ICL
Pothway
FUJ00058278
FUJ00058278
INITIAL GO LIVE Ref: BP/REP/oo10
Version: 1.0
Date: 12/11/96
LESSONS LEARNT
Document Title:
Document Type:
Abstract:
Status:
Distribution:
Author:
Comments to:
INITIAL GO LIVE - LESSONS LEARNT
Report
As a result of introducing the Benefit Payment System into
the 10 offices in Stroud early, ICL Pathway has gained
valuable experience which can be applied to the
introduction of the subsequent releases for the national
rollout. This document covers some of the significant
lessons learnt from IGL.
Issued
John Bennett
Tony Oppenheim
Martyn Bennett
Terry Austin
Liam Foley
Steve Muchow
Raj Aurora Anna Campopiano
Paul Curley Barrie Davies
Janet Dore David Groom
Lorraine Holt Ian Honnor
John Hunt Denise Ivermee
Gill Jackson Graham King
Richard Long Bob McDermott
Harvey Potts Martin Riddell
Roger Todd Paul Underwood
Barry Vaughan Alan Ward
Steve Hyde
Library
David Howells
David Howells
COMMERCIAL IN CONFIDENCE Page 1 of 12
FUJ00058278
FUJ00058278
ICL INITIAL GO LIVE Ref: BP/REP/oo10
Version: 10
Pothway LESSONS LEARNT Date: 12/1/96
COMMERCIAL IN CONFIDENCE Page 2 of 12
© 1996 ICL Pathway Ltd
FUJ00058278
FUJ00058278
Ref: BP/REP/oo1o
ICL INITIAL GO LIVE ;
Version: 1.0
Pothway LESSONS LEARNT Date: 12/1/96
O Document controt
O.L Document history
Version Date Reason
Ou 06/11/96 Issued to IGL Team to add to the text here and to gain
comments.
0.2 Approval authoriticy
Name Position Signature Date
Terry Austin Programme Director
0.3 Associated documenty
Reference Vers Date Title Source
fa] BP/PLA/ooo1 1.1 02/08/96 Initial Go Live Stage Management Plan
0.4 Abbreviotiony
0.5 Changes i thy vernow
Updated as a result of additional items received, refer to Acknowledgements.
COMMERCIAL IN CONFIDENCE
Page 3 of 2
FUJ00058278
FUJ00058278
ICL INITIAL GO LIVE Verse, BRIREP/oo10
Pothway LESSONS LEARNT Date: 12/1/96
COMMERCIAL IN CONFIDENCE Page 4 of 12
FUJ00058278
FUJ00058278
INITIAL GO LIVE Ref: BP/REP/oo10
Version: 1.0
Pothway LESSONS LEARNT Date: 12/1/96
0.6 Table of content
3 LESSONS LEARNT .......cssceeseeseees
3.1.2 BA/POCL REQUIREMENTS..
3.1.3 STAGE MANAGEMENT PLAI
3.1.4 SINGLE INTERFACE TO PDA...
3.2.1 DETAILED DESIGN SPECIFICATION .......scesesssessesseesseeseessesseesteeneesee
3.3 TEST AND INTEGRATION ......cessessesseessesseeseesesseessensecsesnesseesseaseessenneeseesee
3.3. JOINT TESTING PROGRAMME
3.3.2 GOING LIVE .
3.4 IMPLEMENTATION.
3.6 COMMUNICATIONS
3.6.1 IGL LAUNCH
3.6.2 BRIEFINGS...
YN DAADAUDAUDR RA
NON
CoO 0 DMD HMDmDaAU A
0.7 Acknowledgementy
1, Anna Campopiano for Communications
2. Gill Jackson for Testing Going Live
3. Lorraine Holt for Training
COMMERCIAL IN CONFIDENCE Page 5 of 12
FUJ00058278
FUJ00058278
ICL INITIAL GO LIVE Verse, BRIREP/oo10
Pothway LESSONS LEARNT Date: 12/1/96
4. Graham King
COMMERCIAL IN CONFIDENCE Page 6 of 12
ICL
FUJ00058278
FUJ00058278
INITIAL GO LIVE Ref: BP/REP/oo10
Version: 1.0
Pothway LESSONS LEARNT Date: 12/1/96
INTRODUCTION
As a result of introducing the Benefit Payment System into the 10 offices in
Stroud early, ICL Pathway has gained valuable experience which can be applied
to the introduction of the subsequent releases for the national rollout. This
document covers some of the significant lessons learnt from the IGL stage
within ICL Pathway.
This report has been divided into the key functional areas within ICL Pathway
and is based upon the collective experiences of the IGL team to date following
the successful introduction of releases 0.1 and 0.2. Further details are defined in
the IGL Stage Management Plan, ref [1].
SCOPE
This report covers the lessons learnt to date from each of the key functional
areas within ICL Pathway only, as a result of the Design, Development, Testing,
Implementation and Support of the Initial Go Live releases into the 10 Stroud
offices.
