I think it’s always important that the people who are reviewing documents like this understand the scope of their review because S80 was complex.
So we can all make comments about some of the technical aspects but if the technical aspect is not your domain, those comments wouldn’t necessarily, you know, carry any weight.
Ms Kennedy: If we could turn to page 6, please, we can see there in the introduction it says:
“The purpose of this document is to provide visibility and understanding to the IMPACT programme and relevant BAU domains …”
“BAU”?
Stephen Grayston: Business as usual.
Ms Kennedy: “… of a high level business implementation plan for BT and POL-FS and the main activities for the initial pre-implementation stage. This document is largely derived from the migration strategy and meetings held with the business area representatives. It outlines the high level implementation approach that will govern and guide a lower level BT and POL-FS implementation plan.”
If we move to page 7, scope, it says:
“The high level plan scope includes …”
So when it talks about the “high level plan”, these are the things that are going to happen as a kind of headline point; is that right?
Stephen Grayston: Yes, yes.
Ms Kennedy: And we can see it sets out a number of things that are going to happen, and if we look at paragraph 9, it says:
“Distribution of materials to branches and the NBSC, including training and operational instructions.”
Number 10:
“Development of branch error scenarios and scripts for the NBSC.”
Number 12:
“Training of NBSC in types of calls and changes to BT.”
After IMPACT and after the S80 release, the NBSC was going to be extreme important, wasn’t it?
Stephen Grayston: Yes, NBSC was extremely important.
Ms Kennedy: Before but even more so after these changes?
Stephen Grayston: At any release and any change, there is a curve of increased volume calls, et cetera. So, yes, the support services, the support desks, should expect to receive an increased volume of calls, yes.
Ms Kennedy: But over and above, surely, what you would normally expect with a release because, as we were previously discussing, this is now the way that you can dispute what Horizon is showing you, right?
Stephen Grayston: Yes.
Ms Kennedy: So on the long-term you would expect not just a peak after the release but a peak going onwards, wouldn’t you?
Stephen Grayston: Correct. I mean, that’s part of the volumetric analysis that is undertaken for support services. You know, what is the baseline position, how is that baseline likely to change, and what is the curve or what is the BAU wave of increased calls likely to look like?
Ms Kennedy: Do you remember that being something that was considered or thought about carefully at the time?
Stephen Grayston: I believe it was, yes.
Ms Kennedy: Do you think that, all things considered, the IMPACT programme was a success in that it met its objectives?
Stephen Grayston: I don’t remember seeing a closure report. I might be wrong about this, but I can’t recall seeing a closure report or closure analysis. But in terms of the points that you have gone through and the implementation that took place, I believe it was – you know, it achieved what it set out to achieve at the headline level.
Ms Kennedy: Did you investigate with the NBSC what the impact of the IMPACT programme was or how those calls increased over time?
Stephen Grayston: I think, you know, the approach – which, you know, as far as I’m concerned is a standard approach – is that the implementation takes place and there is a handover at each branch or from the programme to business as usual, and NBSC in the early stages is supported by the programme. So, yes, we would have been looking at or should have been looking at the calls being raised with NBSC and the Horizon System Helpdesk. There should be analysis going on to see if there is an improvement required in training or communication or what are we seeing, yes.
Ms Kennedy: At this time, were you aware that Fujitsu were able to access the data generated by the counter remotely and input into it?
Stephen Grayston: No, and I – you know, this is something that, you know, I’ve seen referenced, but at the time, no. To me, it just seems troubling. Perhaps there was a full – there is a full audit log but giving somebody access to the back end to inject data, you know, I would be very uncomfortable with that.
Ms Kennedy: If you had known that at the time, how would that have impacted on your view of how appropriate it was to place such stock on Horizon data?
Stephen Grayston: Well, it would be extremely concerning. You know, you cannot – I mean, I don’t – if there is a – I’ve not seen the reasoning behind it, so if there is justification behind it and there is visibility and it is auditable and it is clearly articulated as a record somewhere of what was done, who did it and why, then there may be a legitimate business reason. But sitting here, knowing what I know, it doesn’t sound appropriate.
