POL00028094
POL00028094
BA/POCL Automation Programme.Review
HM Treasury Independent Panel
. ; Report
Adrian Montague
Bill Robins
Alec Wylie
July 1998
al
POL00028094
POL00028094
CONFIDENTIAL - COMMERCIAL .
BA/POCL Review - Independent Panel report
CONTENTS ; Page
an
¢€
z
2
°
a
a
in
fe}
fe}
=
=
m
S
>
go
fe}
=
a
rs
PART I - INTRODUCTION
Terms of Reference .
“Methodology... ..
Business drivers ....
Brief description of the syste
pom>
PART Il - THE PANEL'S FINDINGS AND RECOMMENDATIONS
A Current Status of the Programme ......
B. Solution Design and Fitness for Purpose
Cc Future Proofing
D. Timetable for completion ...
E Programme management, organisation and structures
FL Strategic Commitment a
G.
‘BRRERBBE iui pate
. Likely costs of delivery ...
PART Ill - RESIDUAL ISSUES AND RISKS - 418
A Strategicissues .. 18
B. Critical path issues 49
Cc. Operational issues ig I
PARTIV- SUMMARY OF FINDINGS ... . adeaee eee eee cent ence eee cece ee eeeeeteeeees i
ANNEX A - THE PANEL'S VIEW OF A POSSIBLE WAY FORWARD
Option 2: Partial Restructuring .
Option '1: Full restructuring .
+ 22
22
23
: Weighing the options . 24
ee Financial Implications . 25
Conclusion ..... 26
{mplementation of a rest 26
Contractual/legal issues - 22
ANNEX B - THE INDEPENDENT PANEL'S RULES OF PROCEDURE ..............0000005 28
ANNEX C - THE PANEL'S MEETINGS WITH THE PARTIES - 34
ANNEX D - OUTSTANDING ISSUES .
POL00028094
POL00028094
CONFIDENTIAL - COMMERCIAL,
BAPOCL Review - Independent Panel report
EXECUTIVE SUMMARY
Background
In May 1996, the Benefits Agency (BA) and Post Office Counters Limited (POCL) signed PFI
contracts with ICL Pathway. Pathway would set up and maintain a secure, automated
infrastructure for POCL, to allow benefits to be paid to customers presenting a special
magnetic strip card and to ) provide a platform for other POCL business.
Through the programme, BA aims to eliminate benefit encashment fraud, provide better
accounting and Management information, reduce administration costs and improve service
delivery. POCL aims to maintain its income from benefit payments and other transactions
with benefit customers, replace old systems, modemise the network and develop new
services and commercial Opportunities, with Pathway as a partner.
In the light of concems over progress, this Panel, chaired by the head of the Treasury Task
Force on Private Finance, was set up to make an independent assessment of whether the
programme was technically viable, if so how quickly it could be completed and at what cost.
Findings
The programme is complex, probably the biggest of its kind. Its scale, and particularly the
development work required, were underestimated initially. The parties have since.increased
the resources devoted to the programme, but a range of issues remain to be resolved.
Our view is that the programme is technically viable. There must be some risk around
scaleability and robustness because the system has had to be tested at the level of
component parts, but we are satisfied these risks are being well managed by. Pathway.
There is good evidence of future proofing at all levels. The basic infrastructure is very robust
for the future and, in the main, industry standard products have been used. The system
should allow POCL to compete for new business in a variety of markets, including banking
and financial services. New applications based on smartcard technology should be relatively
straightforward and economic. If on-line applications are required, they may take longer and
require more investment.
A further nine months delay to the programme is our best forecast, with September 2001 for
national rollout completion. Critical path issues will have to be resolved fast to make this
possible, and the date could be brought forward with commitment and goodwill on all sides.
The new Horizon Programme Office in POCL has a vital role. The parties need to ensure it
has the power and the resources to drive forward the whole programme end-to-end and to
resolve critical outstanding (and future) issues.
Driven mainly by timetable slippage, the sponsors’ business cases are eroding. The direct
cost of delay is estimated at £180m, over half of which falls to the sponsors. Potential
savings from fraud reduction would also be delayed. Pathway, on the basis of the figures it
has provided, would make an overall loss if the contract continued on its present terms and
would require an extension to break even.
POL00028094
POL00028094
CONFIDENTIAL - COMMERCIAL
BA/POCL Review - Independent Panel report.
Possible way forward.
Given these findings, we sought to find a viable way forward for all the parties, subject to
negotiations about the detail. Although the parties did not all agree, we believe that a
restructuring of the full programme cold offer such a way forward. . 4?
This restructuring would extend the use of the card beyond the current contract énd date. BA
and POCL would prepare for a rapid increase thereafter in benefit payments via the banking
system (ACT). POCL could by then be ready to offer a competitively priced service for
customers who still wished to use Post Offices for access to cash in this new environment. :
Pathway would be closer to recouping its investment.
Asecond option, less risky in programme management terms, would be to de-scope the
programme by stopping the benefit payment card, while still allowing time for BA and POCL
to prepare for ACT. However in our view this‘option would run a higher risk of protracted and
costly contractual dispute, possibly leaving POCL without an automated network. It would
also prolong the use of paper based methods of payments.
. The first option (full restructuring) may offer the prospect of better value for money in the _
longer term although, to achieve this, the programme would need to be driven much harder.
and more professionally than so far. .
If Ministers were to decide on any restructuring, it would be for the parties in negotiation to
agree the details. ; :
We suggest a neutral “troubleshooter, directly responsible to Ministers and with their full
confidence, be appointed to’assist this process, to ensure all the parties give théir full
“commitment to implementation of the programme and that rollout is completed at the earliest
agreed date.
POL00028094
POL00028094
CONFIDENTIAL - COMMERCIAL
BA/POCL Review - Independent Panel report
SUMMARY OF RECOMMENDATIONS
To help the reader, we include below ‘headlines’ for our recommendations, with paragraph
references to the ‘main body of the report:
* BA can do more to ensure that the other parties have confidente in its plans. Br
Paragraph
The need to commit to a firm baseline and plan becomes critical over the next few weeks. 23
As preparations for national rollout proceed, the need to clarify t! the resourcing : and
functioning of POCL's Service Management organisation will become increasingly urgent. 32
We see considerable weight falling on the Horizon Programme Office, which needs to be
properly resourced in terms of numbers and skills... .....sseeeeeeeeeeee v... 52and 92
All parties should agree to an “open book" approach so that an end to end plan for the
programme can be arranged and driven through by the HPO. ...........eeeeeeeeeee 56
The HPO must drive for resolution of critical path issues before end July.
BA's constructive engagement in the programme management process will be needed to
bring the programme to a successful conclusion. 3
The HPO should manage the leaming process from Release 1c and testing/trial activities. 62
Resolution of a number of strategic issues is fundamental to the programme ......... 184 :
Pathway also has work to do to convince the other parties that the solution is complete and
scaleable. sf... cee cece e ccc ed eee cere e eee cen cece e seen eeeeeereeeees 89
‘The current adversarial approach of the parties will have to change.~ ...........seeeee “91
HPO and the Horizon Programme Board need real authority within POCL, with aremitto
drive forward progress on outstanding acceptance ISSUES... ..... 6 cee e eee e cece eee eee 92
Depending on the way forward Ministers decide, we would also suggest
Aneutral ‘troubleshooter’, directly responsibié to Ministers and with their full confidence,
should be appointed, . Annex A.
‘The Post Office Review should consider whether refinements to POCL's present
management structure would better allow it to.implement the changes required. .. AnnexA
The opportunity should be taken to restructure the contractual relationships ..... -. AnnexA
POL00028094
POL00028094
CONFIDENTIAL - COMMERCIAL
BAIPOCL Review - Independent Panel report
PART I - INTRODUCTION
A Terms of Reference
1. - In May 1996, after a two-year PFI procurement process, the Benefits Agency (BA) and Post
Office Counters Limited (POCL) signed contracts with ICL Pathway (Pathway) under which Pathway
would provide BA and POCL with a wide range of services. The'programme involves the setting up
and maintenance of a secure, automated infrastructure for POCL, which will allow payment of social
security and other benefits across Post Office counters to customers holding a magnetic strip card
and provide a platform on which other POCL business could be transacted. °
2. In the light of concems over progress with the programme, a Panel chaired by the head of
the Treasury Task Force on Private Finance was set up to make an independerit assessment of
whether the project was technically viable and if so how quickly it could be completed and.at what
cost. The Panel was to report to a working group of officials from HM Treasury, DTI and DSS.
