Markit Systems Limited v Fulcrum Group Propriety Limited (A5071/2021 ; 39743/2018) [2023] ZAGPJHC 429 (5 May 2023)

begin wrapper begin container begin header begin slogan-floater end slogan-floater - About SAFLII About SAFLII - Databases Databases - Search Search - Terms of Use Terms of Use - RSS Feeds RSS Feeds end header begin main begin center # South Africa: South Gauteng High Court, Johannesburg South Africa: South Gauteng High Court, Johannesburg You are here: SAFLII >> Databases >> South Africa: South Gauteng High Court, Johannesburg >> 2023 >> [2023] ZAGPJHC 429 | Noteup | LawCite sino index ## Markit Systems Limited v Fulcrum Group Propriety Limited (A5071/2021 ; 39743/2018) [2023] ZAGPJHC 429 (5 May 2023) Markit Systems Limited v Fulcrum Group Propriety Limited (A5071/2021 ; 39743/2018) [2023] ZAGPJHC 429 (5 May 2023) Download original files PDF format RTF format make_database: source=/home/saflii//raw/ZAGPJHC/Data/2023_429.html sino date 5 May 2023 IN THE HIGH COURT OF SOUTH AFRICA (GAUTENG DIVISION, JOHANNESBURG) CASE NO: A5071/2021 A QUO CASE NUMBER: 39743/2018 NOT REPORTABLE NOT OF INTEREST TO OTHER JUDGES REVISED In the matter between : MARKIT SYSTEMS LIMITED Appellant And FULCRUM GROUP PROPRIETARY LIMITED Respondent Neutral citation: Markit Systems Limited v Fulcrum Group Propriety Limited (Case No. A5071/2021) [2023] ZAGPJHC 429 (05 May 2023) Summary: Contractual damages arising out of cancellation of contract under cancellation clause - appellant averring cancellation clause not lawfully invoked – appellant averring respondent repudiated contract – appeal dismissed – court upholding court a quo’s finding that respondent had lawfully cancelled contract under cancellation clause – counterclaim also upheld. JUDGMENT KEIGHTLEY J: (WINDELL J and MIA J concurring) INTRODUCTION 1. In April 2017 the parties in this appeal concluded a technology agreement (the agreement).  On 13 December of that same year the respondent, Fulcrum Group (Pty) Ltd (Fulcrum), delivered a notice of cancellation to the appellant, Markit Systems Limited (Markit).  Markit contended that Fulcrum had breached its obligations under the agreement and unlawfully repudiated it by purporting to terminate the agreement when there were no lawful grounds for doing so.  It instituted an action in which it claimed contractual damages from Fulcrum in the amount of some R90,9 million.  Fulcrum disputed the claim and, in turn, instituted a counterclaim in which it sought damages from Markit in the sum of R4,5 million.  The High Court dismissed Markit’s claim and upheld Fulcrum’s counterclaim.  With leave of the trial Judge, Adams J, Markit appeals to this Full Court. 2. Markit is a software company based in the United Kingdom.  It has offices in London and, for purposes of the agreement in dispute in this matter, also used the services of developers situated in India.  Fulcrum is in the business of providing financial services in the insurance industry.  Its clients are insurers, brokers and underwriting managing agents.  At the time that the agreement was concluded Fulcrum utilised the Flexibroker business information and operating system which had been in place for 27 years.  The Flexibroker system was an internal networked system.  It resided on a server in the offices of Fulcrum’s respective clients, rather than being hosted on an online platform.  Fulcrum provided the system to its clients, and they would use the system to run their businesses.  Fulcrum had recognised that the Flexibroker system was outdated, and it was for this reason that it engaged Markit to provide it with an updated and expanded online platform for Fulcrum’s clients. 3. Fulcrum wanted a system that would take its business into the future with the intention of making Fulcrum a leading service provider in the insurance industry.  It was envisaged that the new platform would enable Fulcrum’s individual clients to manage policies, claims and premium transactions; perform online quoting; ensure dynamic reporting and enable support services to their own clients.  This gives some idea of the complexities involved in the system that was to be built by Markit.  Adding to the complexity was the nature of Fulcrum’s business in that not only did Fulcrum require a multi-tenancy system, but also one which could service multiple users. 4. Markit’s Chief Technical Officer, Mr Mclean described the company as being a technology services provider to clients in the insurance industry.  He explained that Markit has extensive experience in providing multi-tenancy technology solutions to clients in the insurance industry.  It aims to align and configure the online platform with the commercial model and business needs and requirements of its clients.  At the time the agreement was concluded, Markit had provided and built systems for two other South African clients in the insurance industry.  One of these was Infiniti, of which we will hear more later. 5. One of the factors that appealed to Fulcrum was that Markit had already developed and built software modules that the parties understood could be used and adapted to suit Fulcrum’s particular requirements.  It was also understood that the existing modules would have to be augmented.  Mr Schreuder, who was Fulcrum’s Chief Information Officer and had been involved in the pre-agreement negotiations with Markit, testified that Fulcrum understood that 80% of its requirements would be met from system modules Markit already had at its disposal.  Mr Mclean put the ratio at 60% use of existing modules and 40% new build.  It is not necessary to determine which estimate is correct.  The important point is that Markit had existing modules at its disposal which it was understood would be used in developing a system for Fulcrum. 6. A further point to note by way of introduction is that the parties were agreed, and it was recorded in the agreement, that the project would commence immediately.  As events unfolded, however, this did not occur.  For one reason or another, it was not until about July 2017 that anything of note happened.  I will deal with the chronology of events in more detail later in my judgment. TERMS OF THE AGREEMENT 7. The agreement recorded that Markit would provide Fulcrum with technology and related services.  The scope of the project was described with reference to the Business Development Agreement (BRD), which was defined as: ‘ a document detailing the business processes the system will need to address as set out in Schedule 2 ’ .  As appears from the remainder of this judgment, the BRD looms large in the appeal. 8. Clause 2 of the agreement is important.  It provided, in relevant part: ‘ Scope of the project [Markit] and [Fulcrum] have entered into a technology agreement, whereby [Markit] will be appointed as the primary technology partner for the project. The scope of the project is explained in more detail in the BRD as detailed in Schedule 2. The Project shall be implemented in phases, to be agreed upon in a detailed project plan. Once [Markit] has analysed and documented [Fulcrum’s] business requirements, both parties shall jointly agree on : a. A BRD which shall set out what [Fulcrum’s] requirements are subsequent to a detailed analysis of [Fulcrum’s] business by [Markit] . The BRD shall include (but shall not be limited to): I. Application Functional and Non-Functional Requirements II. Migration Requirements including; a. details of insurer products required to be migrated; b. customer data; c. details of the migration process, stipulating the level of automation to be implemented in the process. d. … .. Either party shall be entitled to cancel this agreement in accordance with clause 16 (c) below in the event that the Parties are unable to reach agreement on the details of the BRD.’ (My emphasis) 9. Clause 3 described Markit’s responsibilities which included its responsibility to: ‘ a. Analyse and document [Fulcrum’s] requirements, to be encapsulated in the BRD as provided in clause 2(a) above … .’ (My emphasis) 10. Fulcrum’s concomitant responsibilities were outlined in clause 4.  These are described, again in relevant part, as follows: ‘ a. [Fulcrum] acknowledges that the [Markit’s] ability to provide the Services is dependent upon the full and timely co-operation of [Fulcrum] (which [Fulcrum] agrees to provide) , as well as the accuracy and completeness of any information and data [Fulcrum] provides to the [Markit]. Accordingly, [Fulcrum] shall provide [Markit] with access to, and use of, all information, data and documentation reasonably required by [Markit] for the performance by [Markit] of its obligations under this agreement . b . To assist [Markit], [Fulcrum] will appoint a dedicated Business Analyst to assist with the creation of the BRD . [Fulcrum] shall be responsible for the accuracy and completeness of the Materials on the System.’ (My emphasis) 11. Clause 16 is central to the appeal.  It provides: ‘ Term and termination a. This agreement shall commence on the Effective Date and shall continue indefinitely unless cancelled upon written notice as provided for in this clause 16. b. Notwithstanding the above, neither party shall be entitled to cancel this agreement within the first five years from the Effective Date, save for termination in accordance with this clause 16(c) . c. Either party may terminate this agreement immediately at any time by written notice to the other party if: (i) that other party commits any material breach of its obligations under this agreement which (if remediable) is not remedied within [14] days after the service of written notice specifying the breach and requiring it to be remedied; or (ii) … (iii) … (iv) … (v) The Parties fail to reach agreement in respect of the details to be documented in the BRD ; (vi) … (vii) … (viii) … .’ (My emphasis) 12. Clause 17 is the breach clause: ‘ Should a party breach ("the Defaulting Party") any clause or warranty in this agreement, the other party ("the Aggrieved Party) shall be entitled to claim payment and/or immediate performance by the defaulting party of all of the Defaulting party's obligations that are due for performance, without prejudice to the Aggrieved Parties ( sic ) right to a claim for damages.’ 13. Clause 19 read with Schedule 5 detailed the fee schedule payable by Fulcrum to Markit.  In the first year of the project, development costs were payable for the development and configuration of the system.  These were based on an assumed build period of 12 months.  These costs were payable at different stages, as explained in Clause 19: ‘ c. An initial deposit amount (1st instalment) shall be paid on the Effective Date as an upfront payment ( prior to completion of the BRD) of the development costs. i. 50% of this initial deposit shall be made by [Fulcrum] to the [Markit]. The remaining 50% will be paid into [Fulcrum’s] elected attorney's sterling … bank account, which attorneys shall be based in the United Kingdom. ii. The 50% held in trust by the appointed Attorney shall be released to [Markit] subject, firstly, to agreement to the BRD by both parties and, secondly, to completion of the BRD in accordance with the terms and conditions of this agreement. This will give all parties the opportunity to revise the project costs based on the agreed Requirements. iii. Should this agreement be terminated in accordance with clause 16, due to non- compliance with the BRD or failure to reach agreement on the BRD; the balance of the initial deposit shall thereafter be paid over to [Fulcrum]. d. On the Effective Date, the 2nd instalment shall also be deposited into [Fulcrum’s] elected attorney's sterling … bank account, which attorneys shall be based in the United Kingdom. This shall be released for payment to [Markit] on Commencement of build (expected commencement date is 1st July 2017) .’ (My emphasis) 14. Schedule 5 made provision, in addition to the payment of development costs, to the payment of a licence fee: ‘ A license fee will be payable based upon the value of the Gross Written Premium (GWP) transacted through the System. There will be a minimum and deposit fee adjustable at an agreed annual rate on and applicable to all GWP processed through either Flexibroker or Delta. The minimum deposit will become payable when the System is live.’ 15. Finally, clause 29a dealt with variation.  It provided that a variation of the agreement shall be in writing and signed by or on behalf of both parties. EVENTS LEADING UP TO TERMINATION OF THE AGREEMENT 16. The agreement was concluded on or about 26 April 2017.  Markit issued an invoice, for the first 50% of the first instalment, on 4 April.  By agreement, the invoice was paid directly to Markit, rather than to the attorneys.  This was an amount of R1,5 million.  The second invoice for the second 50% of the first instalment, also in an amount of R1,5 million, was paid by Fulcrum in early June 2017. 17. Neither of the parties appointed a new project manager at the commencement of the project.  Instead, Markit relied on Roger Walters to head up the project from its side, while Fulcrum relied on Mr Schreuder.  A steering committee (Steerco) met monthly on a digital platform.  On 4 May Mr Schreuder provided Markit with access details to log into the Flexibroker demonstration server.  At the June Steerco meeting Mr Thurlow of Markit reported that the project was still in the scoping phase and that by end of July/August it should get to a stage where an overarching BRD had been created.  Further, Mr Walters had started to fill out the BRD in more detail. Mr Thurlow explained that the BRD was ‘a living document’ and that the time would come, when the groundworks were in place, that sections would need to be signed off once everyone was comfortable with them. Once this happened it would be necessary to ‘drill down and prioritise what needs to be done and allocate responsibilities.’  Elsewhere in the minutes it was recorded by Markit that a ‘full BRD’ may not be practical. 18. It is also clear from the minutes of this Steerco meeting that both parties were still in the process of bringing resources into the project.  In fact, in the July 2017 Steerco minutes it was recorded that the project was 2-3 months behind due to resource constraints on both sides.  It should be noted that the second payment to Markit was made in June 2017 despite the BRD not having been agreed.  Under the agreement, payment was only due on agreement of the BRD. 19. Mr Walters had drafted a Project Initiation Document (PID) before the agreement was concluded.  Mr Schreuder added to this document in June, indicating that the BRD ‘owner’ was Fulcrum.  Also in June, Mr Walters provided a template, high level draft BRD, which Mr Maclean agreed was an initial draft for review.  It contained references to ‘Lloyds brokers’.  In July Mr Walters updated the draft but the reference to Lloyds brokers remained.  Shirley-Anne Oakley-Brown was engaged as project manager by Fulcrum. 20. In August 2017 Fulcrum appointed Ms Rika Maclean as an interim business analyst.   Mr Walters updated the high-level template BRD by adding some user scenario example sketches.  On 16 August 2017 Markit invoiced Fulcrum for the second instalment under clause 5 of the agreement in the amount of R3 million.  As per the agreement, the amount was payable on commencement of the system build.  Fulcrum had misgivings as to whether Markit had indeed commenced the build.  This led to an important meeting on 24 August in Johannesburg attended by Mr Turner and Mr Thurlow of Markit.  Mr Schreuder referred to it as an ‘emergency’ meeting, at which the two Markit executives were asked questions about the data build.  They assured Fulcrum that the build had commenced, and they undertook to provide a demonstration at the workshop that had been arranged in South Africa between Fulcrum, Markit and its Indian developers at the end of August until 6 September.  Based on these assurances, Mr Ian Bain, of Fulcrum gave the go ahead for Markit’s invoice to be paid. 21. It was at this stage that a shift seems to have occurred in the conceptualisation of how the project would run.  In an email from Ms Oakley-Brown, noting points from the 24 August meeting, she recorded among other things that the project would not be run on the lines of a conventional ‘waterfall approach’, but would use a hybrid system.  The significance of this is that according to Markit, it was not necessary for each element of the project to be completed in linear fashion until the next could commence: it allowed for a more ‘agile’ system, where components could be worked on simultaneously. 22. The workshop from the 30 August to 6 September 2017 was attended by 16 Fulcrum staff, including subject specialists.  Markit brought 2 executives and 4 developers from India.  Fulcrum picked up the bill for the workshop.  The demonstration of the system build promised by Markit took place at the workshop.  However, according to Fulcrum, contrary to what they had been led to believe, what they were shown was nothing new: it was the Kuda system, which had been developed for another of Markit’s South African clients, which Fulcrum had already seen.  Markit disputes that the demonstration did not show any new build. 