Ms Kennedy: Thank you, Mr Grayston. Those are all my questions.
I’m just turning to see if any of the Core Participants have questions. I can see Mr Stein does.
Mr Stein: Sir, there’s a matter that has been brought to my attention in an email that I would like to take some instructions on. It is now 11.15. I wonder whether I could use this time and ask for 20 minutes to have a break.
Sir Wyn Williams: Yes, certainly.
Mr Stein: It may be I will have no questions but I just want to make sure.
Sir Wyn Williams: That’s fine.
Is anyone else intending to ask any questions, just so that I know?
Ms Kennedy: Yes, Ms Patrick and Ms Page.
Sir Wyn Williams: So what is it now? 11.15 or thereabouts?
Ms Kennedy: Yes. 11.30?
Sir Wyn Williams: 11.30, Mr Stein, unless you send a message that you need a little longer, all right?
Mr Stein: Thank you, sir.
Ms Kennedy: Thank you, sir.
(11.16 am)
(A short break)
(11.30 am)
Ms Kennedy: Thank you, chair. I believe Mr Stein has some questions.
Questioned by Mr Stein
Mr Stein: Sir, very grateful for the time. It allowed me just a couple of minutes to gather my thoughts and take instructions.
Mr Grayston, I represent a large number of subpostmasters and mistresses. I’ve only got couple of questions that relate to your evidence you have given today.
You’ve spoken to Ms Kennedy about the branch suspense accounts and about the IMPACT programme that then, as a result of that programme, removed the suspense accounts, okay. You have also discussed with Ms Kennedy the fact that at one stage within the branch suspense accounts that it reached a surprising amount of money, it went up to about 10 million. Okay?
Now, help us, first of all, with what you believe that £10 million in those branch suspense accounts meant. What did it represent?
Stephen Grayston: I wasn’t sure. The context of knowing about this 2 million to 10 million is unclear to me. It wasn’t – it didn’t come to me formally but somewhere it came up. Now, for it to go from 2 million to 10 million in suspense means that there’s errors that were being posted to suspense. Now, I don’t know what those errors were but that – the purpose – my understanding was analysis was ongoing.
Mr Stein: When you say “errors” – if we can just tease this out gently, when you say “errors” do you mean errors within the Horizon System, errors being made by, in your mind, subpostmasters and mistresses, other reasons to account for and lead to errors and shortfalls?
Stephen Grayston: Any or all of those, yes.
Mr Stein: Just pursuing this as far as we can, you’ve answered Ms Kennedy’s questions about this, but what was done that you can recall now to look into the difference – those different possibilities?
Stephen Grayston: I do not know. That’s the position.
Mr Stein: All right. Can we then look at the flip side, which is this: we reached the stage whereby the IMPACT programme suggests that the ability to put the error or the shortfall into the branch suspense accounts was eliminated. Now, what happened to that money? Now, it’s not real money or is it?
Stephen Grayston: Sorry, which money?
Mr Stein: The 10 million in the suspense accounts. Now, is that real money in your mind or is it notional money?
Stephen Grayston: Well, if I refer back to a document that Ms Kennedy showed me, when she talked about 10 per cent being real debt, it could be that some of that 10 million was related to discrepancies or potential debt arising from timing discrepancies in – as data flowed through the system.
Mr Stein: Right. That’s 10 per cent?
Stephen Grayston: Well, I don’t know. That’s one possible constituent of 10 million.
Mr Stein: That’s leaving 9 million. The other 9 million –
Stephen Grayston: No, I think it’s the other way round. I think it’s – if 10 per cent is debt and 90 per cent is timing, then –
Mr Stein: I see. So when this branch account – when this ability for the branches to put money into the suspense accounts was eliminated, what happened within the accounting system of POL to that figure? It can’t just be eliminated, can it?
Stephen Grayston: Well, it can’t just be eliminated, but the purpose of, you know, where finance is, if that is an amount of money that is deemed owed, or debt, then the analysis must show what has caused – what is it that’s causing it. It won’t be – I’m positive that it won’t be one single factor, there will probably be a number of factors involved in it and finance would then seek to deal with each of those factors, is the way that I would expect it to be approached.