3... This report fulfils that remit. Specifically we, the Panel, were asked to assess:
. whether the project can deliver a fully functioning system which meets the project
+. °. specification, and integrates fully with BA and POCL computer systems;
. ‘ whether the timetable for completing the systems development, and starting and
4 completing roll out, is deliverable and whether the necessary managerial and
e organisational structures are in place;
. the likely costs of delivery, under current contract dates and with extension; and
en in each of these areas, the risks associated with these assessments, and whether -
tobust monitoring arrangements and disciplines are in place. .
4. All parties expressed a willingness to consider suggestions the Panel might wish to make for
taking the programme forward successfully, and we have proceeded on the basis that our terms of
reference should be read as including a power to make recommendations. Our view on a possible
way forward is at Annex A. The Panel's rules of procedure are at Annex B.
B. Methodology
5. We have received written submissions and oral presentations from all the parties both
separately and together. We have seen Pathway products demonstrated and been briefed on the
wider. Government information systems agenda by the Central IT Unit (CITU). The panel's schedule
of meetings with the parties is at Annex C.
6. PA Consulting has assisted us with detailed technical and programme planning
investigations, conducting further interviews with the parties. The report draws on the parties’
submissions and discussions with us, PA's findings and our independent view. A preliminary view of
our findings has been discussed with the parties. .
7. We would thank each of the parties for their open and co-operative approach to our
investigations, recognising that their contributions have been given without prejudice to their
contractual rights and remedies. may
POL00028094
POL00028094
CONFIDENTIAL - COMMERCIAL. -
BA/POCL Review - Independent Panel report
c. Business drivers
8. In assessing the viability, timeliness and cost of the programme, we started from the parties’
business drivers, taking account of how they may have changed or been added to since the start of
the programme. The programme has to satisfy the various interlocking objectives of the parties and
is, therefore, complex and wide-ranging.
ene} je .
9. For BA, the business drivers are: te
. to eliminate benefit encashment fraud, estimated at £190m per year, ~*
. to provide adequate accounting and management information for BA;
. to reduce administration costs; and
. " to achieve a fundamental improvement in service delivery.
Post Office Counters
10. For POCL, the drivers are:
. ,.to maintain income from benefit payments and ‘footfall’ (ie other POCL transactions, .
with benefit customers);
. to replace legacy systems and modemise the nationwide Post Office network;
. to provide a platform for modemisation and development of other client services;
and :
. to acquire a strategic private sector partner to bring additional commercial
a, opportunities. . .
thwai
41. Pathway’s objectives at the start of the programme were:-
. to.meet sponsors’ service needs by integrating a number of systems:
“* todo this ata competitive price while achieving a reasonable rate of retum over the
period of the contract;
. to establish a world class solution which could be marketed intemationally; and
wt to become POCL's preferred strategic partner in the development of its wider
business.
12. These remain Pathway’s objectives, but the first has had to be expanded to include
additional design, development, building and testing of the systems needed to deliver the services
required. The enhancements required and delay in their implementation has caused a deterioration
in the financial retums the programme offers to the point where Pathway asserts that an extension of
the contract is required to restore its financial equilibrium.
POL00028094
POL00028094
CONFIDENTIAL - COMMERCIAL
BA/POCL Review - Independent Panel report
Wider Government
13. Beyond the parties to the contracts, the Government has an interest in the Programme’ 's
potential contribution to a number of objectives, including:
+ the preservation of the social value of a nationwide network of POs;
. the reform of the welfare system;
. a move to an ‘active modem service’ for benefit customers;
. the target of completing 25% of Government transactions electronically by 2002; and
. _ the delivery of government.direct.
D. Brief description of the system
14,” The programme comprises a number of interdependent systems and services, Tequiring
work to be completed by Pathway, BA and POCL. Somewhat simplified, these systems are as
described and shown in the diagram below.
15. On the benefit payment side, in the Pathway domain:
. . the Post Office counter clerk uses the Benefit Encashment Service (BES) to make a
. _ benefit payment to customers presenting a benefit payment card. The clerk swipes -
the card through a magnetic strip reader and is then prompted by the system through
the rest of the transaction;
. BES uses data from the Card Management Service (CMS), which holds details of
cards issued, in service, stopped etc, and from the Payment Authorisation Service
(PAS), which holds details of all payment instructions received by Pathway;
. CMS and PAS link via the CAPS Access System (CAS) to BA systems;
. Payment Card Data Files (PCDF) link CMS to de la Rue which produces and
despatches cards and pick-up notices to customers;
. separately, the Order Book Control Service (OBCS) is a anti-fraud measure which
checks bar coded order books (until such time as they are superseded by the card)
against a national ‘stop’ list.
16. Inthe BA domain, CAPS systems exchange data with CAS. These systems are:
. the Personal Details Computer System (PDCS), storing personal details for BA
customers;
. the Customer Payment Computer System (CPCS), containing details of payments in
and out; and
. ‘an accounting system (PACS);
these systems rely on Feeder Benefit Systems, which support each benefit payment
system and provide details of payments due; and separately
. the Electronic Stop Notice Control System holds details of stop notices.
17.. For POCL and its other clients, Pathway provide:
. an Automated Payments Service (APS) for bill payment and pre-payment using
7
POL00028094
POL00028094
CONFIDENTIAL - COMMERCIAL
: BA/POCL Review - Independent Panel repot
18.
magnetic strip cards and smartcards; and
ve Electronic Point-of-Sale Service (EPOSS) for electronic cash register functions - the
‘glue’ connecting together other services at the PO counter.
Pathway’s systems rely on, or provide data to, others in the POCL domain, particularly:
. the Reference Data system, which holds details of post offices, products, prices, etc
. the Host Automated Payment System (HAPS), handling existing automated systems
and taking a feed from Pathway's APS; and
. Transaction Information Processing (TIP) taking data from Pathway for POCL’s
accounting and management information systems.
Benett cand
e bs ‘Maker
BA Systems Pathway Systems A POCL Systems
CAPS
Bevel (Payne
Sytems. <p Systern H~_
POL00028094
POL00028094
CONFIDENTIAL - COMMERCIAL
BA/POCL Review - Independent Panel report
19. _. Service charges from,BA and POCL to Pathway are calculated from a matrix of unit
transaction prices. BA have a separate contract with POCL for counter services, the charges
including fixed, semi-fixed and variable elements, with a ‘floor’ providing that payments cannot fall by
more than.a specified amount in any.one year, regardless of volumes. A simplified diagram of the
flows is shown below.
Horizon programme payment flows
Benefit Payment Card and other applications
Variable counter
charges
-\ Fixed charges
Card Management
and Payment
Authorisation
Service charges Encashment
Service charge: Income from
: other clients
Payments to”
subcontractors *-
20. ‘The service interdependencies mean that releases of Pathway systems, CAPS and POCL
feeder systems need to be coordinated. Separate testing of systems, followed by integrated testing
of ‘model offices’ and live trial, lead to the full national rollout of the service to all post offices and BA
district offices, subject to acceptance tests at appropriate stages. Benefits migrate to the new system
according to a schedule drawn up by BA. Finally the programme is complete and responsibilities for
continuing service provision transfer to the supplier and sponsors.
POL00028094
POL00028094
CONFIDENTIAL - COMMERCIAL
BA/POCL Review - Independent Panel report -
PART Il - THE PANEL’S FINDINGS AND RECOMMENDATIONS
A. Current Status of the Programme
21. The programme has moved on since PA reviewed it towards the end of 1997. Indicators of
Progress are:
. _ Release 1c (Ric), a partial Solution providing the benefit payment card and OBCS,
«has been working satisfactorily in just over 200 offices since November 1997;
:« "' Pathway has brought in new technical skills and management resources, increasing »
. headcount to around'270.staff and introduced new procedures to support the high
level of software development needed;
. BA has increased its resources on the programme and Release 3.0 of its key feeder
system (CAPS) has been given DSS Seals of Approval;
. POCL has also increased its resources on the programme, establishing a pilot
service management function and a National Implementation organisation, to
support Pathway in preparing outlets and training; the Horizon Programme Office
(HPO) also started work on 14 April 1998, .
22. However there remain problems, and difficulties in formally signing-off requirements and
solutions so that delivery dates can be planned and agreed. For example: .
ve there is not yet a stable baseline requirement formally agreed by all Parties on which I
plans and key milestones can be agreed;
. the parties have yet to sign off proposals to de-scope Release 2 (originally intended
. as the full solution) into New Release 2 (NR2) a partial solution ready for the start of I
national rollout and New Release 2+ (NR2+) the full solution to be available later,
. there is no agreed Acceptance Plan or time-scale for acceptance, which puts at risk I
the timetable for contractual acceptance of the system; .
