POL00090015
POL00090015
~y
@ PA REPORT - FIRST DRAFT
A GAPS
. Any obvious “golden rivet” solutions? > no mention of the “one key thing...”
e No mention at all of CAPS and its part to date (glaring!)
e¢ No mention of TMT/TSG/Horizon links within POCL (ie our own automation.
programme planning)
¢ No mention of common workload models for volume linked - ICL guarantees -
unless BA are using different volumes?
e Limited analysis of policy context for the programme in the first place
(underplayed) - hence strange comments about partial network automation and the
economic value of some post offices
B CLARIFICATIONS
What’s your definition of “end roll out” please?
BA’s low confidence in POCL to deliver (+ DSS “at the top”) - what’s the source?
What about role of PDA Board, if any, given PSC recommendation?
What do you mean by strengthening “key management roles” (too vague) - advice
to whom specifically?
e ‘What does separation of ICL’s and PDA’s delivery/contract/business development
management” mean?
¢ — Why didn’t you take up our offer to see more/other POCL people? (eg Dick/Dave
Morphey/Lesley Lawson - Dave Smith?)
. What does “explore alternative boundaries” mean in recommending a single
business case?
. Be more explicit about “no shared view of root causes” statement?
¢ — Where’s the £100m” extra costs to BA number sourced?
C CHALLENGES/AREAS FOR DISCUSSION
. PSC taking a hands on executive role - is this realistic?
. “A programme that has been out of control in the past” - whose programme? is this
wholly true?
e Do you really mean “common business case!” - or is it more towards the idea of a
common set of quantified aims and budgets cross-programme?
e Assumes POCL’s business case is basically OK - not true (underplays POCL
massive profit strain and linked points to our competitive position and our own
automation programme - did PA see complete set of right POCL people?)
. Says POCL is a monopoly supplier - not even “preferred supplier” to BA except in
a retail context!
. Finance reports are taken regularly by PDA Board > + CASG Reports within the
Post Office also monitor finance position regularly
e Incomplete view of extent of “longer term implementation planning” already in
place?
© Weren’t “agreements to agree” resolutions handled in fact by separate resource
from contract delivery within ICL and PDA - how distracting really was this?
PAUL RICH (Mena’s input)
5/9/97
POL00090015
POL00090015
IN STRICTEST CONFIDENCE
COUNTER AUTOMATION STEERING GROUP: FILE NOTE
Horizon: ICL/Post Office Meeting
5 September 1997
148 Old Street
Attendees
Post Office - John Roberts, Jerry Cope, Richard Close, Stuart Sweetman,
Paul Rich
ICL - Keith Todd, John Bennett, Mike Coombs
Copy: Scott Childes, Mena Rego
Key Points Made
1. John Roberts set out the purpose (per his letter to Keith Todd), and
emphasised he needed:
* aclear picture from ICL on how they think the programme is going,
and their confidence in solving issues around delay; and
* assurance about the programme for the PO Board.
2. Keith Todd gave his overview:
ae
bye
53.27
Horizon is critically important to ICL and Fujitsu, and to Post Office,
and (hoped) it is for DSS;
his belief that the programme is do-able, and that ICL will commit
all necessary funds to deliver its part;
his view that his own macro-objectives in setting out on the
programming had not changed, ie
~ to put in a UK national infrastructure via post offices that could
be developed long-term for society as part of a “national information
flow’;
~ to take the first serious steps, through using cards for DSS fraud
control, to take the “information society” to the technology-resistant
“mass market” in the country and then build on that (eg via
smartcards);
he has taken a calculated gamble that delivering Pathway’s
programme will enable ICL to become the PO’s main technology
partner;
that Horizon remains the best practicable option for DSS;
that Horizon is a world-class system (evidence of many postal
administrations wanting it).
That there were short term practical issues around Releases 1c and 2,
which meant a three week slippage to the former, and at least a three
month slippage to the latter. The latter still needs another 4-6 weeks
for Pathway to validate and re-baseline Release 2.
—y
POL00090015
POL00090015
4.
5.
58.27
IN STRICTEST CONFIDENCE
* that a key lesson learnt was that the joint sponsor contractual
relationship had made delivery much more complex, and that ICL
had been over-ambitious at the outset;
* that the DSS attitude was less than helpful in some respects.