Mr Stein: The way that you’re speaking about this is with considerable amount of caveat. You’re saying that, first of all, you accepted a point made by Ms Kennedy as to the possible makeup of the money, the 10 million.
Stephen Grayston: Yes.
Mr Stein: Secondly, you believe that the finance will have dealt with it. Do you have any actual direct knowledge of what happened?
Stephen Grayston: No, I don’t, I don’t. I’m sorry.
Mr Stein: So is one possibility that the subpostmasters and mistresses were pursued for that amount of money as debt?
Stephen Grayston: Yes. It’s a possibility, yes.
Mr Stein: Excuse me for one second.
Thank you, sir.
Questioned by Ms Page
Ms Page: Just a few short questions from me. It’s Flora Page, also representing some of the subpostmasters.
You’ve told us in your statement that you weren’t able to sort of put your hand on any particular user feedback although you know some was created. I can take you to that if you like but –
Stephen Grayston: Yes.
Ms Page: Yes? Do you have any idea of why it’s not been possible to locate that at this stage?
Stephen Grayston: No. But, you know, all I can say is that there should have been a document library and an archive created that contains the full set of documents relating to the impact from start to finish, business change included.
Ms Page: Would that document library have potentially included records of board papers or anything of that nature?
Stephen Grayston: Yes.
Ms Page: Possibly even records of important meetings at which it was discussed?
Stephen Grayston: Yes. You know, it’s standard practice that, you know, a document library is created and then held, you know, for a considerable period of time.
Ms Page: So it’s slightly unusual, is it, that we find ourselves in a situation where we’ve got some papers but we don’t seem to have any meeting notes, we don’t seem to have any of your user feedback, in other words that what we’ve got is rather patchy?
Stephen Grayston: Yes. I think that’s – you know, although there’s a long period of time between today and what we’re talking about, you know, it’s unfortunate that there isn’t the record there to help the conversation that we’re having.
Ms Page: Thank you.
You’ve mentioned that you think that there should have been – whether there was or not we don’t know, but there should have been something of a report into this fivefold increase in the suspense accounts. Who do you think would have been responsible for that?
Stephen Grayston: It would sit in finance, with the finance team, to understand, investigate, analyse and produce appropriate outputs.
Ms Page: So perhaps Graham Corbett sitting at the top of that?
Stephen Grayston: Yes. I can’t – you know, I can’t remember the names, the particular names, at this point in time, but, you know, senior finance managers and, in particular, those that worked with the suspense account, yes.
Ms Page: Yes. Then you’ve told us also that you would envisage a report into the types and numbers of criminal prosecutions for discussion at Post Office management level. Again where would the responsibility for that sit and what managers would you have envisaged having those sort of discussions?
Stephen Grayston: Well, the investigation team, as a function, was at one point in time with Royal Mail Group, but then each of the businesses took on investigation in-house by taking some people from Group. So there was an investigation team. In terms of organisational structure I’m not sure whether the investigation team for Post Office Counters would have sat in finance or separately somewhere in operations, but, you know, if you were looking at weaknesses in your systems which are resulting in investigations taking place, then there is analysis that takes place at a national level to understand how many, what time, what amounts, so that it gives you the opportunity to close out and take rectification steps where, you know, there are weaknesses.
Ms Page: Did you ever see a document of that nature?
Stephen Grayston: Not – no. Maybe in the early ’90s, at a group level, because, of course, when you’re looking at the situation in the businesses, you do need to understand what’s going on in terms of investigations.
Ms Page: But you don’t believe you saw one during the period that the IMPACT programme was being developed –
Stephen Grayston: Certainly not, no.
Ms Page: No. But you believe one should have been done or something along those lines?
Stephen Grayston: I can only say that I would expect that the people involved in that would be doing that. They should be doing that, yes.
Ms Page: Just finally, you have very fairly acknowledged that the IMPACT programme required Horizon cash account data to be reliable and, of course, we know now that it wasn’t in a very large number of cases, perhaps not by any means a majority but a significant number of cases.