. there is no consensus on the length of Model Office testing, live trial (the final stage \
before rollout) and the contingency to be allowed;
. ~ there is no agreement on the rate of rollout, or ‘beat rate’; {
. -there are no agreed timescales for change control decisions;
. version 4 of the Master Plan (covering the whole programme) has not been signed {
off and there is no formal agreement about the conditions for deciding that rollout
has been completed; and. ~
. hence the dates proposed for the start of live trial and rollout ‘to all 49,000 post I I
Offices are at risk.
23. Under Pathway’ 's current plans, preparations. for national rollout begani in eamest on 8 June.
1998. This is not a point of no retum, but it does mean that increasingly more significant.
commitments will be made on the way to the start of national rollout in April 1999. So the. need to
commit to a firm baseline and plan becomes critical over the next few weeks.
B. Solution Design and Fitness for Purpose ~
24, We believe the programme can deliver the contracted functionality. The technical
architecture is necessarily highly complex: systems must be separated for contractual reasons; they
10
POL00028094
POL00028094
CONFIDENTIAL - COMMERCIAL
BAYPOCL Review - Independent Panel report
have to have high levels of resilience and security; and they employ a number of different computing
platforms. These factors have been well considered i in the design‘and structuring of the project and
technical architecture.
25. The main architectural issues are scaleability and robustness. We are advised that a solution
of this scale and scope with so many different platforms and products has, as far as PA is aware, no
precedent. We are satisfied that Pathway's approach to design, development and performance
testing is sufficiently rigorous for such a major undertaking. At this stage of development, testing has
to be based on the component parts rather than the entire system. We believe this has been
completed in an appropriately structured way; indeed it is a strength of the modular nature of the
architecture that this approach can be taken. Significant attention is being given to testing at the
correspondence server level where the highest risk of congestion occurs.
26. Given the size of the system, there is an unavoidable risk that it cannot all have been tested
end to end in eamest. However we are satisfied that Pathway has contingency plans to upgrade
individual components of the architecture, should that prove necessary.’ We therefore assess the risk
of the entire solution failing to operate as expected ' be as low as could be achieved in the
circumstances. . *\ ‘
27. Although we believe the architecture to be viable, there is a concem that the system is
(necessarily) heavily dependent on the third party middleware product 'Riposte’. This risk will persist
and steps must be taken to manage this risk over the operational lifetime of the system (in addition to
those steps already taken in the development stages by ICL and Pathway). If, as is confidently
predicted by ICL, this product becomes a Postal industry standard, this risk is significantly mitigated.
Pathway has also taken steps to cover their dependency’ on Riposte by holding a copy of the source
code and by training their staffl inits use.
28. The complex architecture demands a supply of highly skilled technicians across a wide
range of disciplines. Pathway has put in place extra staff with strong development expertise. There
are now forward tesourcing plans and extensive documentation supports the knowledge and
expertise built up in people's heads. PA has been most impressed with the progress Pathway has
made since last year in this respect.
29. The project is probably the biggest of its kind and many of the component parts, although
sourced from industry strength products and companies, are being used towards their current limits
and scale. Pathway has recognised the risks and has in place the controls we would expect to see in
a development project of this scale.
30. ° The architecture is modular and as the number of on-line offices grows, So specific server
types can be added incrementally. Pathway has ensured that, at all levels of the architecture, more
computing power is available from chosen Suppliers should that currently I planned for reach the limit
of its capability.
31. There is no reason to suggest BA will not be able to deliver the. CAPS functionality required.
Its resourcing and testing of this part of the programme, and its track record so far, all give
confidence. As with the Pathway systems referred to above, there must remain some risk until the -°
systems are connected together in eamest. The same is true of POCL's parallel work on its feeder I
systems (Reference Data, TIP, APS etc.) —
32. POCL has been establishing plans for its Service Management organisation and we are
satisfied that the requirement is properly understood, although POCL has not been able to describe
to PA exactly how the proposed Service Management organisation will be resourced nor say how it
will fit into POCL's organisational structure. What is clear however. is that the system will demand a
W
POL00028094
POL00028094
CONFIDENTIAL - COMMERCIAL
BA/POCL Review - Independent Panel report
degree of central contro! and management not evident in the present regionally based POCL
organisation. As preparations for national rollout proceed, the need to clarify the resourcing and
functioning of POCL's Service Management organisation will become increasingly urgent.
jo. C. . FutureProofng - - wt
33. - " There is good evidence of future prodfing at all levels. We have been satisfied that all
reasonable steps have been taken to ensure robust sources of supply and compliance.with industry
standards in designing the architecture. Upgrades to software Platforms and ic components
are provided for, should they be necessary.
34. *° The networking infrastructure and the Riposte messaging system use the ToPaP protocol.
This protocol is becoming an industry standard. The ISDN backbone network is anothiér technology ° .
that will be around a long time.
35. Pathway has taken steps to ensure the future of the Riposte messaging software, but there
must continue to be a Concem about this product unless and until such time as it becomes much *
more of an industry standard. Having said that it is worth noting that there are now Post Office
automation projects based on Riposte committed in: at least five more countries since this contract
was placed.
36. There is evidence that the architecture will support a variety of applications. The
ly. infrastructuré technology used will allow Intemet type-developments. HTML and Java could be used
I . for expansion into other services. The application code has been written using industry standard
I
I
tools and languages such as Visual Basic and Pro C, which ensures that it can be adequately
covered in the future by another agency if necessary.
37. Part of the'system design was to make the equipment very user friendly, with touch screens
and the like. Conversion to a “kiosk” type system would be relatively straightforward.
38. The system has been designed to handle smartcard-based applications. Benefit payment
applications use a magnetic strip ‘swipe’ card, but future applications could use a smartcard instead.
. If new applications are focused on smartcard technology, then to implement each one should be a -
7 relatively straightforward and economic process as far as the basic infrastructure is concemed,
although the necessary changes to correspondent systems may be more complex.
! 39, The basic structure to support banking applications will bein place. Pathway will have NT
machines in the local branch, ISDN lines to all branchés and a managed TCP/IP network available -
across the UK. This will provide the basic, generic structure needed to run local applications on the
counter and to support specialised terminals such as cash dispensers.
40. One caveat is that the whole system is basically designed as a batch system. It can, and
does, go on-line, but the Riposte system is not centred around On-Line Transaction Processing
(OLTP); it works on a regular collection (harvesting ) of messages. The architecture may be able to
support a reasonable OLTP performance (and Pathway may have tested this) but it is not part of the
-contracted functionality. Real time on-line applications, such as those used by the National Lottery, \
could therefore require major changes to the system.
41. Wedo not know whether OLTP will be required to support POCL’s future banking
applications. It will depend on the security needed whether, for example, cash withdrawal requires
on-line authorisation. The advantage of smart cards is that significant parts of authorisation can be
212
POL00028094
POL00028094
CONFIDENTIAL - COMMERCIAL
BAVPOCL Review = Independent Panel report
local, with less need to go on-line.
42. Until POCL decides which banking applications are to be provided, how they are to be
provided and, if using a smart (or other) card, on which technologies and standards they are to be
based, there will remain some uncertainty about how economically the system will support such a
venture and whether the Riposte architecture poses any limitations. A smartcard route should pose
few problems. ' :
'D. Timetable for completion
43. There have been a number of re-plans since the contracts were signed. The latest plan to
which the parties are working (although not contracted) shows national rollout to start in April 1999,
some 21 months later than was initially planned. .
44. ~ There is no end-to-end schedule for the entire programme agreed between the parties, and
no end-to-end critical path analysis has been done, We,consider this to be a serious deficiency in the
management of the programme. In order to establish what risk there might be of further delay, PA
carried out their own critical path analysis. .
“45. The likelihood of further slippage is increased by a number of factors:
. detailed acceptance criteria and the process for acceptance have yet to be agreed;
22 - the parties have yet to agree on: the duration of model office testing and trialling, and
the contingency to be allowed; the descoping of Release 2 into New Release 2 and
‘New Release 2+; and which benefits are to be included in the start of national rollout;
. programme management artangements are still not satisfactory;
. insufficient strategic commitment to the delivery of the programme by all parties; and
. the absence of properly co-ordinated end-to-end planning, and the sheer difficulty of
co-ordinating the rollout of such an ambitious programme.
46." -: Given the disagreements between the partiés on timescales, ‘PA had to make a number of
- judgements about the duration of key activities. These judgements tended to be on the cautious’ side,
closer to BA's view, but not inconsistent with the history of slippage seen so far in the programme.