23. Mr Schreuder told the court that he was concerned about what appeared to be an absence of any real indication that Markit, in fact, had commenced the system build.  He was not the only Fulcrum staff member who was concerned.    Karen Bouic had been appointed to Fulcrum in September as an additional resource on the project. Ms Oakley-Brown and Ms Bouic arranged to meet with one of Markit’s pre-existing South African clients, Infiniti, as they shared Mr Schreuder’s concerns.  Their evidence was that this meeting raised concerns about project delays that Infiniti had experienced with Markit and about whether Markit had sufficient knowledge and experience of the South African insurance landscape required for the task.  They were also concerned because the Infiniti project was much smaller than the Fulcrum project.  I need to record that none of these misgivings was communicated by Fulcrum to Markit at the time. 24. Fulcrum had started working on the sub-BRDs after the workshop.  In October 2017 Mr Mahabeer was appointed as Fulcrum’s business analyst to replace Ms Mclean.  He and Mr Walters engaged in email discussions on the development of the sub-BRDs (encapsulated in the ‘multi-coloured email’ Mr Berridge, for Markit referred to in his submissions).  Mr Mahabeer produced a draft of the first BRD, the ‘product and entity BRD’ and sent it to Markit for review.  This was between 25 and 28 October.  The document was over 60 pages long.  In the meantime, there were ongoing emails between Markit and Fulcrum regarding the project, including question-and-answer emails. 25. In the third week of November Fulcrum received some indication that Markit had not, in reality, commenced with the build.  This occurred when Markit gave Fulcrum an update on its team structure by way of an email on 22 November.  Ms Oakley-Brown noted that Mr Pine was now the project manager, rather than Mr Walters.  She asked Mr Walters when this change had occurred.  He responded: ‘ Just to elaborate on the team structure, as we move to the build phase we have naturally had to increase our team size and specialise in terms of roles. I still oversee the whole project, Chris as PM will focus on the day to day detail of each release. I am not going away… .’ (My emphasis) 26. Fulcrum’s case is that the underlined portion of this email confirmed its fear that Markit had not yet commenced the build, despite having been paid the R3 million instalment linked to build commencing, and despite Markit’s assurances that the build had commenced.  I shall deal with this in more detail later.  I should note, however, that despite its misgivings, at this stage Fulcrum kept its concerns to itself and did not confront Markit with them. 27. A further development increased Fulcrum’s concerns.  On 27 November Ms Oakley-Brown emailed Mr Mclean requesting sign-off of the first sub-BRD, which Mr Mahabeer had completed and forwarded in the last week of October.  She sent a reminder email on 28 November.  On the same day Mr Mclean responded: ‘ Ha ving reviewed the Product BRD with the team I am able to sign o ff condition a l ly with the following caveats . The Product BRD i s at a high level and lacks detail in key areas. We need this detail in order to complete the build. We can commence the build but we require further detail in order to process the build to completion. Question and answer sets in particular have not been supplied - we require worked examples of products. We require the underwriting specification to put the product build into context. Can you advise when we are likely to receive the outstanding information please. In the meantime we will commence the build based on our current understanding and information.’ (My emphasis) 28. Ms Oakley-Smith was not best pleased.  She made this clear in her response.  She indicated that she was ‘very disappointed’ with it as Mr Mclean had had the document for five weeks (it was slightly less than this length of time).  She added that Mr Mclean was supposed to be delivering some sort of functionality on how the product set up would look by the coming Friday.  Mr Walters emailed back on 29 November in order, as he put it, to ‘calm things down’.  He confirmed that Markit intended to issue a functional spec that week based on the information that had been provided.  However, he qualified his response by saying: ‘ I understand why you want formal sign-off because that is what you would normally do in a standard waterfall type delivery, but as we are having to work with this incrementally we need to be realistic about the accuracy of each batch of requirements until you have completed and reviewed them all in the context of each other.  We are happy to proceed based on the information we have at the moment … .’ 29. Ms Oakley-Smith acknowledged the points but also noted that they were not working in a typical ‘agile’ methodology because the team was not all co-located in the same room.    She reported that the ‘financial BRD’ was 99% complete from Fulcrum’s side and they had already started working on the next BRD.  This should: ‘… help things become a lot clearer and help both parties become a lot more comfortable.’ 30. In the Fulcrum camp, there was growing concern.  Mr Mahabeer responded to Mr Walters’ 29 November ‘calm things down’ email noting several concerns: ‘ Its ( sic ) going to be bit difficult running true agile with key SME's, business and dev situated in 3 different geographic locations. Also taking the communication barrier into consideration, requirements can easily get lost in translation … … Honestly, I do not know what modules you have vs don't have. We are hoping to leverage off your functional solution/ design if you have one. Perhaps this is the reason why we are misunderstanding each other ( sic ) I think first and foremost- Please advise, is there a base platform? If not, and we are building this Policy Admin System Greenfields, then we would know how to best approach this? Would it be better for you to commit some of your Indian resources here to South Africa for a period of time to assist in eliciting the information and building? Do you need me to sit in India with your team in order to ensure design is as per spec, and testing runs smoothly? See, these are all the questions to the Project approach that we need to answer.’ 31. Despite his concerns, Mr Mahabeer continued to communicate with the Markit team.  One of the requests he made in early December was whether Markit could provide a demonstration on how the product builder would work.  Mr Mclean responded that Markit could give a demonstration ‘of parts of it as it becomes available.’  On 6 December, following a discussion between Fulcrum and Mr Turner and Mr Thurlow the CEO of Fulcrum, Vaughn Jones, sent the following email: ‘ Dear Simon and Nicholas. Thank you for the open and frank discussion earlier today. I would like to reiterate that Fulcrum is committed to try and make the partnership work with Marklt ( sic ) as it is undoubtedly in the best interests of all parties concerned to see the project succeed. This notwithstanding. we have some very real concerns around the progress of the project to date particularly in regard to seeing any tangible evidence of the delivery requirements to support the third instalment payment of R3m. To this end, and following on from our conversation this morning, before any further invoices are submitted we require the following: 1. Proof of commencement of build and evidence of work commenced; 2. Proof of the progress of that build and ability to test what has been built thus far; 3. E v iden c e of your ability to te st yo ur Multi-tenan cy functionality ; 4. Evidence of your ability to complete functional testing of data segregation based on elicitation document proposal; 5. A design proposal of the migration architecture; 6. A revised functional specification(s) aligned to the Product & Entities BRD ’s ; and 7. A revised version of the Test Strategy following the feedback that was supplied by Fulcrum. Please could you provide the above by close of business on 8 December 2017. In order for all parties to prepare for the meeting to be scheduled for 11/12 December 2017. I am sure that you are in agreement that the above requirements are not unreasonable given where we are in the project cycle. We look forward to resolving the items outstanding shortly so that we are in a position to move forward with our partnership in ensuring a successful delivery of the Markit project.’ 32. On 8 December Mr Turner replied that a comprehensive response to this email would be forthcoming on 11 December.  