Looking back, do you think that as S80 was designed and created, alongside it, and perhaps not fully intentionally but certainly at some stage intentionally, there was a sort of development of a myth that Horizon cash account data was absolutely reliable?
Stephen Grayston: Myth … I think business decisions have to be based on an understanding that what is coming out of the system is accurate and reliable. If at a management level there is a suspicion that it may be flawed in some way then that causes or should cause, you know, a lot of thought and creation.
Myth – I’m not sure about “myth” but …
Ms Page: If there was perhaps an unwillingness to sort of investigate those possibilities?
Stephen Grayston: Yes, I think – you know, this is something that, having listened to some of the testimony, you know, they – stepping back and looking at what’s going on, making use of the various types of different view or data that would exist in the business, may have helped. I don’t know if that took place or not. But, having heard what I’ve heard, you know, in the lead up to being here today, you would expect there to be some stepping back and looking.
Ms Page: Thank you. Those are my questions.
Examined by Ms Patrick
Ms Patrick: Good morning, Mr Grayston. My name is Angela Patrick and again I act with Mr Moloney and Hudgell Solicitors for another group of subpostmasters.
I don’t have a lot of questions for you but Ms Kennedy has asked you a number of questions about your involvement in Horizon during the development stages, testing and acceptance, and during the rollout. I don’t want to go back quite that far but I want to look and ask a few questions about the end of the rollout, so before IMPACT.
Stephen Grayston: Right.
Ms Patrick: I want to look at two documents and ask a few questions about them. First is POL00104602.
Can you see that, Mr Grayston?
Stephen Grayston: Yes, yes.
Ms Patrick: We can see that it’s an email headed “Electronic memo”, from Dawn Howe to Keith Baines, sent on 6 September 2000. Can you see that?
Stephen Grayston: Yes, I can, yes.
Ms Patrick: It’s headed “Horizon NRO Close Down Reporting”.
“NRO”, would that be national rollout?
Stephen Grayston: It would, yes.
Ms Patrick: If we scroll down a little – we don’t need to look at the substance of that email but we can see it’s got a second email attached to the bottom part of that, and that’s an email from Don Grey copied to a number of people including, I think, yourself. You can see Steve Grayston there; would that be you?
Stephen Grayston: Yes.
Ms Patrick: This was one that was sent on 5 September 2000, and we can see again same title but it says “Initial draft for comment please … confirm requirements within NRO Board”.
So this is a document being sent to you for comment; is that fair?
Stephen Grayston: Yes.
Ms Patrick: If we can go over the page we can see what the document is, and we can see:
“This paper documents the process to be adopted by the Horizon Implementation Team to close the … National Roll Out project.
“… Issued for initial comment.”
So were you being asked here to comment on the plans for close down reporting, so – or how the close down reporting for the end of the rollout project was to be conducted?
Stephen Grayston: Yes, it was put together by Don Grey and, at that point in time, I think I was working for Douglas and part of Don Grey’s team, yes.
Ms Patrick: So you were part of the Horizon Implementation Team for the rollout?
Stephen Grayston: Yes.
Ms Patrick: Yes, and involved in conducting the review or part of it?
Stephen Grayston: Yes.
Ms Patrick: Thank you. This may be very basic but this was how the Post Office was proposing to learn any pertinent lessons they could from how the rollout had gone?
Stephen Grayston: I’m not sure in terms of scope whether it talked about lessons learnt. I mean, I’d need to sort of have a look at more of the document, but yes, I mean, it should refer back to lessons learned and, you know, opportunities for improvement, et cetera.
Ms Patrick: We don’t need to go into the detail of this document because it’s planning for how the review would be conducted. I’d like to look at the second document that I’d like to ask some questions about, and it’s POL00104482, please.
We don’t have a cover email for this but I can see on the top right-hand side, can you see that, Mr Grayston, there’s a date?
Stephen Grayston: Yes.
Ms Patrick: And it says “Draft”, and it seems to be 5 April 2001. So this is some time on from the initial email.