47, Based on this analysis, we conclude that the start of national rollout might be further delayed
_ by up to 9 months to January 2000. This assumes that all outstanding issues with New Release 2
and New Release 2+ are resolved quickly by the HPO (by end July) and a baseline agreed. If so the
forecast date might be brought forward. . 7
485 Once rollout starts, it is planned to continue ata ‘beat rate’ of 300 offices per week. There
are risks to the achievement of this rate, for example: :
+.” * there are no pauses for review, and.scaiing up beyond, say, 10,000 outlets may give
rise to problems; . ai ty yg .
ve the rollout of EPOSS, if it requires process changes at the counter, runs the risk of -
disruption to the benefit payment service; : .
. - BA card rollout sequence plans may need to change at short notice;
. parallel activity to upgrade parts of the Post Office estate may impose constraints on
rollout; and P
“3
POL00028094
POL00028094
CONFIDENTIAL - COMMERCIAL
BA/POCL Review - Independent Pane! report
. agreement between the parties about the criteria for rollout completion may be'hard
to achieve. :
49. These factors have been taken account of in PA’s analysis. Harder to model are the
demands rollout will place on POCL organisation, and the stress likely to be caused’as ‘its regionally *
based organisation tries to cope with a service management approach that demands a more
centralised organisation. With this caveat, we believe that the 300 beat rate has been properly
planned for by Pathway and should be sustainable, leading to a best estimate for completion of
rollout by September 2001.” :
50. In order to de-risk software development and improve the management of software releases,
* Pathway has committed resources both at a significantly higher level and with a different skills mix
than it had envisaged at the outset, requiring a total re-design of the project team structure and
processes. Pathway's plans are now much more robust, realistic and better resourced in numbers
and skills and, if the issues mentioned above can be settled “ there is every likelihood that
Pathway can deliver to timescales.
51. BA appears consistently to be able to plan 2 and deliver to timescales.
52. ‘There remains a great deal of work to be done by POCL to develop its programme and
service management capabilities. POCL's planning, for both. implementation and service
management of the programme, does not appear to have been given sufficiently high a priority within
the PO Group. It needs to be properly resourced in terms of numbers and skills. Planning should
provide for and manage an end-to-end critical path, including links and dependencies with its other
key systems. This planning must be set in the context of POCL's longer term business strategy. ~
E. Programme management, organisation nd structures
53. The programme is characterised by inefficient decision making processes. The way change
tequests are managed is a particular problem; Pathway are obliged to respond to change.requests
within 3 weeks and impact assessments have to be made in 5 days, but no time is stipulated for
decision-making by the sponsors. This imbalance is, in our view, harming progress.
54. The new programme management arrangements since April 1 are intended to improve this.
The Horizon programme office (HPO) has a vital role, but it is not yet operating as planned - for
example joint commercial teams are meeting while the technical and implementation boards are not
- although there are signs of its increasing effectiveness.
55. POCL and Pathway feel that they are collaborating more closely than before. This process of
“joining up” the parties must accelerate and embrace BA. Just as important are stronger links with
POCL's emerging change programmes. F
56. We understand there are no end-to-end descriptions of the system and process
architectures, which makes it difficult to plan and manage the programme. Without these, and a
master critical path activity network, alignment between the parties will always be difficult and
contentious, All parties should agree to an “open book" approach so that an end to end plan for the
programme can be arranged and driven through by the HPO.
57. _ The HPO must drive for resolution offal the Issueswe have identified as on the critical path
(see Part Ill below) before end July, if there is to be any chance of improving on the forecasts we
have made for br completion of the Horizon programme.
ye
POL00028094
POL00028094
CONFIDENTIAL - COMMERCIAL
= BA/POCL Review - Independent Pane! report
58. Pathway has demonstrated its capacity to deliver the programme. The.evidence suggests
that the primary focus for any leaming from Release 1c has been at Pathway. We found no evidence
of an explicit leaming process across the programme, although there is evidence that all parties
have benefited from their experiences of Ric. Pathway has processes (problem management,
service review, quality audits) to capture Ric experience and have changed Release 2 design and
_ the production service as a result.
59. While BA has demonstrated its ability to develop and rollout CAPS, itis more remote from
the programme management process. Its constructive engagement will be needed to bring the
programme to a successful conclusion.
60. As noted above, the HPO i is not yet working as was envisaged. There is evidence of under-
resourcing in terms of numbers of people, the necessary skills mix and funding. While BA and
. Pathway are buying in the resources and skills they need, POCL are using in-house staff where
> possible. If the HPO is to drive the programme as a whole, it is essential that key roles in it are
resourced to standards comparable with those being achieved across the rest of the Programme.
. 6. In our opinion, there are not enough high calibre staff in the HPO particularly in key positions a
such as programme office: only afew HPO staff have relevant experience in managing such a large
and complex programme. As a result, itis our view that the HPO director has been pulled in too:
many directions - contractual discussions, POCL business and so on: He so far has had insufficient
time to drive the Programme on a day to day basis and so resolve ‘the issues that are causing the
programme to falter. Although as this review has. progressed there have been positive signs of .
change, we believe that additional high quality resource is needed to support the HPO Programme
Director so that, for example, end to end visibility of all parties’ plans are available, challenged and
then integrated into a critical path network and master plan.
62. There is anecdotal evidence that suggests POCL is not making the most of opportunities
. afforded by Ric for improvements to its processes. We have not found any evidence of a process
that ensures sharing between the parties of experiences and lessons leamt. The HPO should
manage the learning process (from Ric and from testing and trial activities still to come) to ensure all °
parties leam appropriately.
F, Strategic Commitment
63. The HPO's ability to drive the parties to agree and to achieve programme milestones will
continue to be undermined in the absence of shared business objectives for the programme.
64. There is uncertainty about both sponsors’ long term intentions: BA about altematives to
cards for benefit payment; POCL about its strategy to exploit automation. There is a recognition at
POCL that successful implementation of the automation programme demands a significant change
in culture, however plans to effect this are not well advanced.
65. Over and above the resource constraints noted elsewhere, we believe that the HPO does
not have sufficient authority for the role it has been given, We see this in two key respects:
. the Horizon programme director does not manage directly the dependencies of
POCL projects e.g. Reference data, TIP, yet the overall programme is dependant on
_ these Projects delivering on time;
.* - the programme requires substantial support of the POCL business as a whole when
it comes to implementation, use and-support of the service. The HPO director is a
15
POL00028094
POL00028094
aye
yee oie
CONFIDENTIAL - COMMERCIAL,
: BAIPOCL Review- Independent Panel report
member of the POCL top team but we are not yet convinced that the. HPO is locked. °
into POCL’s organisation, management and decision making processes sufficiently
to ensure timely decision making on . programme issues.
66. With the proper level of senior management’ ‘commitment and delegated authority tothe -
Programme Director, he would be much better placed to drive through the programme with the
minimum need to refer upwards.
G. Likely costs of delivery .
67. Increases to programme costs are driven primarily by timescale slippage. BA’s costs include
those attributable to CAPS development. In addition, BA will lose potential fraud savings, POCL’s
opportunity to generate other non-BA business is delayed, and Pathway lose income from system
charges.
68. Based on our forecast of 9 months further slippage, current programme run rates provided :
by the parties suggest 'that direct costs will increase as.follows:
BA obit
POCL £15m
st Pathway: . £72m
69. For comparison, BA's forecast total costs on the programme are over £1bn and annual ~
payments to POCL ond Pathway will be around £500m in steady state.
70. .We have not verified the parties’ run rates. They may contain an element of negotiating
margin and may.not take full account of the possibility of reprogramming other work to mitigate the
impact of further slippage. However they seem to us: to be a reasonable guide to the costs of delay
for the purposes of this report. .
athwat
71. Pathway tells us it has spent or committed £250m so far and expects to spend a further
£250m, much of it already contracted for, to complete the design and build stages. Pathway's peak
cash requirement will be around £425m and financing costs will therefore be substantially higher
than budgeted for. On current planning assumptions, income from operations is estimated at £220m,
leaving a deficit of £205m. This takes account of a financing cost of £170m. The cost ‘run rate’ is
about £8m per month and steady state income lost for each month of delay is about £14m.
72, POCL has questioned the accuracy of Pathway's cost projections and the true extent of
deterioration in its rate of retumn. We have ‘not verified Pathway's figures, although it will be
necessary to do so for the purposes of any commercial negotiations. Meanwhile we believe they
should be accepted as the basis for a decision on the future of the programme.
BA
73. In 1996, the NPV to 2003/2004 of BA’s business case was £1bn, taking account of benefit
fraud savings but excluding CAPS costs and savings. When CAPS was taken into account (although
we question whether it is appropriate to do so), the NPV dropped to £730m. On current assumptions
POL00028094
POL00028094
—
CONFIDENTIAL - COMMERCIAL
BA/POCL Review - Independent Panel report
{ie rollout starting in April 1999), this latter figure has dropped to £469m. Currently, the cost 'run rate’
of continuing with the programme in its development state is about £10m per month (including CAPS
work) and, in steady operational state, the losses are about £15m per month based on the benefit
fraud and administrative savings lost through delay’ in migrating to the new system.