Given the complex nature of this type of programme and vast number of people
involved at various stages, it is not possible to collate all of the experience
gained across ICL Pathway within this report. This report covers only some of
the more significant lessons which have been highlighted to the IGL team over
the past 4 months.
Release 0.2 has only been in live operation for three weeks at the time of writing
this document, and therefore the steady state teams will continue to learn over
the next 6 months during the life span of IGL.
LESSONS LEARNT
For ease of reference this section has been divided into functional areas within
ICL Pathway, although some of the points are general in nature and may apply
to more than one of the areas described here.
COMMERCIAL IN CONFIDENCE Page 7 of 12
FUJ00058278
FUJ00058278
ICL INITIAL GO LIVE Refi BP/REP/oo1o
Version: 1.0
Pothway LESSONS LEARNT Date: 12/1/96
31 PROGRAMME MANAGEMENT
3.1.1 IGL PROGRAMME
IGL was implemented as a result of a political imperative from Peter Lilley
which called for the introduction of payments of benefit by card, with the
deadline of the 23"¢ September. From the award of contract to Pathway in May,
this left just 18 weeks to go live with no option to slip this date.
The IGL programme was compressed to the minimum work required in order
to meet the objectives within the required timescale. As a result, functionality
within the system that could not be developed within the time available, was
worked around by defining manual processes within the Post Offices and BA.
Without the political influence and deadline, it is likely that more time and
effort would have been put into developing and testing the products, with less
reliance on manual processes.
Pathway itself was just evolving in all areas during this period. Many of the
resources required for this stage were brought on part way through and the
supporting business processes were defined and refined as Pathway grew.
In summary, by the nature of IGL significant work was required during the
programme to recruit staff and define supporting processes and procedures
from scratch, alongside the primary task of building the IGL system and service.
3.1.2 BA/POCL REQUIREMENTS
Throughout the stage, additional requirements on Pathway appeared over and
above those agreed initially within the Stage Management Plan. These
originated from groups within BA and POCL who were not directly represented
within the Step 1 reviews. For example, significant work was required late in the
programme to satisfy the reconciliation process between BA and POCL,
although no MIS reporting requirements were identified by the Step 1 team
when the plans were defined.
It is essential that PDA own and manage the interfaces to other parties within
the automation programme and that strict change control is applied to new or
amended requirements.
COMMERCIAL IN CONFIDENCE Page 8 of 12
FUJ00058278
FUJ00058278
ICL INITIAL GO LIVE Verse, BRIREP/oo10
Pothway LESSONS LEARNT Date: 12/1/96
3.1.3 STAGE MANAGEMENT PLAN
As mentioned previously, despite all of the discussions between Pathway and
the PDA to define the requirements, there was a tendency for people to look at
what was delivered and then decide what they really required. Also
requirements which had not be defined, i.e. receipt formats, became “essential”
to the business when we declared it was too late to develop new functionality.
The key document during these discussions was the Stage Management Plan.
This document stated up front what our baseline assumptions were, specific
details in our solution not mentioned elsewhere. It also described our approach
in certain circumstances, i.e. when CAR’s were not achieved in time.
This highlights the importance of document management and change control.
It is essential to ensure that the change control process is followed even if it
requires ad hoc meetings. Pathway must learn to say “no” to changes that
become imperative to the business if these are not defined in time.
3.1.4 SINGLE INTERFACE TO PDA
Early on in the programme it became apparent that the Step 1 team were being
given varying answers to the same questions depending on who they were
talking to in Pathway at the time. This occurred particularly during the testing
phase whilst the procedures were being produced, where answers to specific
questions were based upon either what the system should do or what it does
do. In addition different parts of the programme would come up with the same
questions. The Stage Manager then became the single point of contact for
Pathway and all questions from their side were directed to a single contact on
the Step 1 team. Queries were recorded in the “Daily Fax” and responses could
be monitored. This resolved the problem of conflicting messages and avoided
wasted effort replying to the same questions. However, at particularly busy
periods this process did become a bottle neck in communications between
Pathway and the rest of the programme.
For release 1, the communication process in Pathway needs to be considered in
terms of Authorised Contacts for the PDA.
3.1.5 PROGRAMME COMMUNICATIONS
During the lead in to going live in September and October daily conference
calls were found to be necessary. An “A Team” was set up which consisted of
representatives from the Step 1 team, Pathway, CAPS, BA region and the Post
Office region. This provided a rapid communication mechanism across the
whole programme at a critical time.