Stephen Grayston: Yes.
Ms Patrick: The heading is “Project Implementation Review – Horizon National Roll Out”.
Is it likely this was a draft of the review that you may have seen for your input?
Stephen Grayston: I don’t know. Documents – there should be one – or I would expect one report that Don was pulling together. There may be different takes on material in that report for different audiences.
Ms Patrick: If we can turn to page 10 – and there are appendices or annexes to this document, but if we look at page 10 to start with it may help with your memory. We can see appendix A is “Post Implementation Review of Field Management”, and if we scroll to the bottom of that page there’s a distribution list, which you aren’t on, but if we can scroll over to page 13, there’s an acknowledgments list at bullet point 2.
We can see there the second paragraph main contributors include Don Grey, Douglas Craik, Steve Grayston. So is it likely that you would have been a contributor to at least part of this review process?
Stephen Grayston: Yes, I mean, clearly from paragraph 1 what you have got is inputs from the field teams, the four field teams, and the management of those four field teams. I recognise all those names. Paragraph 2 is the head office team, yes.
Ms Patrick: Okay. As you said, there are some things that would have been within your domain, others that wouldn’t, but you may have been involved in reviewing different documents. You said that to Ms Kennedy earlier.
Stephen Grayston: Yes.
Ms Patrick: I don’t propose to go through all of this document. There are two issues I want to look at to see if you can help the Inquiry. Whether you have seen it or not it may refresh your memory if we look at it as we go through.
If we turn to page 5, the first issue that I wanted to ask some questions about arises there. We can see there’s a heading there headed bullet 5, and it says “Performance - Operational.”
Can you see that, Mr Grayston?
Stephen Grayston: Yes.
Ms Patrick: I want to scroll down to the sixth bullet point – sorry, the sixth point on that page, I apologise. It’s “Performance - Technical”. It’s actually the next section down. Thank you for your patience.
We can see the section there reads:
“Technical oversight and validation of ICL Pathway activities was almost non-existent compared with the preceding live trial and development phase. Although this was not really a problem it is an area that should not be overlooked either in the Horizon maintenance phase or in future projects.”
I think you can see immediately below there it says a full lessons learned report was going to be at appendix A, which we just looked at, and appendix B.
If we turn down to page 6, please, we can see some recommendations there, and 9.1 is headed “Supplier issues”, and I want to look at bullet point 2, which reads, if we read it together:
“We should never again put ourselves in the position of dependence on either a sole supplier (or, indeed, supplier dominated project progress information) without first establishing a defined and adequate contingency. At the outset we should assure customer pre-eminence with any future supplier who must commit to identify, agree and deliver to our requirements including detailed performance metrics and integrated reporting structures. Furthermore, any future supplier must empower their local field teams to mirror the responsibilities we invest in our people.”
Then if we can look at the third bullet it says:
“Improving the way we manage our chosen supplier; having more than one route without proper technical backup can make us look both unprofessional and vulnerable.”
I simply want to ask – I don’t know if this refreshes your memory of this at all but can you recall at the time this review at the end of the rollout was being conducted, was there a recognition within POL that POL had been very reliant on Fujitsu in the development and also during the rollout of Horizon?
Stephen Grayston: Well, in terms of what you’ve shown me and the comment that you have referred me to, this was about implementation, not about Horizon more generally. So on the point that I think is being made here in recommendations, ICL Pathway had subcontracted various pieces of work to different organisations and that led to difficulties through – and challenges through the implementation.
In terms of, I think, your question, which I think is wider, the reliance on ICL Pathway, yes, Post Office Limited – Counters Limited was reliant on ICL Pathway understanding the nature of their role and executing it appropriately and I think – sorry, I just add to that, I think you’ve already seen, and I’ve seen in the material, concerns over visibility and openness and the nature of the contract and the limitations of the contract.
So yes, I mean, I think this particular point was about implementation. I understand it. I do remember it.
Ms Patrick: I just go back to that phrase that was used on page 5. We don’t need to turn it back up again, but:
“Technical oversight and validation of ICL Pathway activities was almost non-existent compared with the preceding live trial and development phase.