POCL
74, In 1996, the NPV to 2003/2004 of POCL's business case was £65m. This has dropped to
£46m on current planning assumptions and could drop further to about £22m depending upon the
resolution of outstanding issues and risks. It is estimated that the cost ‘run rate’ is about £14m per
month (excluding costs associated with Reference Data and TIP). Operating cost savings of asimilar
amount are also being delayed.
75. With the exception of POCL, the parties have now resourced up to.the numbers, skills and
experience required for a programme of this scale and complexity. There remain questions about
costs at POCL, partly the resourcing of the HPO bit also reflecting our view that POCL has not yet
worked:through‘all the implications of automation. Preparing for and implementing an automated
counter environment and its exploitation, which could involve major organisational change, could
mean a substantial un-budgeted cost.
76. We believe that POCL continues to take an over-cautious approach to resourcing this major
Programme, relative to the other parties. The POCL business case shows, for example, that the .
resource for national rollout is budgeted at £1000 per post office, that the HPO costs peak at just
~ under £1m per month and the Service Management Function is budgeted at £2.5m per annum. It is
understandable that POCL will want to keep implementation costs down. Indeed that presumably is
why the'PFI route was chosen in the first place. Having noted that, we believe that POCL may well
have to find additional resources to ensure the success of the Programme.
77. A second area of risk at POCL concems the state of its organisational readiness to accept
an automated network. POCL themselves note thatthis:programme ‘requires a significant culture _
change’. This:must be registered as a potentially costly risk to the programme until such time as.
POCL sets out detailed plans for implementing any organisation change for automation. These plans
must be able to demonstrate that the confirmed Pathway plans for national rollout can be supported
before this risk can safely be removed.
78, In'summary, it can be seen that the financial retum from the programme has deteriorated
significantly for all parties. Pathway, on the basis of the figures it has provided, would make an
overall loss if-the contract continued on its presént terms. For BA and POCL, the NPVs of the
programme, although positive, have been substantially eroded, andin POCL's case may have
become marginal. °
POL00028094
POL00028094
CONFIDENTIAL - COMMERCIAL,
BAIPOCL Review - Independent Panel report
_ PART Ill - RESIDUAL ISSUES AND RISKS
79. ~- The programme from the start seems to have had two distinct sponsor visions, with no clear
Goverment-wide view. There has been no clear single owner of the programme. Perhapsasa_
consequence of this, we have seen‘a lack of grip at project level (at least until early ‘98), leading to
poor conflict management and a lack of prioritisation. There is some evidence of this changing, but
the’change needs to accelerate.
80. Despite recent changes to the programme management arrangements, the supply chain
temains unclear. This might be addressed in contractual negotiations ifa sestructuring of the
programme were to be approved.
81. Ona short operating franchise PFI, with business cases sensitive to delay, specification of
sponsor requirements was initially only at a high level. The supplier was.prepared to enter into a
fixed price contract to deliver against them, but underestimated the programme complexity and the
Tesources required in light of the sponsors’ specification. Delays so far have meant that for Pathway
to generate a positive rate of retum is now heavily dependent on generating non- -BA business for
POCL.
82. The relationships between the parties can be characterised bya low jevel of confidence in
various aspects of the eventual solution and in each other.
83. The parties presented us with a number of outstanding issues for resolution. Some have
been referred to already. Many we see as ‘operational’, ones which we would have expected to have
been resolved by routine programme and change maiagement procedures. Others - identified below ,
as ‘strategic’ or ‘critical path’ - have more serious implications. The full list, with our assessment of
which party should be teking the lead in inresolving each one, is at annex D.
A. Strategic issues : ‘s #4
84. The following issues are fundamental to the programme, with severe risk of its failure if they
remain unresolved:
oe all parties need sound business reasons for a full strategic commitment to the
programme, with a common yision and incentives for all to make it work ina
teasonable timescale; original business cases are substantially eroded or no longer
viable, and there is no process for resolution (we will retum to this in Part V below);
oe BA's commitment is perhaps hardest to achieve while the lower unit cost of an
alternative payment mechanism - Automated Credit Transfer (ACT) for at least the
majority of its customers - appears so attractive; any lack of commitment on BA's
part could be highly corrosive; conversely a strategically committed BA could play a
major role in the success of the programme;
. the parties need to ensure the HPO is resourced and empowered to champion and
drive forward the programme; POCL must also convince the other parties of its
organisational capacity for rollout and service delivery, all within the context of a
clear, well-implemented future business strategy;
. Pathway must convince first POCL, then BA that Horizon can be exploited
. cost-effectively to support banking, financial services and other applications for
clients in and outside goverment, bringing economies of scale and reduced unit
costs for POCL clients;
. the lack of a single owner for the project is a major drawback.
I
POL00028094
POL00028094
-CONFIDENTIAL - COMMERCIAL
BA/POCL Review - Independent Panel report
B. Critical path issues
85, There is a further group of issues whose resolution is essential for the parties’ agreement
about programme milestone dates, especially the start of live trial and national rollout. Agreement is
needed quickly and should not wait for commercial and legal discussions to be complete. _
86. It Is.our view that, with goodwill from the parties and the right drive from the HPO, all critical
issues can be satisfactorily resolved; indeed resolution is close on some, for example Soft EVP. The
big issues of uncertainty relate to:
. the duration of Model Office Testing and Live Trial; current disagreement centres on
the amount of contingency needed;
“e the ability of each of the parties internal programmes to all come together atthe right
time to allow Live Trial to commence,
. the ability of the parties to agree acceptance criteria and process so that acceptance
is achieved in a timely way‘and,
. the risk that BA and POCL business process will need to change once the end-to-
end impact of the system has heen fully assessed. *
87. In most cases it is the HPO which should be era the resolution of these issues. It must,
for example, take early steps to:
. develop a master, baselined plan for delivery of contracted functionality, which takes
account of sponsors’ other work programmes and reaches across interfaces with
other systems, including POCL feeder systems;
. finalise NR2+ contents description, resolving outstanding agreements-to-agree;
. agree an achievable start date and duration for live trial;
. get agreement on how far national rollout can proceed before NR2+ is avattable; and
. speed up acceptance and change control processes, :
88.- -" BAcan do more to ensure that the other paities have confidence in: -
2” its plans to rollout CAPS; tas
oe * the schedule for migrating benefit types to the new system; and
. that CAPS personal data quality \ will be sufficient not to affect rollout adversely.
89, Pathway also has work to do to convince the other parties that:
. the essential nature of the security requirement has been taken seriously;
. service management products are fully available and scaleable;
. subcontractors will be properly managed; and
. there is a consistent, complete and scaleable technical design. -
a. Operational issues
90. There remain a further group of ‘operational ‘ issues, for example around:
19
POL00028094
POL00028094
yah ge
CONFIDENTIAL - COMMERCIAL
BA/POCL Review - Independent Panel report
. the effectiveness of BA data cleansing;
. * the use of contingency in the programme timetable;
. problems with EPOSS; ~~ °
. staff training and competency; and
. the proposed solution for small offices where full Horizon installation is difficult.
91. We would have expected these to have been resolved at working level. With goodwill and a
willingness to compromise on all sides, these should be solved once key milestones are set.
Howe’ ties will have to change. oe
92, We see considerable welght falling on the HPO, with the Horizon Programme Board used
- effectively to reconcile conflicts and ensure progress: Both will need real authority within POCL, with
a remit to drive forward progress on outstanding acceptance issues. They must demand tough
delivery schedules of Pathway.and command the confidence of POCL's biggest client (BA) to
concede authority over the programme to the HPO: ‘Relationships with other POCL clients will also
be important in developing new business,
93, In summary, if there is a commitment to the programme at the highest level (ie with the
strategic issues resolved), the main risk lies in the ability (or otherwise) of of the 3 parties to clear all the
critical path issues (listed below for convenience) by end July.