He indicated that several of the points raised needed to be addressed via a series of conference calls and he requested that Fulcrum advise on the availability of its team for 12, 13 or 14 of December.  Markit’s comprehensive response was contained in an email of 11 December, as promised.  In this email, Markit raised some of its own problems with Fulcrum’s performance including: ‘ We feel where there has been a delay, this has been materially contributed to by a delay in Fulcrum's delivery of requirements: e.g. Accounts BRD which was promised end of October, delayed to 30th November but we still do not have. … An area of new concern is the lack of technical knowledge by some members of your team. This has two effects; firstly, we have to re-visit and/or explain decisions that were made some months ago. Secondly, we risk making crucial decisions without appropriate levels of technical input.’ 33. An online meeting was scheduled for 13 December at which Markit would demonstrate its system and the test environment.  Ms Oakley-Brown and Mr Mahabeer attended the demonstration.  The former reported back to the broader Fulcrum team as follows: ‘ Good Afternoon All, This session held on Wednesday 13 Dec at 12 midday was held in order for Markit to take the Team through the Test environment, and the system that Markit uses to track the dev and testing / issues/ communication - Engage. Session was attended by Roger, Malcolm - Markit & Ash, Shirley-Ann & WJ. Originally Nicholas was on the call, and he wanted to clarify what the call was for. I noted that the call was to a) view the Test environment, b) walk through Engage and c) talk though the test strategy doc. The sessions was led by Roger and direct comments such as: • "the test environment is currently set up for the London market” • "product config not is not set up as yet” • "currently populated with Lloyds risk codes” • "test environment has been branded with Fulcrum” • Roger indicated that they could provide us with access and passwords, but there was not much there as yet and as a result was "reluctant to let us play as yet”. Commented by Malcolm that over the coming weeks "considerably more will be added to the site as they are not going on holiday”. Further discussions on Test strategy was that it requires further discussion, as the testing could become overload ( sic ). They would build in small chunks and then test and let us test. Once done then there should be less errors when doing end- to-end testing. I did iterate that full and rigorous testing will be required.’ 34. On the same date Fulcrum delivered its letter of cancellation.  It read: ‘ F ollowing from continuous engagement and several conversations held between the technical and management teams at both Fulcrum and Markit, Fulcrum hereby declares that Markit has materially breached the Technology agreement entered into between the parties. Markit is in breach of its obligations under the above-mentioned agreement for the following reasons: 1.  Markit has misrepresented significant deliverables under the project, including (but not limited to) the date of commencement of build as well as progress made. 2.  Markit has failed to cooperate with the instructions provided by Fulcrum in respect of service levels, deliverables, details of technical documents and has failed to provide tangible evidence of the progress made under this project, despite requests to do so. 3.  Markit has failed to analyse and document Fulcrum's business requirements, and produce a business requirement document (hereinafter “ BRD") which is critically required to form the basis of the end system to be built. 4.  Markit has failed to provide and/ or jointly agree to the BRD, and/or has failed to complete and sign off the required BRD. 5. Markit’ s technical documents have failed to address the specific requirements of the BRD. Markit's technical documents have insufficient detail and amount to scoping documents only, falling short of the technical detail required to develop and build the complex multi- tenancy model required by Fulcrum. 6.  Markit has failed to deliver in terms of the requirements as set out in technical documents. The above confirms that Markit will be unable to provide the services required for the development and configuration of a replacement system to Flexibroker and will be unable to deliver the system required to be developed in accordance with the system specification provided by Fulcrum. Furthermore, the above material breaches have caused an irretrievable breakdown in trust between the parties making it impossible to continue the business relationship. Accordingly, Fulcrum hereby provides notice of immediate cancellation of the Technology agreement between the parties, in accordance with paragraph 16 (c) of the Technology agreement, and specifically in terms of paragraph 16 (c) (v), (vi), (vii) and (viii). We trust that we can proceed with an amicable separation.’ 35. Markit disputed Fulcrum’s entitlement to cancel the agreement and denied that it (Markit) was in breach.  It sought to hold Fulcrum bound under the agreement.  Ultimately, however, it took the stance that Fulcrum’s conduct amounted to a repudiation of the agreement, which repudiation it accepted.  Markit thereafter instituted its action for damages. COURT A QUO 36. Markit’s action was founded on a variety of alleged breaches by Fulcrum of its obligations under the agreement.  The main allegations of breach were that, contrary to its obligations under clause 3(a), Fulcrum had failed to co-operate with Markit to provide the latter with the necessary information about its business requirements in order to finalise the BRD; Fulcrum had reneged on its undertaking to complete the BRD; its purported notice of cancellation was an act of repudiation; and Fulcrum had  breached its payment obligations by failing to pay an invoice of R3 million issued by Markit in early January 2018. 37. Markit claimed damages in the amounts of: 1. R3 million in respect of Markit's invoice dated 31 January 2018. 2. R3 million in respect of the final instalment of the development costs which Markit would have received had Fulcrum not repudiated the agreement. 3. R84,961,742.67 in respect of the license fee which Markit would have earned following upon the new system going live, had Fulcrum not unlawfully repudiated the agreement, based on the Gross Written Premium (GWP) projections provided by Fulcrum at the outset of the project. 38. During the trial, the latter claim was reduced to R40 414 803. 39. Fulcrum denied it was in breach of the agreement.  Its defence was placed squarely on clause 2(d) read with clause16(c)(v).  It pleaded that under these clauses either party could cancel the agreement immediately, at any time, by written notice to the other party if the parties failed to reach agreement in respect of the details to be documented in the BRD.  The BRD was required to be completed within a reasonable time.  This did not occur, as by 13 December 2017, the parties had still not reached an agreement on the BRD.  As such, Fulcrum pleaded that it was entitled under clause 16(c)(v) to terminate the agreement with immediate effect.  For the same reason, it denied that its termination constituted a repudiation of the agreement. 40. As to the particular allegations of breach, Fulcrum admitted that it had an obligation timeously to co-operate by providing Markit with information necessary to enable the latter to comply with its obligations.  However, it pleaded that Markit had an obligation to solicit this information and that it had failed to do so.  Fulcrum denied that it was indebted in the amounts claimed. 41. In its counterclaim, Fulcrum pleaded that it had paid Markit R6 million by way of the first and second instalments under the agreement.  Consequent on its cancellation of the agreement, Fulcrum had suffered the following damages: 1. the amount of R1 500 000,00, being 50% of the first instalment which was paid to Markit despite the fact that there was neither agreement in respect of the BRD nor a completed BRD at the time of the payment; and 2. the amount of R3 000 000,00, being the second instalment, which was paid to Markit based on its misrepresentation that it had commenced with the build of the system, which later transpired not to be the case. 42. After a lengthy trial, the court a quo dismissed Markit’s claim and upheld Fulcrum’s counterclaim.  It identified two core questions for determination.  First, did the parties fail to reach agreement in respect of the details to be documented in the BRD? Second, and if so, did this failure to reach agreement occur as a result of a breach by Fulcrum.  The court a quo noted that these questions required an interpretation of the relevant clauses in the agreement. 