Stephen Grayston: Yes.
Ms Patrick: Whether it’s implementation or not, the conversation there is about technical oversight and validation –
Stephen Grayston: Of implementation.
Ms Patrick: – being non-existent. Then there’s a reflection, continuing on, on “Improving the way we manage our chosen supplier”, and I think that’s forward looking for new projects, but can you recall if there was any concrete plan for change in the relationship between POL and Fujitsu to improve technical oversight and validation going forward?
Stephen Grayston: Well, the technical oversight and validation was around the steps that were required to undertake implementation, which was effectively, you know, a migration to the new world and so, to answer your question, no, because there would not be another technical rollout or implementation of a similar type with ICL Pathway. That activity had been done.
However, for Post Office’s purposes, you know, should we be working with another supplier (and we had a large banking programme, Post Office Card Account, for example), that the learning point about how we manage implementation, those points should be taken on board for future programmes.
Ms Patrick: Of course, I think you were continuing to work with ICL Pathway, and thereafter Fujitsu, on what we start calling the “business as usual” operation of Horizon.
Stephen Grayston: Yes.
Ms Patrick: And any other projects connected with Horizon that would be conducted by Pathway and then Fujitsu; is that fair?
Stephen Grayston: Correct, yes.
Ms Patrick: I’m going to leave that point and go to the second point and it’s to look at some of the detail in the lessons learned in this document.
If we could turn to page 30 to start, there are a few points I’d like to look at to see if they are consistent with your recollection of the review at the end of rollout. If we look at appendix B and we start at the bottom of this page, page 30, what I want to look at runs over the page on to page 31. If we can see the very last paragraph:
“The overall strategy towards training was not in tune with the contractual relationship that exists between Post Office network and subpostmasters. The requirement for subpostmasters and their assistants to be pass a PSA (Personal Standard Assessment) after training caused some inconsistent anomalies within the network in terms of offices reaching the minimum training compliance to enable migration to be completed. The lack of a proactive approach by Territories in this area. Detailed information on PSA failures and provision of training material from ICL Pathway have exacerbated the problem.”
On training, you’ve said a little to Ms Kennedy already this morning about your recollection of training. Is that something consistent with your recollection of concerns around training during the rollout?
Stephen Grayston: Well, even prior to rollout – first of all, you know, training was part of the programme that I think the Inquiry’s heard from one of my colleagues, Trevor Rollason. But, as a team, head office or regional, we were getting feedback on what was, you know, the struggle.
Yes. I mean, I think the work done pre-rollout to improve training, which was AI 218 I think, was seen as extremely helpful. But nevertheless, with a user population so large, there were people who could not cope with Horizon and they failed a test that had been introduced to assess competency.
Ms Patrick: We see just the paragraph below that one. It continues:
“The policy for ‘out-of-hours’ transactions is at best a stop gap. There are [key] client and account team issues that need to be addressed.”
This was being written in maybe 2001 at the end of the rollout. Can you recall what the key client and account team issues that still needed to be addressed were?
Stephen Grayston: No. It would have been clear at the time but out-of-hours transactions were used on occasions for certain product types but I can’t remember, in the context of what’s said here, what the implication was.
Ms Patrick: Okay. If we can go down to page 32, please, and I want to look at bullet point 3.4, please. Can you see that, Mr Grayston?
Stephen Grayston: Yes.
Ms Patrick: Thank you. You can see there – I don’t want to look that whole thing but the third paragraph down there is an entry which says:
“Cash account training was not comprehensive enough within the training delivered by ICL Pathway. The training delivered by ICL Pathway was poor in terms of the instructors had little or no knowledge of Post Office procedures.”
Again, just to be absolutely clear, is that consistent with your recollection as to the conclusions of the Post Office at the end of rollout in 2001?
