WITNO7270100
WITNO7270100
Witness Name: Robert Arthur George Ellis
Statement No.: WITNO7270100
Date: 17 March 2023
POST OFFICE HORIZON IT INQUIRY
FIRST WITNESS STATEMENT OF ROBERT ARTHUR GEORGE ELLIS
I, MR ROBERT ARTHUR GEORGE ELLIS, will say as follows:
INTRODUCTION
1. I am currently employed by Fujitsu Services Limited (“Fujitsu”) as a workshop
engineer and have been employed by Fujitsu since around March 2001. I am not
involved in the Horizon project at present and have had no involvement in it since
2006.
2. This witness statement is made to assist the Post Office Horizon IT Inquiry (the
“Inquiry”) with the questions put to me in a Rule 9 Request dated 19 January 2023
(the “Request”). It is based on my direct knowledge of relevant matters. I was
assisted in preparing this statement by Morrison Foerster, who represent Fujitsu in the
Inquiry.
Page 1 of 9
WITNO7270100
WITNO7270100
3. The topics in the Request relate to matters that took place more than 15 years ago. I
have tried to remember relevant material so as to assist the Inquiry as far as possible
but have not always been able to do so.
BACKGROUND
4. I studied Computing at A-Level and completed City & Guilds qualifications in
Computing at Sitec College.
5. My professional experience prior to joining Fujitsu was as a car mechanic.
6. As noted above, I joined Fujitsu in February 2001. I was employed as a first-line
Horizon System Helpdesk (“HSH”) operator, based in Manchester. My role was to
provide initial support over the phone to Horizon users on hardware and software
issues.
7. IIn 2003, the HSH’s operations moved from Manchester to Stevenage. As I did not
wish to relocate with the desk, I left the HSH and became a field engineer on the Post
Office account. My role included repairing and replacing hardware.
Page 2 of 9
WITNO7270100
WITNO7270100
8. In 2006, I stopped working on the Post Office account and moved to a role within
Fujitsu’s defence contracting division. I have not been involved with the Post Office
account or Horizon since then.
TRAINING
Four-week initial course
9. Before I was permitted to work on the HSH, I was required to complete a four-week
training course. I completed this course at the same time as other new HSH joiners.
10. The main part of the course involved each of us working through a substantial training
manual individually. The manual contained information about working on the HSH and
detailed scenarios that we might encounter as a first line operator. We had access to
a training system (which was an offline version of Horizon) as the manual contained
various exercises, we needed to complete each day.
11.Our progress through the manual was supervised by senior HSH operators (who were
called team leaders). The team leaders would regularly check in on us and answer
any questions we had. They would also verify that we had completed the required
exercises each day.
12.In addition to completing the manual, there were also sessions in a more “classroom”
style environment, where we would be shown how parts of the Horizon system
Page 3 of 9
WITNO7270100
WITNO7270100
worked. For example, I recall one session where the trainer stripped the printer that
was used in branches, and explained how it worked and common problems that might
occur.
13.We were working through the manual in the same office as the HSH itself, so there
were also opportunities to listen in on real-life calls and see how the more experienced
operators dealt with them.
Buddy training
14.After completion of the initial four-week course, we were able to begin answering
actual calls into the HSH. For the first two weeks of doing that, a team leader would
listen in to most or all of the calls we took.
15.The team leaders would provide guidance on how to approach issues raised by
callers. On some occasions, the team leaders would take over the call if we were stuck
and unable to help the caller. I recall receiving regular feedback during this period
from the team leaders listening in to my calls. This included being asked to go over
certain sections of the manual again to improve my knowledge of them.
Adequacy of training
16.In my view, the training I received before starting work on the HSH was very good. I
found the course was thorough and included a lot of detailed examples of scenarios I
might be required to deal with as an HSH operator. No course can prepare you for
Page 4 of 9
WITNO7270100
WITNO7270100
every part of a job. However, looking back, I felt the training I received was good
preparation for the actual job of being a HSH first-line operator.
EXPERIENCE WORKING ON THE HSH
Daily experience
17.While I worked on the HSH, all the operators sat within a single large room, with rows
of desks in it. First-line operators shared the space with the second-line team. I do not
recall if we worked in shifts, but I generally only worked standard business hours.
18.On average, I recall answering anywhere between 20 and 50 calls a day. The number
of calls I was able to take depended a lot on the complexity of the issues that came
up that day, as some calls could require a lot of time to resolve. The issues that came
up could be varied, and so there was no “standard” day.