Critical Path Issues for resolution before end of July 1998
Interface management
Resolution of outstanding Agreements-to-Agree
Baselined pian for delivery of contracted functionality:
Finalisation of NR2+ contents description
NR2 acceptance for iv by Jan 99
NR2 limited to 4000 outlets
Sufficiency of 11 weeks for live trial
Change control processes
Need to take account of sponsors’ work programmes
Programme master plan
Speed of acceptance process
Management of Pathway subcontractors
Consistent and complete technical design
Service management products available and scaleable
“Card production
Scaleable design
Security requirement taken seriously
POCL estate readinéss__
Horizon Programme office viability
Issues with reference data
Large scale FAD codé changes
Scheduling benefit migration
Card rollout ~
. CAPS performance - - -
‘by HPO
D4
=z
by Pathway
D-
by POCL
a
Bt
byBA =
z
20
POL00028094
POL00028094
CONFIDENTIAL -'COMMERCIAL
BAPOCL Review - Independent Panel report
PART IV - SUMMARY OF FINDINGS
94, The BA/POCL programme is complex, probably the biggest of its kind. Its scale, and
particularly the development work required, were underestimated initially. The parties have since
increased the resources devoted to the programme, but a range of issues remain to be resolved.
95, Our view is that the programme is technically viable. There must be some tisk around
scaleability and robustness because the system has had to be tested at the level of component
parts, but we are satisfied these risks are being well managed by Pathway.
96. There is good evidence of future proofing at all levels. The basic infrastructure is very robust .
for the future and, in the main, industry standard products have been used. The system should allow
POCL to compete for new business in a variety of markets, including banking and financial services.
New applications based on-smartcard technology should be relatively straightforward and economic.
If on-line applications are required, they may take longer and require more investment.
~ 97. - A further nine, months delay. is our best forecast, with September 2001 for national rollout
” completion. Critical path issues will have to be resolved fast to make this possible, but the date could
be brought forward with commitment and goodwill on all sides.
98. The new Horizon Programme Office has a vital role. It needs to be given the power andthe .
resources to drive forward the whole programme end-to-end and to resolve critical outstanding (and
future) issues.
99. Driven mainly by timetable slippage, the sponsors’ business cases are eroding. The direct
cost of delay is estimated at £180m, over half of which falls to the sponsors. Pathway, on the basis of
the figures it has provided, would make an overall{Gs$ if the contract continued on its present terms
and would require an extension to break even." -”
2 I
POL00028094
POL00028094
CONFIDENTIAL - COMMERCIAL
BAPOCL Review - Independent Panel report
ANNEX A - THE PANEL’S VIEW OF A POSSIBLE WAY FORWARD
1. Given the findings in the body of the report, we sought to find a way forward. on which all the
parties might agree in principle, subject to negotiations about the detail. We considered all the
options: .
. Restructured full programme
. Restructured partial programme - with no benefit payment card
Extend the current programme
. Continue the programme as planned
. Partial termination - no benefit payment card “bee
. Complete termination
PRpONs>
2. Taking the options in reverse order, each of the last four has fatal flaws if an’agreed way
forward is sought - one or more of the parties would be unable to accept it
. termination of the complete programme would leave POCL's automation plans set
back for at least two years (and very possibly more after potential litigation, re-
advertisement, procurement and implementation), with the likely loss of non
Government business in the meantime; Pathway would face a significant loss of
prospects, reputation and revenue;
° .. partial termination with no restructuring.- ~ to scrap the benefit payment card and
reduce the programme to POCL automation plus OBCS - would unacceptably
reduce Pathway’s revenue stream and leave POCL with an infrastructure too highly
specified for its short term needs;
. continuing the programme as currently planned would leave Pathway below break-
even on its investment and POCL with its infrastructure incomplete for a. move into
banking and financial services; the introduction of the card for the short period
remaining of the initial contract term would expose customers to disruption if BA
moved to full ACT immediately thereafter,
. .a simple extension would delay BA’s move to increased use of ACT, protonging its
exposure to high unit cost of benefit payments; POCL would have little incentive to
modemise further.
3. We took the opportunity, when presenting our emerging findings to the parties, to set out
Options 4 and 2 as the most likely to provide an agfeed way forward. We invited the parties to
respond indicating whether either might be acceptable. POCL and Pathway supported Option 1, BA
preferred Option 2. We offer our assessment of these two options below.
’ Option 2: Partial Restructuring —
4, Under this option, paper based payments would continue until POCL were ready to compete
for ACT payments. BA would go to open market for ACT-fed payments after a lock-in period to
assure income stability for POCL to prepare for the transition. The elements would be:
* no benefit payment card system, but BA continuing its present arrangements with
POCL until it had a capacity (technical and commercial) to compete for ACT
delivery;
= OBCS implemented in key (high fraud) areas;
oe automation of POCL infrastructure continues but migrates to support banking and
financial services applications, usi
compete in new markets;
smart card technology, with freedom to
fiex
POL00028094
POL00028094
CONFIDENTIAL - COMMERCIAL
BA/POCL Review’ Independent Panel report
_5._° Advantages of this approach would be: ‘
* POCL would get an automated infrastructure, capable of supporting banking,
Pathway compensated for reduced scope of contract.
financial services, etc
BAwould have an anti-fraud tool (oBcs), a anda route to reengineering its payment
processes to ACT; .
the programme would be descoped, with therefore lower implementation risk; and
Pathway would be well placed to compete for govemment.direct business.
. & On the other hand its disadvantages would be:
the perpetuation of a high cost paper-based method of payment until POCL were
ready to compete in the banking market,
POCL's infrastructure would be too highly specified for its needs, and yet the lack of
an atitomated payment component would handicap (so it contends) POCL's efforts
to win alternative non-Government I business;
delays to BA migration to automated payments and to robust BA accounting and
reconciliation systems (although ai believe OBtS might be adapted to provide the
_ hecessary information);
lengthy, costly negotiation belvesn the parties, with the ating possibility that
*- Pathway would be unwilling to de-scope in this way;
ICL would lose the substantial part of a prestigious project, suffer damage to its
-Feputation, and could pursue claims for substantial compensation; and
the failure to capture the full extent of the fraud savings expected from the complete
system, the burden of any compensation payable to Pathway and POCL’s inability to
sustain any significant reduction in the payments it receives from BA combine to .
make the option uncertain value for money. :
Option 1: Full restructuring
fe Under this option, the benefit payment card would'go ahead, with card based payments until
POCL were ready to Compete for ACT payments. BA. would then go to open market for ACT-f “A
payments. The elements would be:
card based payments from the start of national rollout; order books and giros
withdrawn completely by the end of rollout; BA to compete its payment business
. once POCL had the capacity to compete (technically and commercially) for ACT
delivery;
” POCL infrastructure migrates to support banking, financial services and other
applications with smart card, with freedom to compete in new markets;
planned BA transfer-to full ACT ona schedule beyond the current contract end date,
+ -to be agreed in negotiation, by which tine POCL would have re-engineered its
business to compete electronically and Pathway would have come closer to
recouping its investment;
card (magnetic strip and potentially ‘Smartcard) based and-ACT payments coexisting
_ throughout.
POL00028094
POL00028094
CONFIDENTIAL - COMMERCIAL
BA/POCL Review - Independent Panel report
8. Restructuring the programme in this way would have a number of advantages:
. benefits would be automated, with administrative savings for BA and payment fraud
eliminated;
* .” benefit recipients would be educated in the use of card-based technologies; .
. BA would see a clear route to ACT and an end to POCL's ‘monopoly’; i
. POCL would gain an automated infrastructure, with incentives to compete for.
business (including from BA) and a platform for social banking, financial and
government.direct applications, but with time to re-engineer IisI business as
necessary;
+ Pathway would be well placed to compete for new business;
. 3 there would be little likelihood of litigation.
9. There would, however,-be disadvantages and issues to be resolved:
. = > BAWwould have to accept some delay. to ACT migration, but this should be similar to
the delay under Option 2;
. the residual issues and risks with the Programme would have to be resolved
urgently; ..
+ ___ the parties would have to satisfy themselves they could pursue this option without
» fear of legal challenge.
Weighing the options
10. In both Options 1 and 2:
. the maintenance of an efficient natiénal network of Post Offices is achieved;
. “the provision by Government of the necessary commercial freedoms to POCL to
compete in banking and financial services is a pre-requisite;
+ "the timescale for BA’s move towards ACT-fed payments is similar,
+” there are incentives for POCL to re-engineer its business quickly and in good order,
“ "(but with time to do so); and
@% “ overall cost to the public purse, while difficult to assess sat this stage, should be
* “similar.
41. The potential for litigation and programme blight is greater under Option 2. The attitude of
the parties suggests there is the almost certain prospect of litigation, prompted either by BA (on
grounds of breach of contract through delays by Pathway) or by Pathway (on grounds of termination
by BA for convenience through de-scoping the requirement) or by POCL (against BA), or possibly by
the other contractors who were unsuccessful in the initial bidding. In any event, long, acrimonious
and costly legal disputes would be likely, during which Progress on any repiprament programme
would be blighted. ! Fs
12. Under Option 1, the prospect of serious dispute (though not of difficult negotiations) is
virtually eliminated and the investment already made jin time, effort and money on the payment card
is not wasted. There would be no call for compensation by either of the parties and Pathway would
have a reasonable (although not risk free) opportunity to recoup its investment. The contracts
between the parties would, however, need to be restructured and certain legal issues resolved.