Stephen Grayston: Yes, although looking at it today I think that there probably would need to be a reflection on what was done to boost that training but, you know, if I looked at it in a different way, Peritas (I think that was their name at one point in time) who had been appointed to run the training, didn’t have a Post Office background, didn’t understand all of the processes associated with it. So for Peritas, or the supplier of training, to come in and run good training courses, even with time and good material, was again, I think, a learning curve on their side.
I think there was a reflection that the cash account training wasn’t comprehensive enough and, through AI 218 and the negotiation that I think Bruce McNiven was involved in, that was improved. That was improved.
Ms Patrick: But this AI 218 takes us to acceptance and rollout which starts in January 2000. This is being drafted in May 2001.
Stephen Grayston: Yes.
Ms Patrick: It’s being recorded here that the training on the cash account had not been comprehensive enough. Was that, in your recollection, the view of Post Office in May 2001?
Stephen Grayston: Sorry, I wasn’t clear. I think in my mind this reflection should have had two elements to it that it absolutely wasn’t and that there was an intervention as a result of AI 218 that had improved things. It does not say that here and your interpretation, you know, is reasonable from what is said here.
But the quality of the training that was given, I believe was deemed to be adequate and the reason I say that is that there were four implementation teams nationally, and the head office team had worked with the regional teams through the lifetime of this programme, and the regional teams represented the business operations around the country and also reflected the needs of the programme in implementing in the various parts of the country.
If that feeling as expressed here was so black and white, then it would have been stopped. The regional management of Post Office Counters Limited would have stepped in. So I think – in my mind, you know, I am taking an interpretation that it wasn’t good enough and it improved. There was nothing coming out from the implementation teams or regional management that said every week this training is not could enough, it is not good enough. So you know that’s my thought on this.
Ms Patrick: But that’s not reflected in the draft that we have here?
Stephen Grayston: As I say, your interpretation of what’s said here, yes.
Ms Patrick: If we can go over to 3.6, which is over the page on page 33, it might help elaborate on this thinking.
Can you see that now, Mr Grayston? I think it’s come up. At point 3 6, which deals with pre and post Go Live support –
Stephen Grayston: Yes.
Ms Patrick: – and it says:
“More in-depth training for those people who supported second, third and fourth balance support especially around suspense account entries. The scheduling of Retail Network Manager was not consistent with instances of more than one arriving at an office to offer support. The allocation of support for balances worked better when the scheduling was undertaken by the cluster groups. Offices were given the impression that they would have a trained person with them for the first balance, far too many did not have anyone leaving them to ‘flounder’ with an inadequate balancing guide.”
If we scroll down further to 3.8, 3.8 deals with the documentation given to subpostmasters and it says:
“In the latter stages of the project changes arising from revised documentation have been deployed before the documentation had been signed off. Operational instructions and balancing guides were excellent, the quick reference cards poor as were the arrangements for CSR+. The distribution of documentation on the whole was poor with a number of offices receiving their balancing guides well after their Go Live.”
It goes on that the diagrams in the Horizon user guide were not well accepted as it contained too many flow charts, and it says some more about training.
Coming back to your understanding of the position of subpostmasters during the rollout, was this the reflection of the implementation team at the end of the rollout looking back that some SPMs, some subpostmasters, had been left to flounder?
Stephen Grayston: Well, from the position of the implementation processes and the role of the HFSO, which I knew because I’d been involved in the design of that role, it was an agreed process that, at the point of implementation and migration, the Field Support Officer would guide the manager and staff through the process and would be there at the first cash account after implementation and that subsequent cash accounts, if necessary, would have some level of support from the retail line; so business as usual retail operations as the implementation team was moving on.
So there was no intention of subpostmasters or any of their staff being intentionally left to flounder.
Ms Patrick: I wasn’t asking about what was intended – I apologise if there’s any confusion – simply that the reflection here, looking back on what could be learned from the rollout process, in May 2001 it was being recorded here that the Post Office was recognising that some subpostmasters had been left to flounder.
Stephen Grayston: That’s what – yes, that’s what it says.
Ms Patrick: Thank you. I have one last question. If we could look at page 34, please, at the bottom and I want to look at 3.10 which is headed “other”. Can you see that, Mr Grayston?
Stephen Grayston: Yes.