43. On the first question, the court a quo found that the evidence established that there had been no agreement between the parties on the details to be documented in the BRD: the PID was not the BRD, nor was the Product and Entity sub-BRD (which Markit had only conditionally approved) sufficient to meet the requirements laid down in the agreement.  As the court a quo noted in its judgment, [1] to date Schedule 2 of the agreement, which was the envisaged BRD, remained a blank document.  There had thus been no agreement on the details to be documented in the BRD despite the lapse of a reasonable period of time. 44. As to the question of whether the fault for this could be laid at Fulcrum’s door, the court a quo concluded it could not.  Under clauses 2 and 2(a) of the agreement Markit had a positive obligation to analyse and document Fulcrum’s business requirements, whereafter they would be included in the BRD.  It followed from these provisions, and as the party with the necessary expertise and skill, that Markit had to take the lead on this first step.  Markit had the primary obligation to solicit the relevant information from Fulcrum in order to analyse and document its business requirements for purposes of the BRD. 45. The court a quo concluded that the failure to reach agreement on the details to be included in the BRD did not result from any breach by Fulcrum of its obligations under the agreement, but rather from Markit's failure to comply with its obligation to analyse and document Fulcrum's business requirements.  The court a quo upheld Fulcrum’s counterclaim on the basis that post cancellation, all payments made by Fulcrum to Markit should be refunded under clause 19(c)(iii).  It ordered Markit to pay Fulcrum a combined sum of R4,5 million. ON APPEAL 46. The crux of Markit’s appeal is that the court a quo erred in finding that Markit had failed to discharge its onus of establishing that there were no lawful grounds for cancellation under clause 16(c)(v) read with clause 2(d).  According to Markit, the right to terminate under that clause is triggered not simply by the absence of consensus on the details to be documented in the BRD, but by a demonstrated inability or failure by the parties to reach consensus.  Markit contends that the parties were on their way to reaching consensus and that they would have done so had it not been for Fulcrum’s unlawful termination.  For this reason, it is Markit’s case that the court a quo erred in concluding that Fulcrum was lawfully entitled to rely on clause 16(c)(v) to terminate the agreement.  Fulcrum’s conduct amounted to a repudiation and the court a quo ought to have reached that conclusion. 47. Moreover, says Markit, properly interpreted, clause 16(c)(v) is not available to a party that is in breach of its own obligations under the agreement.  Markit contends that Fulcrum was in breach of its obligations in that it failed to co-operate with Markit and provide the information necessary to enable Markit to analyse and document Fulcrum’s business requirements, which was a precursor to the completion and agreement of the BRD.  Fulcrum also failed to appoint a business analyst until October 2017 which, Markit says, was in breach of its obligations under clause 4(b).   Further, Markit makes the case that it was Fulcrum who bore the obligation of producing the BRD and any failure or inability to reach consensus on the details to be included in it lie at Fulcrum’s own door. 48. Fulcrum’s case is that its termination of the agreement was based on clauses 2(d) and 16(c)(v) of the contract. These clauses provided that if the parties failed, or were unable, to agree on the BRD either party could cancel the agreement immediately: this is what Fulcrum legitimately did. These cancellation clauses were designed specifically to allow either party to cancel the agreement if they could not agree on the BRD.  It was common cause that no BRD had been agreed.  According to Fulcrum, this meant that Markit had to satisfy the court that despite the non-existence of a BRD, consensus nonetheless would have been reached.  Fulcrum contends Markit failed to discharge this onus and the court a quo correctly dismissed its claim. 49. As to the question of whether reliance on clause 16(c)(v) is impermissible for a party that itself is in breach, Fulcrum submits that this interpretation cannot be supported.  However, even if that interpretation is adopted, Fulcrum contends that Markit failed to discharge the onus of establishing such breach. 50. As I see it, to succeed in this appeal, Markit must satisfy the court that the following issues should be determined in its favour: 1. The right to terminate under clause 2 read with clause 16(c)(v) is not triggered by the mere absence of an agreed BRD, but by a demonstrated inability or failure to agree. 2. This being the case, on the facts, and despite the absence of an actual BRD, there was no inability or failure to agree. 3. Had it not been for Fulcrum’s conduct in terminating the agreement, the parties would have reached consensus on the details to be documented in the BRD within a reasonable time.  Thus, clause 16(c)(v) was not triggered and Fulcrum’s termination of the agreement amounted to an unlawful repudiation. 4. Alternatively, even if clause 16(c)(v) was potentially triggered, it was not available to Fulcrum. 5. This is because, on a proper interpretation, this clause cannot lawfully be used by a party that is itself in breach of its obligations under the agreement. 6. Fulcrum had breached its obligations and thus its reliance on clause 16(c)(v) was unlawful. 7. For this reason, too, its conduct amounted to a repudiation and Fulcrum must be held liable to Markit for its contractual damages. 51. I realise that there may well be some overlap between the issues demarcated above, but I believe they provide a useful roadmap for the appeal. Was there an inability or failure to agree on the details to be documented in the BRD? 52. On the first issue, clause 2 permits termination if the parties are ‘unable’ to reach agreement on the details to be included in the BRD.  Clause 16(c)(v) uses the word ‘fail’.  For purposes of this judgment, I accept that something more than the mere absence of an actual BRD is required before it can be concluded that the parties failed or were unable to agree on the details to be documented therein. 53. The requirement that there should be a failure to reach consensus on a BRD before a party is entitled to terminate must logically include a temporal element if it is to have any practical effect.  In other words, a party could hardly claim in week two of the project that because there is no BRD yet in existence the requirement of an ‘inability’ or ‘failure’ to agree has been satisfied.  One must understand the clause to mean that the failure or inability should manifest within a reasonable time. 54. Obviously, a document recording an agreed BRD would be evidence of consensus having been reached.  The absence of a documented BRD is trickier.  At least at a prima face level, the absence of a BRD constitutes evidence that the parties failed or were unable to reach the required consensus.  However, as noted earlier, the reasonable time qualifier must be factored into the inquiry: it may be that, given time, a BRD would have been produced, and so clause 16(c)(v) was not triggered.  Conversely, if the parties did not manage to produce the agreed BRD within a reasonable time the logical inference to draw is that they failed, or were unable, to do so.  Markit’s case requires us to consider these issues. 55. One needs to consider, first, what would be a reasonable time?  Markit submitted that, on the facts of this case, the clock only started ticking in October 2017, when Mr Mahabeer was brought on board by Fulcrum as its business analyst.  On this approach, less than three months had elapsed before Fulcrum terminated.  Markit says that this was too short a period within which to conclude that the parties had failed to reach consensus on the BRD. 56. Markit’s submission is not supported by the evidence.  As I described earlier, it was recorded that both parties accepted that they had resource constraints in the early days of the project.  As a result, both parties made use of their existing in-house resources to fill existing gaps.  While Fulcrum undertook, as one of its obligations, to appoint a business analyst to assist Markit in the BRD formulation and completion, this did not prevent the parties from fulfilling their assigned tasks in the interim.  