24
POL00028094
POL00028094
CONFIDENTIAL - COMMERCIAL
BAIPOCL Review - Independent Pane! report
13. A major influence in the choice between the two options is the speed of migration to ACT
payments and the feasibility of enforcing 100% adoption of ACT. Under either approach, migration to
ACT-fed payments will take several years, requiring two systems to co-exist. The benefit payment
card Is a more cost effective basis for non-ACT payments than the existing order book system.
Against this, BA argue that, if a move to 100% ACT-fed payments is accepted as an immediate
objective, it would be disorientating and inefficient for benefit recipients to have to familiarise
themselves with the use of a benefit payment card which would be in use for only’a short while until
100% ACT became effective. But, unless an element of compulsion is provided, there will be a
continuing need in the medium term for a non-ACT system to provide payment for customers who
are unable to have.a bank account or are simply unwilling to do so, preferring payment in cash. So
unless Ministers are. prepared to compel BA's customers to open bank accounts (and to subsidise
the costs of doing so for those customers for whom banks would otherwise be unwilling to provide
banking facilities), Option 1 will represent a more efficient way of providing cash payment facilities in
the medium term than Option 2. : mo
14, Option 1 provides a platform for the Goverment to make an early start on its social banking
and electronic access policies, and provides much needed customer education in the use of card
technology. . .
15, ‘There are serious risks to POCL's business under Option 2. The partially restructured
Programme should carry less risk of further delay then the continuation of the full programme, since
most of the complexity is associated with the payment card system. However, the absence of an
automated payment system could leave POCL exposed commercially both in maintaining existing
clients and in attracting new clients from the banking and financial services sector. It will be in the
interests of POCL and Pathway to de-risk the continuation of the full programme under Option 1.
iy) ;
Financial Implications . . -
16. Both restructuring options would require the preparation of new business cases and realistic
Programmes. The financial terms of any restructuring should recognise:
. BA's need to establish a fixed timescale for the progressive introduction of ACT
payments; .
. POCL’s inability to sustain a significant reduction in payments from BA until itis ina
Position to compete with other banking service providers for delivery of ACT-fed
payment services to BA customers; and
. ICL's concem t6 secure a reasonabl Tetum (though not a guarantee, as Pathway
must share with POCL the risks of diversifying the commercialisation of the system).
17. The variables would therefore be: . . :
. the length of the plateau until competition for ACT delivery, and the speed of the . ~
. transition; *
. the level of POCL’s guaranteed income; F
. the balance between (i) contract extension to remunerate Pathway's investment, (ii)
the rate of retum fairly allowable to Pathway and (iii) the split of retums fro1
commercialisation of the system between POCL and Pathway; soo
. ~ POCL's financial prospects after transition;
. review of the risk profile at completion of national rollout.
25
POL00028094
POL00028094
CONFIDENTIAL - COMMERCIAL
BAPOCL. Review. Independent Panel report
18. Any restructuring of the programme must be based on a realistic reappraisal of dates for the
start and completion of national rollout. As noted above, we believe a delay of up to 9 months on the
present scheduled date for the start of national rollout is likely. We understand that, for contractual -
reasons, the parties may not be prepared formally to acknowledge the likelihood of further delay at
present for fear of influencing decisions regarding the future of the programme. But it would be a
serious mistake to embark on a restructuring process without an end-to-end programme for the
commissioning of the entire system, agreed by all parties. At the same time the restructuring plan
must allocate the costs of further delay between the parties and establish fresh incentives to
motivate each to seek to minimise any further slippage and achieve the earliest practicable
completion dates. >”
19. It would be for the parties in negotiation to agree the details. The way forward should seek to
reconcile key business drivers of all stakeholders. Factors in the decision would be which option best
met the priorities of the Post Office review and promoted the social benefits of the network, and how
much strategic factors weighed in a value for money ; assessment.
20. We suggest that,as a basis fora decision between Option 1 and Option 2, the parties should.
address the detailed cost implications of each, and which option will allow POCL to compete most
effectively and expeditiously for ACT-fed benefit it payments in competition with other banking service
providers.
: Conclusion
21. Option 2 (partial restructuring) is, in our view, attractive only in relation to the lower risks it
might carry by comparison with Option 1. It has the major disadvantage of resulting almost certainly
in prolonged contractual dispute, possibly leaving POCL without any automated network until the.
legal position is resolved and a fresh procurement completed.
22, Option 1 (full restructuring) may offer the prospect of better value for money in the longer
term although, to achieve this, the programme I will need to be driven much harder and more
professionally than so far. This will entail a major change in the status of POCL and a major shift in
its management and customer service culture from mor poly supplier to an efficient competitor in an
open market for banking and financial services. *
23, The parties have not agreed on either option as a way forward.
Implementation of a restructuring
24. If Ministers decide to adopt either option, we would like to see a neutral ‘troubleshooter’,,
directly responsible to Ministers and with their full confidence, appointed to ensure the parties give
their full commitment to the implementation process and that the rollout is.completed at the earliest
agreed date. The first priority would be a restructuring plan, agreed with the parties, containing:
- an action plan to remedy the risks and deficiencies identified in this report; with
. incentives for. prompt delivery and wider commercialisation of the programme; and
. open book resolution of outstanding commercial issues.
25. The troubleshooter would have a challenging réle. For any restructuring to succeed, it will be
necessary for all parties to bury their differences and to build a new commitment to see the
programme to a successful conclusion. The troubleshooter’s contribution would be essential in
POL00028094
POL00028094
CONFIDENTIAL - COMMERCIAL
: BA/POCL Review - Independent Panel report
setting the tone for the continuing relationships between the parties. The troubleshooter could have a
longer term role if Ministers wanted some external monitoring of progress against the agreed plan.
26, Adoption of the programme as a platform for the development of banking and financial
services applications will require POCL, as it recognises, to effect a major cultural and organisational
change to prepare itself to compete in the wider commercial world. The Post Office Review being
conducted by the DTI should consider what refinements to POCL's present status: and management
structure would La desirable to allow it to implement the changes required.
Contractualfegal issues
27. Early indications are that Option 1 would run fewer risks of challenge under the EC
procurement directives, by the Commission or by other bidders, than Option 2. However the parties
will need to assess this question in light of the details of the original competition.
28. Either of the options will lead to modification ‘of the programme contracts. In our view,
although the scale of the work involved will be substantial, the opportunity st should be taken to
> ‘restructure is contractual relationships to:
. * establish POCL as the clear owner of. the programme;
: clarify its status as a supplier to BA; and ©
. reflect the tighter programme management arrangements suggested in this feport.
Bed
27
POL00028094
POL00028094
CONFIDENTIAL - COMMERCIAL
BAPPOCL Review - Independent Panel report
ANNEX B - THE INDEPENDENT PANEL’S RULES OF PROCEDURE
1.
441
1.2
1.3
Introduction
Difficulties have arisen with the project to automate payment of social security benefits
across Post Office Counters, for which the Benefits Agency (BA) and Post Office Counters
Limited (POCL) have contracts (the “Contracts")with ICL Pathway (ICL). BA, POCL and ICL
-. are referred to collectively in these rules as “the Parties”. In order to inform decision-making
in relation to these difficulties it has been decided to prepare an urgent assessment of -
whether the project is technically viable, and if so how quickly it can be completed and at
“what cost.
To ensure that this assessment is based on a thorough and impartial review it is proposed to
establish a panel chaired by the head of the Treasury Task Force on Private Finance, Adrian
Montague. The Treasury Taskforce was set up as a result of the recommendations‘of the
Bates Report on the Private Finance Initiative (PFI). One ofits functions is to assist in PFI
projects facing difficulties. Its responsibilities include monitoring PFI Projects to ensure
progress is being made in accordance with agreed timetables.
The Parties have agreed with the Panel and each other to participate i in the proceedings of
the Panel in accordance with these rules (the Proceedings’).
Terms of reference
1.4
The Panel's terms of reference will be to con ict the Proceedings (which will be without
prejudice to the Parties’ legal rights) with the intention of preparing a confidential report to consider:
1.5
‘The focus of the Panel's review will therefore:
- whether the project can deliver a fully functioning system which meets the project
specification, and integrates fully with BA computer systems; this will include
examination of BA’s and POCL’s systems and arrangements for roll out;
- whether the timetable for completing the systems development, and starting and
* completing rollout, is deliverable; and whether the necessary managerial and
organisational structures are in place;
- the likely costs of delivery, under current contract dates and with extension;
- in each of these areas, the risks associated with these assessments, and whether
robust monitoring arrangements and disciplines are in place.