19.In my view the HSH was professionally managed. There was a number of team
leaders on the desk, who supervised the other operators. I do not recall having any
problems with them. I felt the desk had a positive environment, and do not recall others
being unhappy with how the desk was managed.
Types of calls received
20.The most common types of calls I recall receiving related to hardware issues (for
example, a printer not working) and how to reverse transactions (for example,
Page 5 of 9
WITNO7270100
WITNO7270100
because the user had entered a transaction onto the system in error). There was no
standard approach to resolving hardware issues, as it would depend on the hardware
that was not working, and the specific problem that was occurring. In relation to
reversals, the call was usually resolved by guiding the user through the steps of the
reversals process while they did it in branch.
21.In order for a call to be closed, the person calling in had to agree to the closure. I do
not recall ever closing a call without the agreement of the person who had called in.
22.1 understand the Inquiry may be particularly interested in calls from sub-postmasters
(“SPMs’”) that related to balancing issues. I do not have strong recollections of these
calls. I recall receiving much fewer of them than calls relating to hardware issues or
reversals.
23.The approach I usually took to balancing issues was:
a. I would first guide the SPM on how to reverse the balancing exercise they
had just done and which led to the error.
b. I would then work through the whole balancing exercise step by step with
the SPM on the phone. The SPM could provide their pre-balancing account
information and I could populate this data into an offline version of the
Horizon counter, which I had access to. As a result, I could do each step of
the balancing on my offline system alongside the SPM, and check that our
Page 6 of 9
WITNO7270100
WITNO7270100
figures tallied at each stage. This often helped with identifying what had
gone wrong.
c. My recollection is that, in most cases, SPMs were able to successfully
balance their accounts. In some instances, the figures may have been a few
pounds off the amount the SPM believed the system should have, but most
SPMs were usually willing to disregard that and consider the issue resolved.
24.1 do not recall dealing with any balancing calls where large amounts of money (i.e. in
the thousands of pounds) was unaccounted for. If I was unable to resolve a balancing
Call to the satisfaction of an SPM, I would pass the call on to a second-line operator.
Resources available to operators
25.In terms of resources available to assist with answering calls, I recall having access
to the manuals we used in training, which included detailed instructions on how to
resolve a wide range of scenarios. We also had to hand in the office all the hardware
that a user would have in branch. This would allow us to effectively walk a user through
resolving hardware problems, especially if they needed to take apart the hardware to
fix an issue.
26. If I did not know how to resolve an issue or the instructions in the manual were not
working, I would pass the call on to a second-line operator (and sometimes the POL
Page 7 of 9
WITNO7270100
WITNO7270100
Helpdesk) who would take over the call. Once a call had been passed to the second
line, it would not return to me and I had no visibility over how it was resolved.
Attitude towards SPMs
27.Overall, from my experience working on the HSH, I believe HSH operators (myself
included) dealt with SPMs in a polite and professional manner. I do not recall there
being any particular attitude held by HSH operators towards the SPMs or anyone
behaving unprofessionally towards them. While there were certainly calls where an
SPM was very unhappy with encountering an issue or appeared to be under a lot of
pressure, I personally regarded that as just part of the challenge of working in a
frontline customer service role.
Adequacy of support provided
28.From my personal experience, I believe I did a good job assisting SPMs who had
called into the HSH. This is because I remember resolving the relevant problem in the
vast majority of the calls I received, and only escalating a few calls to the second line.
While I do not know how other HSH operators were performing, my sense while
working on the desk was that most cases were resolved at the first line.
29.1 recall that minimum standards of competence for operators were enforced. There
were individuals that did not complete the initial four-week training as they were unable
to cope with the training material. Our calls were also regularly monitored by team
leaders. There was a 6-month probation period after training, and some individuals
Page 8 of 9
WITNO7270100
WITNO7270100
were not retained at the end of that period because they were not performing
adequately on the calls.
BUGS, ERRORS AND DEFECTS
30.1 do not recall being aware of any bugs, errors or defects in the Horizon system while
I worked on the HSH. As I have noted above, if I was unable to resolve an issue on a
call, I would pass the call on to the second line and I would have no visibility over how
the call was then addressed. For example, I would not have been aware if a call was
passed on to third line or if a bug had been identified as a result of a call.
Statement of Truth
Signed:!.
Dated:
Page 9 of 9