Capgemini UK Plc v Dassault Systemes UK Ltd

Hearing date: 14 October 2024

This judgment was handed down by the Judge remotely by circulation to the parties' representatives by email and release to The National Archives. The date and time for hand-down is deemed to be 10:30 on Friday 1 November 2024.

Sean O'Sullivan KC (sitting as a Deputy High Court Judge):

This is the Claimant (“C”)'s application for summary judgment under CPR 24 and/or strike out under CPR 3.4, premised upon what are said to be two short points of contractual construction raised by the Defendant (“D”)'s Amended Defence and Counterclaim. It is important to understand that C does not suggest that it can have summary judgment on its claim, or on D's counterclaim. Rather, its position is that the answer to these two individual issues is so obvious that the Court ought to “grasp the nettle” and resolve them now, on the basis that this would, or at least might, reduce the scope of the dispute before disclosure, witness evidence, etc.

C's underlying claim is for breach of a Settlement Agreement, entered into by the parties on 11 December 2017 (“the SA”). The purpose of the SA was to resolve disputes which had arisen under an earlier agreement (namely the Prime Contractor Agreement or “PCA”). The background to both agreements is a project for the creation and provision of logistics planning technology for the Royal Mail Group (“RMG”).

C's claim is, in essence, that D failed to perform the work required of it and thereby repudiated the SA. D denies that there was any repudiation and counterclaims for further sums to which it says it is entitled for the work which it did perform. As part of both its defence to the claim, and the counterclaim, D relies upon the terms of the PCA, as well as on the SA. It also contends that its work was delayed and disrupted by errors in certain data which was provided by RMG.

It is in that context which the two issues which are the subject matter of C's present application arise. They have been labelled and described by C as follows:

4.1. “The Entire Agreement Clause Issue”. The issue here is whether an entire agreement clause in the SA has the effect of extinguishing the PCA, and hence precludes D from relying on any of the terms of the PCA in its defence to C's claims, or in order to support counterclaims.

4.2. “The Delay Construction Issue”: This issue concerns D's argument that, as a matter of construction, or by virtue of an implied term, C was contractually responsible for errors in data provided by RMG.

Factual background

The parties and the project

Both partes are providers of software and technology services.

They were engaged in a project to provide what is called a “National Scheduling Tool” for RMG. This tool was to replace the systems previously used to carry out logistics planning for RMG's national delivery network. The logistics problem or ‘puzzle’ faced by RMG involves the transportation of mail throughout the United Kingdom by rail, road and sea.

I do not need to descend into the detail of what the tool – the software – is intended to do. In simple terms, it solves logistics puzzles by using a processing engine:

7.1. Customer-specific rules are defined to identify the parameters within which the logistics puzzle must be solved. Examples of customer rules for the RMG project include how long drivers are permitted to work before a break is required.

7.2. The software identifies an initial solution to the logistics puzzle. That initial solution is evaluated by reference to ‘KPIs’ (key performance indicators). These are concerned with outcomes which reflect the cost of the solution (such as the amount of fuel needed).

7.3. The optimiser software makes random changes to the initial solution. Changes that increase the quality of the solution (measured by reference to the KPIs) are retained; changes that do not are reverted. By this process of optimisation, the software should be able to arrive at a high-quality logistics solution within a given timeframe.

7.4. In order to assist the optimiser in reaching a high-quality solution more quickly, the optimiser is ‘tuned’. The tuning process involves identifying the most fruitful types of changes to the initial solution and writing code to ensure that, in the future, those types of changes are considered before other changes.

I am told that a key part of the evaluation of the performance of an optimiser is the concept of a “benchmark”. That involves the customer (e.g. RMG) creating a manual solution for a given dataset. That dataset might contain, for example, details of all of the parcels to be delivered over a particular week, and the resources available to RMG, such as trucks, trains and drivers. The logistics solution provided by the software for that dataset can then be tested against that manual solution.

The PCA

The PCA was executed in September 2015, between C and a company named Quintiq. D subsequently acquired Quintiq and has taken on its rights and responsibilities. It was common ground between the parties that, for the purposes of this application, I could treat Quintiq and D as interchangeable. I will therefore continue to use “D” to mean D or Quintiq, as relevant.

The PCA is in familiar form for an agreement between a head contractor and a subcontractor in the context of an IT project. Picking out a few of the terms which are canvassed further below:

10.1. clause 2 contains provisions in relation to the price and payment terms;

10.2. clause 3.1 contains limitations of liability. Liability for breach is limited to 150% of the project value (clause 3.1(c)); loss of profit and consequential loss are excluded (clause 3.1(d)) and there is no liability for any claim made more than 2 years after termination of the PCA (clause 3.1(f));

10.3. clause 3.2 provides some rights of termination;

10.4. clause 3.4 addresses IP rights, making clear for example what rights each party has in relation to the use and ownership of software;

10.5. clause 3.8 deals with confidentiality in relation to the data, documentation and information which was to be shared;

10.6. clause 3.12 provides for a tiered escalation of disputes;

10.7. clause 3.13 sets out C's requirements for ethical business relationships and sustainable procurement;

10.8. clause 3.14 deals with the parties' anti-corruption and bribery obligation; and

10.9. clause 4 is a bundle of warranty provisions, providing the usual collection of obligations and protections for D in relation to non-conformances emerging after acceptance by C.

In addition, the PCA appended a long and detailed Statement of Work dated 19 October 2015 (“the SOW”). My attention was drawn in particular to section 6 of this document, which set out a long series of “General Assumptions”. For example, one assumption was that “ Realistic and complete test data will be available on time in accordance with the project plan”. Under “ Who”, C was identified. I understand this to mean that (a) it was an assumption underlying the lump sum price and the agreed schedule for this work, that realistic and complete test data would be provided by certain identified dates and (b) that, as between C and D, C was the party taking contractual responsibility for the fulfilment of that assumption.

Section 7 of the SOW also confirmed that the work identified was to be executed for a fixed price of £925,911 (plus VAT). However, it was provided that “ Other activities (such as Change Requests etc.) will be invoiced as per the daily rate of 1100.00 GBP on a T&M basis”.

Deadlock

D's position was (and is) that its workscope under the PCA (see section 6.1.7 of the SOW) required it to test and tune the optimiser against 3 benchmarks provided by RMG and that it did so. It says that, in December 2016, RMG could not provide a large benchmark solution against which the optimiser could be tested. Instead, RMG provided smaller benchmarks (i.e. benchmarks that did not relate to the entirety of the logistics puzzle to be solved, but only part of the puzzle). Those were used for testing the software. D says that at least 3 smaller benchmarks were used in this way.

However, in July 2017, D's case continues, C requested that D test and tune the optimiser against a large benchmark. That led to a dispute between the parties as to whether that work was outside the defined scope. That disagreement formed part of a wider dispute about D's entitlement to additional payments, which resulted in an impasse between the parties.

The SA

The parties entered into the SA in December 2017 to get the project back on track. The SA addressed the two aspects of the parties' disputes separately: (a) it settled the dispute in relation to historic work (the SA defines this as the “Dispute”); and (b) it set out a way forward for the work outstanding under the PCA (the SA defines this work as the “Go Forward Work”).

The distinction was explained in the recitals:

“(D) A dispute has arisen between the Parties relating to sums claimed by [D] under various disputed Change Requests and Invoices issued by [D] under Prime Contractor Agreement all as set out in Appendix 1 to this Agreement (“the Dispute”).

(E) The Parties are also in dispute in relation to the work to be delivered by [D] as set out in Appendix 2 to this Agreement (the “Go Forward Work”) but have agreed that this work will continue under...