John Bennett and Mike Coombs gave a short presentation (copies
attached for Scott) in which they highlighted:
* anumber of contributory factors for delay, which in their view were
not solely Pathway’s fault;
* in particular, the need to get an adequate and agreed requirements
baseline fully in place; the fact that the overall system was more
complex than first envisaged; and the need to understand
dependencies across the programme to get a common critical path and
a better change control process in place;
* the complexities around level of requirements because of PFI;
* Pathway’s need to improve their design, testing and integration
approach;
* some commercial ideas to help POCL longer term.
ohn Roberts invited comments from Post Office colleagues. These were:
* not accepting all of the contributory factors as described;
* acknowledgement that some lessons had been learnt, and plans were in
place to address them;
* that sponsors, via the PDA, had for some time been wanting a more
realistic timetable approach from ICL;
* that confidence levels were still low;
* that any new commercial ideas would be picked up by POCL with
Pathway;
* that some POCL scenario planning had begun as a matter of prudence;
* that POCL remained committed to providing all the resources
necessary to enable these plans to be made.
hn Roberts summarised as follows:
® the Post Office was very disappointed at the current situation;
* the Post Office, and POCL, is still keen to do this project and make it
work;
* ICL’s credibility is at stake, and confidence in Pathway within the Post
Office is still uncertain;
* arealistic baselined plan must be settled, and then frozen, with issues
or changes properly identified and processed after that;
* the open working between all parties must continue;
* that the Post Office could not guarantee its position to sustain
commitment if a realistic plan together with evidence of delivery is not
forthcoming from ICL in the next, say, six months;
¢ that the Post Office was willing to listen to ICL if it felt DSS was
unfairly inhibiting progress in some way.
POL00090015
POL00090015
IN STRICTEST CONFIDENCE
Keith Todd replied by saying he appreciated the spirit in which the
meeting had been conducted, and reaffirmed ICL’s commitment to make
it work. The meeting ended.
Following ICL personnel’s departure, Post Office colleagues discussed the
approach to the PO Board. It was agreed that a short update would be
given as part of the Chief Executive’s Report at the September Board, and
a fuller presentation by POCL would be given on the situation at the
November Board, once the independent PA review was fully known,
DSS's reaction to it was established, and Pathway’s baseline planning
work had been completed. This presentation would also include a
summary of POCL’s strategic scenario planning, being done as a
contingency if Horizon was ceased or radically altered in its scope or
delivery.
Paul Rich
5 September 1997
53.27
POL00090015
POL00090015
aye
Document Name: Document Version Number: I Document Date:
PA review Version 1 5/9/97
Reviewer Name: John Cook
Note that these comments have been complied on rapid read through - a more considered
review will follow.
No. Para Comment . Type
1. I Last para I The term “chronic” seems emotive - and while the full Style
page 1-1 I functionality national roll-out has been delayed there appears I and
to be no recognition of the (limited) successful phased roll-out. scope
2. I First para I The time dependant business case issue is developed further Justify
page 3-3 I later, but I do not feel that the statement here is fully explained
within the report.
3. Page 5-7, I It is not made clear why they believe that the contractual Justify
2nd complexity impedes effective delivery management - nor what
section they perceive as the cause or alternatives to the complexity.
Sth bullet
4. I Page 5-8, I To what extent are we putting all our eggs into the Pathway Key
2nd para I basket in any case. To wait until we can do “large scale real point
time testing” leaves a real risk if there are fundamental
architectural problems
5. I Page 5-10 I We would assert that the delay was in obtaining adequate Style
2nd para I definitions of the security solutions to our requirement - I note
that the following para affirms that our requirements are not
unreasonable.
6. I Page 5-11 I While I would accept that tensions exist on contractual Justify
2nd last I responsibility, the report does not acknowledge that such
para tensions are based on PFI and cost ownership.
7. I5-143rd_ I The PFI contract was not “rushed through”. There are many Error
bullet agreements to agree, but this is inevitable in a programme of
this complexity. There has to be a trade off between
, ‘I negotiating every last detail before award (with more than one
supplier) and moving forward with a single supplier.
8. I 5-14 last I The contract may come into play frequently, but the assertion I Style
bullet that it is too frequent is not supported by other statements in
the report that there needs to be a wider appreciation of the
contract.