COMMERCIAL IN CONFIDENCE Page 9 of 12
FUJ00058278
FUJ00058278
ICL INITIAL GO LIVE Verse, BRIREP/oo10
Pothway LESSONS LEARNT Date: 12/1/96
3.2 DESIGN
3.2.1 DETAILED DESIGN SPECIFICATION
For IGL, the architecture design was only available at a high level except for the
specification of the applications. This meant that detailed design, in particular
hardware configurations were produced mainly in the Solution Centre, in order
to meet the testing schedule. From the configurations set up in the testing cells
the build scripts were developed and issued. Given this process of refining the
specifications from development, testing and implementation, meant that the
design could change when finally implemented. For example, only a single
correspondence server configuration was implemented because there was no
specification available for a multiple server configuration in time for this to be
tested.
3.3 TEST AND INTEGRATION
3.3.1 JOINT TESTING PROGRAMME
At the start of IGL, BA had responsibility for acceptance testing the end to end
processes and counter procedures. As the testing plans were developed for
release 0.1, the PDA produced their own independent test schedule, including
security testing. The result was that the T&I team had to support two testing
programmes with BA/POCL which included duplicate tests. This significantly
tied up valuable resources and limited the amount on internal testing achieved
in the time available. For release 0.2 we ensured that the PDA produced a single
programme which covered all requirements from BA and POCL.
3.3.2 GOING LIVE
The following are specific points picked up by the testing team during the
preparations for going live :
© aconsiderable amount of time was spent doing system back up and restores
e there is a need to determine centrally that the links to the PO’s have worked,
ie. the data has arrived and is correct. The volume of data and number of
PO’s in release 1 will mean that it will not be feasible to travel around and
manually check the records
for release 1 it is recommended that the live data is transferred into another
test machine to validate the basic functions, for example encashments etc.,
rather than loading onto the live environment, testing, backing out and then
replacing
COMMERCIAL IN CONFIDENCE Page 10 of 12
FUJ00058278
FUJ00058278
ICL INITIAL GO LIVE Verse, BRIREP/oo10
Pothway LESSONS LEARNT Date: 12/1/96
e in future we should verify the data migration process against live data and
then validate that the normal processes, for example encashments, work
against the migrated data
¢ better tools/facilities are required for looking at and analysing the data on
the servers and within the PO’s, for example file compare etc.
* non contributing personnel, i.e. interested parties from the PDA, should be
kept away from the local environment to avoid distracting the team during
key periods of going live.
3.4 IMPLEMENTATION
3.4.1 PROCESSES AND PROCEDURES
Defining the processes and procedures for IGL was delayed until the end of the
stage due to the late availability of the products. This did not only affect the
counter procedures but operating/support processes as well. In particular,
significant additional effort was required by Pathway staff to help the PDA
define local procedures i.e. BA office processes etc.
3.4.2 TRAINING
The following is a summary taken from the report on the IGL Training in
Stroud prepared by Elizabeth Battell.
e the training and training documentation worked very well for the post
masters
e the use of the “Counter Workstation Consolidation Workbook” exercises are
essential for user’s confidence and should be endorsed by post masters and
management
e the Counter Procedures manual must match the system operation
¢ trainers need a comprehensive understanding of current Post Office
procedures
¢ awareness of the post masters own “tried and tested” communication/help
systems are necessary, both correct and incorrect information travels very
fast
e the Post Office customer needs extra guidance to ensure the instructions
given are acted upon, This will also assist with continued Customer Care at
the Post Office
e “In Store” training is not ideal, and if used, time needs to be added to take
account of the distractions and interruptions.
COMMERCIAL IN CONFIDENCE Page u of 12
FUJ00058278
FUJ00058278
ICL INITIAL GO LIVE Verse, BRIREP/oo10
Pothway LESSONS LEARNT Date: 12/1/96
3.5 CUSTOMER SERVICES
3.5.1 JOINT LESSONS LEARNT
A separate document is being jointly produced by Rod Stocker and Martin
Riddell which covers the lessons shared between BA/POCL and Pathway.
3.6 COMMUNICATIONS
3.6.1 IGL LAUNCH
The IGL launch of the card was driven by the Secretary of State seeking a high
profile launch. In general the logistics of the event worked well - we received
some excellent local coverage of the card and largely accurate communication
of the main messages.
Having the event in Stroud however, reduced national press interest and
sending out invitations so late reduced press ability to attend.
3.6.2 BRIEFINGS
Changing the date of the Special Interest Group briefing (originally planned for
the 16" September in Stroud) and holding the revised briefing in Leonard
Stanley, coupled with the sending out the new invitations only 5 working days
before the event, meant that only 15 individuals turned up - this is low
compared to the 48 who responded to the first invite. As a result the Special
Interest Groups were not as well briefed as we had originally planned -
necessitating us to follow up with a mail shot to all those who could not attend:
a huge amount of effort and cost. Lessons learnt are:
e make sure the Secretary of State is in agreement with the plans
e make sure the customer education objective remains paramount.
3.6.3 LIVE TRIAL
At Live Trial our imperative should be to ensure that the messages and all
activity is customer education based. This is especially key as Live Trial will be
impacted at some stage by a General Election.
COMMERCIAL IN CONFIDENCE Page 12 of 12