Markit never complained to Fulcrum in the earlier stages of the project that it could not commence analysing and documenting Fulcrum’s business requirements because of the absence of an appointed business analyst.  In any event, Ms Mclean was appointed in August 2017 to take on the role of business analyst before Mr Mahabeer came on board in October.  Markit had access to the Flexibroker system from May.  Steerco meetings were being held regularly and the absence of an appointed business analyst did not, on the evidence, prevent the parties from working on the BRD much earlier. 57. The parties originally envisaged that the BRD process would be completed at the beginning of July 2017.  This can be gleaned from the terms of payment of the second instalment.  It was to be released to Markit on the commencement of build, in other words, after consensus on the BRD, on 1 July.  This indicates that the parties envisaged a relatively short first stage of the project, culminating in the concluded BRD.  Despite the initial delays in the project getting off the ground, I am not satisfied that Markit is correct when it says that the clock only started ticking in October.  The evidence indicates that it was possible for this process to have commenced earlier.  As I discuss below, it was not the delay in Mr Mahabeer’s appointment, but rather the culmination of several factors that obstructed the production of the BRD. 58. What of Markit’s related contention that the parties could still have reached consensus on the BRD within a reasonable time had Fulcrum not delivered its termination notice?  Mr Berridge, relied in particular in this regard on two aspects of the evidence.  The multi-coloured email exchange between Mr Walters and Mr Mahabeer in November 2017, and Mr Schreuder’s alleged concession under cross-examination that consensus would have been reached. 59. Mr Berridge submitted that the multi-coloured email showed that the parties were making real progress on the BRD, the implication being that consensus was not far off.  He pointed, too, to Fulcrum’s timeline of what other BRD’s were in production and could be anticipated.  Mr Berridge suggested that the evidence shows that the parties were on the verge of reaching consensus on the BRD.  Again, the evidence does not support this submission. 60. It is important to bear in mind here that in the preceding months, and without apparent regard to the contract, there had been a shift from the contractual requirement of a single BRD to a list of approximately 20 anticipated sub-BRDs.   As at the time of the exchange of emails between Mr Mahabeer and Mr Walters, all that had been achieved was conditional acceptance by Markit (after approximately a month of consideration) of the draft of the very first of these sub-BRDs, with several caveats prescribed.  The same situation pertained at the time that Fulcrum terminated the agreement.    It simply cannot be inferred from this state of affairs that the parties were anywhere near reaching consensus, let alone that the probabilities are that this would have occurred within a reasonable time. 61. It is also apparent from the exchange of communications detailed earlier that the parties were not on the same page on some fundamental issues.  One example is Ms Oakley-Brown’s response to Mr Walter’s 29 November email, questioning whether the project truly was based on an ‘agile’ approach.  She voiced the lack of clarity and discomfort between the parties which she hoped might be resolved.  Mr Mahabeer’s response to the same email, set out earlier, also demonstrates and expresses a lack of common understanding between the parties.  It is plain from these exchanges that there were fundamental points of disjuncture in the respective understandings of the project by the parties.  Demonstrably, they were not at that stage yet working in concert on the most basic aspects of the project.  For this situation to have existed as late as the end of November is a further indication that Markit’s contention that the parties were on the path to consensus on the BRD is erroneous. 62. Markit’s reliance on the alleged concessions by Mr Schreuder and Mr Hands do not rescue its case.  It is so that these witnesses agreed with Mr Berridge’s proposition under cross-examination that consensus between the parties was not impossible.  However, this agreement was qualified in an important respect.  Mr Schreuder stated that given enough time the project could have been realised, but that what was critical was whether this could have been achieved in a reasonable time.  I have already given an indication that in my view, the probabilities in the case are that this would not have been achieved.  In any event, in my view, clauses 2(d) and 16(c)(v) prescribed an inability or failure to reach consensus, not an impossibility to do so. 63. There is a further significant element arising from the evidence in this case which to my mind puts paid to Markit’s contention that there was no inability or failure to reach consensus on the details to be documented in the BRD.  That element is the breakdown of trust that occurred from when Fulcrum first began to question whether Markit actually had commenced the system build; to the letter of 6 December 2017 from Fulcrum demanding proof, and the immediate aftermath; and culminating in the notice of termination on 13 December. 64. Mr Schreuder, Ms Oakley-Brown and Ms Bouic all testified to this breakdown of trust.  One of the triggers for this was Markit’s invoice for payment of the second instalment even though the BRD was not yet in place.   The BRD was the first building block of the project, and completion of the BRD was the condition precedent for payment of the remaining 50% of the first instalment.  The next step was the commencement of build, which was the condition set for payment of the second instalment under clause 19(d).  Fulcrum was concerned that without the BRD the build could not have commenced.  This led to the 24 August meeting after Markit had issued its invoice for payment of the second instalment.  Markit assured Fulcrum that it had commenced despite the absence of a BRD.  Part of the explanation was that with the adoption of a ‘hybrid’ agile project approach, as opposed to a traditional waterfall approach, it was possible to commence the build without the necessity for an agreed BRD.  After discussions between the parties, Fulcrum’s management authorised payment of the second instalment.  Markit undertook to demonstrate the build at the workshop at the end of August. 65. As I have detailed above, neither Mr Schreuder nor Ms Oakley-Brown were impressed by the demonstration at the August/September workshop: they noted that it was nothing more than a demonstration of an existing system for another client which they had seen before.  Ms Oakley-Brown’s evidence was that she recommended to Mr Bain that the second instalment should not be paid as she did not believe Markit had demonstrated that the work had been done. 66. Mr Berridge made much of an email sent by Mr Mathee, a Fulcrum subject matter specialist, to Markit after the workshop.  It read: ‘ Dear Mr Malcolm Just want to thank you and your awesome team for such a productive week, clearing up almost all uncertainties, and really appreciate building a long lasting working and personal relationship, and more importantly friendship! Let me know if there is anything you need from me to ensure that this project is one hell of a success. See you at the end of October! Kindest regards to Roger and the rest of our friends from India. Looking forward to what the future holds for both our companies. Remember to let me know when Simon next in Cape Town. Cheers bud’ Mr Berridge’s submission was that such effusive feedback from one of the workshop attendees put paid to Fulcrum’s contention that it was not happy with the progress of the project and with Markit’s input at that stage.  As Ms Oakley-Brown pointed out in her evidence, Mr Mathee’s role at Fulcrum was as a subject specialist who understood the workings of the Flexi system.  He did not occupy the kind of position that gave him the authority to speak on behalf of Fulcrum as to its satisfaction with Markit.  Indeed, if one considers the tone of the email, it is self-evidently a ‘chummy’ communication from someone who attended the workshop and was excited about the contacts that he had made and about the prospects of an ongoing working relationship.  This is crystal clear from the sign-off: ‘Cheers bud’.  It was clearly written by Mr Mathee in his personal capacity and simply cannot be read as reflecting the formal view of Fulcrum.  