9. {5,15 last I There have been several meetings of a joint contro! forum. Error
para This may not fully reflect the contracted procedure, but the
documented PDA procedure of an intemal CCB and a joint
CCB are followed and the groups have met a number of times.
10. I 5.19, 5-20 I This section requires more careful Tesponse, but contains Error
factual errors.
» There are not 180 change requests which have not been
formally signed off. There are approximately 21 CCNs in
CADATAWWORDIREVIEWS\PA_VERI.DOC page 1 of 2
saved: OB/09197 12:57; printed: 08/09/97 13:00
2O/TB"*d
POL00090015
POL00090015
”~,. - I
ek ZB"EOUd TWLOL 6K
process - out of a total of around 200 that have been raised.
© Agreements to agree were not the method used to “drop
down”. The “drop down” was based on agreed requirements
and solutions without varying obligation. In parallel a
number of agreements to agree were progressed and became
tied to the process as a lever - they were outside the drop
down process however.
Ti. I 5-20 last I Is the implication that there has been too much “legal Justify
para intervention” to date?
72. I 5-22 2nd I It is not clear what PA mean by a baseline reference of the Clarify
paralast I PDA’s own, especially given the fact that this is in the section
line on business case, Are they suggesting the PDA should have
greater empowerment to trade off between sponsors?
13. I 5-25 3rd_I Is this a criticism of PFI as a delivery mechanism, or the way I Clarify
para in which PFI has been applied to BA/POCL?
14. I Page 6-26 I Chart is difficult to read in fax quality, but seems to indicate Scope
roll-out starting (now) in 1999. I assume this does not take
account of early implementations (to a maximum of 5% of
offices)
15. I Page 6-28 I Not clear whether this comment relates to Pathway, or PDA Clarify
1st bullet _I plans or both.
16. I Page 6-28 I This is a critical point - POCL require the services in all Key
last bullet I outlets. If there is a real risk of anything up to 50% of offices I point
not being suitable it will destroy the viability of ‘the whole
strategy of both BA and POCL.
17. I Page 9-31 I In the centre of the page there is the reference to sorting out the I Clarify
fit between security requirements and solutions. This area
recurs throughout the report. Is this the only major area of
contention in the requirements/solution that PA identified?
18. I Page 9-33 I I believe change control process to be one of the better Error
Qnd bullet I managed areas. It can be slow due to the wide constituency
I that needs to be consulted and this may be open for some
streamlining. The problem I would suggest is that ina
programme with delivery problems (and a long lead time for
implementation of new facilities) there is a tremendous
pressure too conclude change in unrealistic timescales.
19. I Page 9-33 I I would like to understand what PA mean by “simplification of I Justify
3rd bullet I the contract”, and whether this takes jnto account the and
complexity of the BA-POCL commercial contracts as well as I clarify
I the three Pathway contracts.
(CADATA\WORDIREVIBWSIPA_VERILDOC page 2 of 2 saved: 03/09/97 12:57; printed: 08/09/97 13:00
78/28"*d
IOS 22:2T 26. dss BQ
POL00090015
POL00090015
IN STRICTEST CONFIDENCE
Peter Copping
PA Consulting Group
123 Buckingham Palace Road
LONDON 8 September 1997
SWIW 9SR 53-28
Dear Peter
HORIZON OR BA/POCL PROGRAMME REVIEW - PART II!
Further to my fax of earlier today; I said if I thought of any more potential big
gaps, I’d let you know.
Two points do occur: first, on the basic technical question of whether there’s a
basic ICL Pathway design flaw or not that, combined with the sheer scale of
what we're trying to do, makes the programme inherently unstable. The
report is silent on this explicitly at the moment, through it implies the design
if feasible. I wonder if people who worked on the initial technical evaluation
(including, if I recall correctly, a PA consultant) could help here?
Secondly, in relation to the PDA’s future role and the associated work with
that (eg end-to-end process design), some acknowledgement of, or comment
on, the work already in place through Andersons and French Thornton
(commissioned by the PDA Board) would probably be right.
Thanks
Yours sincerely
PAUL RICH
Copy: Mena Rego (on return)
Bruce McNiven, PDA