‘on the deliverability of the currently defined
project, and the risks associated with.estimates of timescale and costs. It is not concemed
: with allocating blame for delays experienced so far, or with negotiations as to whom might
bear any additional costs.
Practical arrangements
1.6 °
The composition of the Panel is:
Adrian Montague - - Chairman
“Alec Wylie
Bill Robins
28
POL00028094
POL00028094
CONFIDENTIAL: - COMMERCIAL —
BA/POCL Review - Independent Pane! report
1.7
1.8
1.9
1.10
1.12
2.
24
2.2
23
2.4
- failure.
The Treasury may, at its discretion, make changes in the composition of the Panel.
,
The Panel will be assisted by a Secretariat provided by the Treasury.
‘The Treasury has appointed PA Consulting'to Act as Consultants to the Panel.
The addresses, phone and fax numbers and email addresses for HM Treasury, the .
Secretariat, BA, POCL and ICL are as set out in the attached Schedule 1.
Meetings between the Panel and the Parties will take place at HM Treasury unless otherwise
specified by the Panel. . . -
The members of the Panel (other than the Chairman) and the Consultants will be retained
and paid by CCTA. The other costs of the Panel will be met by HM Treasury. Each of the
_ Parties will bear their own costs.
The Panel may terminate the Proceedings at any time if it considers that there is no useful
purpose to be served in Continuing. .
Procedure and Timetable sides
The procedure is set out below. A timetable illustrating the possible stages of the
Proceedings is set out in Schedule 2. The procedure and timing of the stages may be varied
by the Panel as it thinks fit subject to the points about timetable in2.6.
The Parties are invited to submit to the Panel and to exchange with each other by the date
set out in paragraph 2 of Schedule 2 a document not exceeding 10 typed A4 pages setting
out: ; sos
2.2.1 . a summary of their view of any issues they believe fo be outstanding arising out of
the issues referred to in the Panel's terms of reference, including amounts at issue;
2.2.2 the proposed resolution of these issues; and
2.2.3 alist of any documents which have an important and direct bearing on the issues. .
outlined inthe summary. * em .
The first full meeting of the Panel will be attendéd by all the Parties and the Consultants. The
Parties will not be legally represented (either by external or employed lawyers) at this
meeting or at any subsequent meetings. At the first meeting the Panel will invite the Parties
to make short presentations to the Panel about the outstanding issues and how those issues
can be resolved. The Panel will, in its absolute discretion, determine the order in which the
presentations are to be given, the number of presentations and the time limits for the
presentations.
Following the first meeting the Panel will establish a list of issues to be investigated. It will
invite further submissions from the Parties and will instruct the Consultants to investigate the
matters in dispute. The Consultants will ask the Parties to provide any information the
Consultants think could be useful to the investigation. The Parties shall at all times give such
assistance as may reasonably be requested by. the Consultants to enable the investigation to
be completed. The Consultants will report to.thé Panel any failure by-any of the Parties to co-
operate in the investigation and the Panel will be free to draw adverse inferences from that
29
POL00028094
POL00028094
CONFIDENTIAL - COMMERCIAL
BAPOCL Review - Independent Panel report
2.5
26
27
2.8
29
2.10
The Chairman of the Panel may deal with procedural questions without consulting the other
two members of the Panel where he is unable to contact them and an urgent decision is
required. In all other matters decisions will be made either unanimously or by majority vote,
at the Chairman's discretion. Where there is no clear majority, the Chairman's decision
‘prevails.
The second full meeting of the Panel will be attended by all the Parties and the Consultants.
At the second full meeting between the Panel and the Parties, the Panel will invite the
- Parties, on the same basis as in 2.3 above, to respond briefly to the Panel's draft findings.
Following the second meeting with the Parties, the Panel will finalise the findings and submit
its report to the interdepartmental working group. A strict time limit of two months from the
date of the Panel's initial letter to the Parties is set for the Panel to submit its final report. The
Panel will actively seek ways of shortening this timetable.
” The Panel may convene further meetings with ‘the Parties, together or. separately, as it thinks
fit. The Panel will fix the date, time and place of those meetings.
No member ofthe Panel, or the Consultants, shall be liable to the Parties (or any of them) for
any act or omission whatsoever in connection with the Proceedings, nor shall they be under
any obligation to make any statement to any person about the Proceedings, nor shall any
Party seek to make them a witness in any legal proceedings arising out of or in connection
with the contracts. ‘
The Proceedings are private, confidential, privileged and informal. All materials, documents
or submissions, whether in writing or oral, made or created for the purpose of the .
Proceedings by the Panel, the Consultants or the Parties (the “Materials*) shall be
confidential and without prejudice to the Parties’ legal rights. The Materials shall not be
subject to disclosure to or discovery by any Party in any legal proceedings arising out of or in
connection with the Contracts or otherwise without the consent of the other Parties.
The report of the Panel shall not be subjec' closure to or discovery by any Party in any
legal proceedings arising out of or in connécticn with the Contracts or otherwise without the
consent of the other Parties. ICL will not be shown those parts that deal with BA’s and
POCL's performance. The chairman of the Panel will debrief the Parties on the Panel's
findings once the Panel's report is complete! but will not comment on or disclose to ICL the
parts relating to BA’s and POCL’s performance.
se.
30”
POL00028094
POL00028094
CONFIDENTIAL - COMMERCIAL . *
: BA/POCL Review Independent Panel repot
ANNEX C.- THE PANEL’S MEETINGS WITH THE PARTIES
The Panel held meetings with the parties, both together and separately, as follows:
11 May Initial shared presentation of issues by parties : All parties
11 May Private pissentaions of issues ey vert : ICL, BA, POCL separately
19 May Presentations of parties’ biidness cases ICL, BA, POCL suche
28 May. Demonstration to Panel of Horizon products , Pathway
8 June 7 Final presentations by parties we - __-ICL, BA, POCL separately
11 June Presentations to parties of Panel's aa findings ICL, BA, ack a
au.
POL00028094
POL00028094
CONFIDENTIAL - COMMERCIAL
BAPOCL Review - Independent Panel report
ANNEX D - OUTSTANDING ISSUES
The following table lists issues to be resolved and, for each one, an assessment of its classification -
strategic (S), critical path (C) or operational (O) - and the party with lead responsibility for resolving it
Classification I Responsibility
Issue - (S, C or O) - for resolution
[. Requirements .
Finalisation of NR2+ contents description c HPO
Resolution of outstanding A2As c HPO
Robust service management organisation Ss POCL
Security requirement taken seriously c Pathway
Interface management c HPO
Position of non-economic offices oO Pathway .
Il. Performance and viability ‘
Consistent and complete technical design . -. - c Pathway.
Baselined plan for delivery of contracted c HPO.
functionality =
Issues with EPOSS ie) * POCL
Issues with reference data Cc POCL
NR2 limited to 4000 outlets .c HPO.
Large scale FAD code changes Cc POCL .
Scaleable design c *. Pathway
Future proof design s Pathway
Service management products available and_ c Pathway
scaleable . °
Ric validation - are lessons being captured?. . fe) HPO
NR2 acceptance for live.trial by Jan 99 Cc HPO
Card rollout and production . c Pathway , BA
CAPS performance c BA
Management of Pathway subcontractors Cc Pathway
Ill. Timescale _
Reasons for previous (and future?) slippage fe} HPO
300 beat rate Ss HPO
Change control processes c HPO
Oct-Dec 98 contingency fe) HPO
Sufficiency of 11 weeks for live trial ~ c HPO
Speed of acceptance process at Cc HPO
32
POL00028094
POL00028094
CONFIDENTIAL - COMMERCIAL
BAIPOCL Review - Independent Panel report
Classification Responsibility
Issue (S, C or O), for resolution
Tw. Costs, resources
Business cases s all
Training of POCL staff (incl. tumover training) (om POCL.
Training competency check ° POCL
Need to take account of sponsors’ work c HPO
programmes a
Effectiveness of data cleansing fe} BA
POCLestate readiness ~ Cc POCL
POCL organisational readiness s POCL
Scheduling benefit migration c BA
V. Management, etc
Horizon Programme office viability c POCL
* Need.for business champions - ownership.not clear Ss” POCL
Escalation process a Te ize) HPO-
Horizon Project Board to meet? - fe) HPO
Adversarial approach Lo} all
Programme master plan c HPO
Better use of contingency. (time) fe} HPO
s all
. Evidence of strategic commitment
33