The letter does not materially advance Markit’s case. 67. Fulcrum’s lack of trust was undermined further by Ms Oakley-Brown and Ms Bouic’s meeting with Infiniti.  Then there were the two emails from Mr Walters and Mr Mclean in November 2017 suggesting that Markit was only at that stage moving towards the build phase of the project.  This raised alarm bells for Fulcrum, based on the previous understanding it had been given that the build had already commenced.  It precipitated the letter of 6 December demanding proof from Markit of what it had done to date.  The demonstration on 13 December did not allay Fulcrum’s concerns and, from its side, it had lost trust in Markit’s ability to deliver. 68. Under cross-examination Mr Schreuder described the situation as follows: ‘ I would say that, by the time we got to mid to end November, that is when my concerns were realised. The data segregation approach was essentially not going anywhere, we hadn't got to any form of agreement as to how to segregate the data, which was an absolute foundational pivotal part of the puzzle. When Mr Roger Walters had indicated to us that build had not commenced, we - and that we were probably, what's it, eight months into the project, and we did not as yet have a signed off and agreed high level BRD and/or a signed off and agreed detailed BRD of any sort - that, to me, was when I believed that this project became untenable.’ And further: ‘ And throughout the communication we have seen throughout the Steercos that there were comments about build commencing and progress being made, and when we got to November and we received emails to the likes of first build environment and build can now commence, and at that point we did not have as yet a single BRD which had been agreed, a detailed BRD that had been agreed, and at that point we didn't have even a high level BRD that had been signed off and agreed, I don't believe. … And I think at that point we had lost trust in the partner, and we asked them to demonstrate the work that had been done to date, which was not demonstrated and hence led to the termination.’ 69. In re-examination he said that: ‘We saw no evidence on build. I suspect that given an environment where the two partners no longer trust each other, I am not sure if any length of time may have healed that.’ 70. Fulcrum submitted that it was this breakdown of trust that was the final nail in the coffin of consensus being reached on the details to be included in the BRD.  Its case is not that Markit breached the agreement by failing to commence the build.  Rather, it is that Fulcrum lost trust in its project partner, and it was this that ultimately caused the failure or inability to reach consensus on the BRD, justifying its cancellation under clause 16(c)(v). 71. Mr Berridge was at pains to point out to Fulcrum’s witnesses, and to the court, that Fulcrum did not specifically raise each of these concerns contemporaneously with Markit.  Consequently, says Markit, the breakdown of trust averred by Fulcrum was nothing more than an ex post facto excuse in an attempt to justify its unlawful cancellation of the agreement. 72. It is so that each concern Fulcrum says it harboured was not raised with Markit on an ongoing basis.  However, this is not an indication that the concerns did not exist and that they were manufactured ex post facto to justify cancellation.  Fulcrum pertinently raised its most important concerns with Markit.  It aired its concerns at the meeting on 24 August about whether the payment of the second instalment was justified; and by 6 December Fulcrum placed a detailed list of demands at Markit’s door.  Ms Oakley-Brown and Ms Bouic testified that Fulcrum’s concerns were discussed within its own circle.  Despite these concern’s Fulcrum’s witnesses testified that it wanted the project to succeed and so Fulcrum pressed ahead until the emails from Mr Walters and Mr Mclean confirmed, for Fulcrum, its fears. The chronology of events, when seen as a whole, demonstrates an erosion of trust, on a long fuse from the end of August, until the ignition point at the end of November to mid-December.  This was not a case of an ex post facto- manufactured breakdown of trust. 73. What would have dealt a blow to Fulcrum’s breakdown of trust explanation would have been for Markit to have produced evidence at the trial to show that Fulcrum’s fears were unfounded: that it had in fact commenced the build, as it had represented to Fulcrum.  Critically, it did not produce any codes, or other evidence at trial to support its case that the build had commenced.  When Mr Maclean was pressed on this under cross-examination, he explained that: ‘ It is accepted business practice that you do not have redundant or unused parts of systems within your various environments. It is a problem in Cloud systems that invites hacking attacks or other exploits. We have a fairly strict approach to these where they are deleted.  Now in addition to that. In every contract that I have ever been a party to in IT terms, when a contract is ended we are under an obligation to destroy or all relevant materials. So if we would have kept this, no doubt we would have been accused of not destroying relevant materials. So, you know, we are caught between two stools here. ’ 74. When he was asked why, knowing that Markit’s work had been called into question in December 2017, it had not stored a demonstration version of what had been built so that, at trial, it could demonstrate that it was entitled to its damages, Mr Mclean responded that: ‘I do not think it was practical. … [I]t was impractical for us to support that.’  He later said that Markit did not immediately delete the demonstration and that it had been kept for a period of time.  It was not produced as evidence at trial and Mr Mclean’s explanation for this raises more questions than provides answers.  The demonstration was on 13 December.  The letter of cancellation came that same day, and litigious letters immediately were exchanged between the parties in the days and weeks following.  The absence of a rational explanation for why, despite the high risk of litigation, Markit deleted evidence of the build supports the conclusion that Fulcrum’s concern that the build had not commenced was not far-fetched, nor unreasonable. They were real concerns which led to a real breakdown of trust.  Once trust was broken, consensus on the BRD was out of the question. 75. Taking all these factors into account, I conclude that Markit failed to establish on a balance of probabilities that the requirements for cancellation under clause 16(c)(v) were not met.  There was an inability or failure to agree on the details to be included in the BRD, which triggered clause 16(c)(v). 76. As discussed earlier, Markit’s case is that this finding is insufficient to protect the court a quo’s judgment on appeal.  This is because even if there was no consensus on the BRD, clause 16(c)(v) could not be used by Fulcrum to cancel the agreement because Fulcrum was in breach of its obligations. Was Fulcrum precluded from relying on clause 16(c)(v) because of its own breach? 77. Markit’s case on this issue is that clause 16(c)(v) must be interpreted to mean that a party that is in breach of its own obligations cannot use this clause to cancel the agreement.  It is not entirely clear to me that Markit’s interpretation of clause 16(c)(v) is correct.  It seems to me that it is arguable that the clause was intended by the parties to be a true ‘no fault’ clause.  That its purpose was to provide a mutual exit clause for either of the parties in the event that they could not see eye-to-eye on the most fundamental building block of the project, the BRD, without fault playing any part in the process at all.  If one party was in breach, it would always be open to the other to resort to the breach provisions set out separately in clause 16. 78. However, in view of the conclusion I reach on the issue of whether Markit established that Fulcrum was in breach of its obligations, it is not necessary to make a finding on the proper interpretation of clause 16(c)(v).  I will proceed on the assumption (without making a finding) that Markit’s interpretation is correct and that, if a breach of its obligations on the part of Fulcrum was established, this would preclude Fulcrum from relying on the cancellation clause. 79. To succeed in its appeal, Markit must establish on a balance of probabilities that Fulcrum defaulted on its obligations.  The first breach relied on by Markit was Fulcrum’s alleged failure to co-operate with Markit by providing Markit with the necessary information to enable Markit to analyse and document Fulcrum’s business requirements for inclusion in the BRD.  