Ms Patrick: Thank you. This section starts:
“The helplines are not seen as an effective support to the network, there seems a lack of knowledge and a reluctance to pass to a higher level for resolution.
“Installing up to the 8th December was a mistake.
“The number of errors generated post Go Live is directly linked to poor cash account training, an extra half-day should have been allowed.
“The legacy left due to the migration use of the suspense accounts needs to be resolved.
“The rollout plan appeared to take no account of office size or pressure periods, this operational information should be included within the scheduling process.
“Overall the size of the project was immense and has been a success which is mainly due to attention to detail, focus, meaningful reviews and a lot of hard work by so many people.”
I have a few questions about this.
Sir Wyn Williams: I thought you only had one, Ms Patrick.
Ms Patrick: It’s one point, sir, but it’s about 3.10 which, as you can see, covers a lot of detail.
Sir Wyn Williams: I’m slightly concerned that we are revisiting Phase 2 exclusively in this part of your question and I’m not sure to what extent I want to do that; but, okay, one last point.
Ms Patrick: Thanks, sir.
We’re at the end of the rollout. Is this an understanding that at the end of the rollout at this point, May 2001, Post Office was acknowledging that the helplines were not seen as an effective support to the network?
Stephen Grayston: Well, that’s what it’s saying.
Ms Patrick: Thank you. Ms Kennedy’s already highlighted some problems would be problems that were flagged by subpostmasters in their branches.
Stephen Grayston: Yes.
Ms Patrick: Were helplines reluctant to pass up to a higher level for resolution when a problem got to them?
Stephen Grayston: I think you would need to speak to the Helpdesk management. They shouldn’t. It weakens and devalues the purpose of a helpdesk or a helpline if the appropriate action isn’t taken in terms of escalation.
Ms Patrick: You see there there’s a number of errors being generated post Go Live. Is that consistent with your recollection?
Stephen Grayston: I think there was a recognition that there were some errors as people were learning to use the system, yes. But there was no feedback that I can recall from the field teams and operational management that the level of challenge was so significant as to undermine the continuation of rollout.
Ms Patrick: So here at the end of rollout in May 2001 the errors are being attributed, it says “directly linked”, to poor cash account training. Is that consistent with your recollection?
Stephen Grayston: I don’t know. The author presumably, or whoever wrote this particular point, would have had the analysis to create that linkage.
Ms Patrick: Just as one of the individuals that were involved in the team putting together this review, we’ve already looked at the acknowledgement, the reference to non-existent technical oversight and validation during the implementation process. Did anybody involved in the review, in your recollection, consider whether these errors that were arising post Go Live might not be attributable only to training but to problems with the technology itself?
Stephen Grayston: Yes, that’s a very good question. At the time – at the time – you know, I think the working assumption was that the system was reliable and robust and producing outputs that could be trusted, and therefore the reflection of cash account or training is what you see here. Whether that was, you know, an assumption that was appropriate is now very questionable.
Ms Patrick: This is the last question: if anybody in your team or anybody else in POL at all, maybe involved in this review or not, can you recall if anybody joined the dots or tried to join the dots between a lack of technical oversight and validation and continuing problems with the cash account?
Stephen Grayston: Sorry, I just need to take you back to your linkage here. The technical oversight was about implementation, technical aspects of implementation, infrastructure, hardware, software, software failures, and aspects of that oversight for implementation.
If you’re asking me the about joining the dots in a more general sense, there were challenges, there were discrepancies and was anybody stepping back and looking at this overall, I don’t know that there was.
Ms Patrick: Thank you, Mr Grayston. We don’t have any more questions for you. Thank you, sir.
Sir Wyn Williams: Thank you everyone. And thank you, Mr Grayston, for coming to give evidence and answering the questions put to you.
So is that it for today, Ms Kennedy?
Ms Kennedy: Yes, Chair. We return tomorrow with Mr Shaun Turner and Ms Anne Allaker.
Sir Wyn Williams: See you in the morning. Goodbye.
(12.17 pm)
(Adjourned until 10.00 am the following day)