Allied to this aspect of Markit’s case is that it was Fulcrum, rather than Markit, that had the obligation to produce the BRD. 80. What the agreement laid down in clause 2(a) was that: ‘Once [Markit] has analysed and documented [Fulcrum’s] business requirements, both parties shall jointly agree on … [a] BRD which shall set out what [Fulcrum’s] requirements are subsequent to a detailed analysis of [Fulcrum’s] business by [Markit].’  A plain reading of this clause places the primary responsibility on Markit to undertake a detailed analysis of, and to document, Fulcrum’s business requirements.  It makes it clear that the purpose of this exercise was to ensure that these business requirements would be included in the BRD.  Clause 3(a) repeated, and underlined Markit’s responsibility in this regard. 81. Markit was not expected to undertake this exercise on its own.  Logically, it required Fulcrum’s assistance, as it was the latter’s business requirements that had to be analysed and documented in the BRD.  Clause 3(a) expressed Fulcrum’s responsibilities: it acknowledged that Markit’s ability to provide the services was dependent upon Fulcrum’s full and timely co-operation, and it agreed so to cooperate. 82. The agreement did not expressly record which party was responsible for producing the BRD.  However, from the above-mentioned clauses, it may be inferred that it would be a joint product emanating from the cooperation between the parties, with each fulfilling their designated responsibilities.  It was, after all, a key requirement of the agreement that there would be consensus on the details to be included in the BRD, failing which, either party could terminate. 83. What is clear, as I have said, is that it was Markit’s responsibility to analyse and document the business requirements.  The terms of payment outlined in the agreement underscore this.  Markit was entitled to payment of the retained 50% of the first instalment on agreement of the BRD.  In other words, it would be recompensed at that stage for the work it had put into analysing and documenting the details of Fulcrum’s business requirements for inclusion in the BRD. 84. The context in which the agreement was concluded is also important.  Fulcrum had an antiquated system which it wished to substitute with an updated and expanded system developed by Markit.  Markit had existing components or modules it could use as a starting point, and it had the IT knowledge and expertise to harness what it had and further develop what needed to be developed to suit Fulcrum’s needs.  Fulcrum understood its business requirements, but it did not have any knowledge of Markit’s modules.  Fulcrum’s expert witness, Mr Hands, put into perspective how this context shaped the parties’ respective responsibilities as regards the BRD.  He said: ‘ I am looking at this purely as a project manager, and I believe it is the case that producing that document, was under the care and control of Markit, and responsibility for the completeness and accuracy of that document, was under the care and control of Fulcrum. Between the two they had to cooperate, but Markit was in the driving seat, by virtue of it being responsible for producing the BRD.…it is good practice for the supplier to determine the format and terminology of the BRD, because it has to be used subsequently by the supplier's development people who are not involved at this stage.’ 85. Mr Hands was asked what the implications for the project would be of leaving it to the client, Fulcrum, to complete the BRD.  He responded: ‘ Well there is the risk that the client would be producing work that may not have been necessary to achieve the 80/20 advantage that we were speaking of because …  it is the supplier who should be looking at the customer's requirements and if the supplier has existing software in its arsenal that can meet those requirements there is not much point in waiting for the customer to scribe them all out in detail if the software to do that already exists. So what I am saying is that for the supplier to just (sit) back and wait to be served up with all these detailed BRD documents by the customer, risks the completion date of the project. It risks a lot of unnecessary work done, because the customer's work is not being properly informed of by knowledge of what the supplier has already got in its software arsenal … .’ 86. Mr Hands is an expert in the field of the project management of IT contracts, and particularly on the BRD aspects of those contracts.  His evidence has been accepted in courts in the United Kingdom.  Mr Berridge was disparaging of Mr Hands as an expert witness and in the manner in which he had prepared for his testimony.  He was criticised for not having interviewed the various role-players and for adopting a biased view on which party was to blame for the project’s demise. 87. The trial court was not, and this court is not, bound in any way by Mr Hands’ opinion on where blame lay for the failure of the project.  I do not believe it is necessary to weigh in on that particular debate between Mr Berridge and Mr Morison.  However, Mr Hands’ evidence was of obvious value to the court in explaining how IT contracts work in general, explaining the terminology involved, the importance of the BRD in such contracts and on the question of what best practice in contracts of this nature is.  From this perspective, I find no reason to differ from the court a quo in accepting his evidence that, generally, it is the supplier, with its arsenal of IT expertise at its disposal, that must drive the process towards recording the client’s business requirements in the BRD.  In this particular case, his view also makes logical sense because Markit already had modules it had developed which the parties understood would be put into use in developing the new system for Fulcrum. 88. Fulcrum’s other witnesses backed up Mr Hands’ evidence in this regard.  According to Ms Oakley-Brown: ‘ It is very difficult for Fulcrum to manage to do those BRD's in a quick and efficient way, if they do not know how the Markit system works. So for them to try and align what is required for Markit is time consuming, which goes to Markit. You needed to have documented and done those BRD's. It is your product; it is your system. You know and understand; you need to solicit the information from the team.’ 89. Under cross-examination, it was put to Mr Schreuder that it was Fulcrum’s responsibility to provide information to Markit and that until that was done, Markit could not fulfil its obligation to analyse and document.  As Mr Berridge put it: ‘I can only analyse what I have been given. Self-evident. Correct?’  Mr Schreuder responded: ‘ That is correct. But in terms of the contractual arrangement, Markit being the vendor and Fulcrum being the customer, the vendor had an onus to own that process and to drive that process, for which they had not. They had sat back and waited for Fulcrum to deliver information, with no input or request or direction that we would have expected from a service provider with their skills and experience.’ 90. Markit’s witnesses did not have much to say to countervail this testimony from Fulcrum’s witnesses.  Mr Mclean admitted that he did not know about the significance of a failure to reach consensus on the BRD for the agreement as a whole.  An exchange between Mr Morison (for Fulcrum) and Mr Mclean during cross-examination bears this out: MR MORISON: You have indicated in your witness statement that you were responsible for compliance with the contractual obligations of Markit. Is that correct, Mr McLean’ ? MR MCLEAN: From an operational point of view, yes. MR MORISON: So were you aware that there was a clause16(c)(v) that permitted either party to cancel the contract immediately if the parties failed to agree on the BRD? MR MCLEAN: I was not aware of that clause until the cancellation was made.’ 91. It is difficult to avoid the inference that the BRD was not regarded as a top priority for Markit.  This is further demonstrated by the shift, driven by Markit, to multiple sub-BRDs, and the shift towards a hybrid/agile project model.  The significance of the latter is that Markit could report that it was working on the system build despite the non-existence yet of a BRD.  As Mr Hands explained, on the traditional ‘waterfall’ model, the BRD would have to be in place before build could commence, as it formed the blueprint for what would be built.  If Markit did not regard the BRD to be of critical importance, which this evidence demonstrates, it is difficult to avoid the conclusion that, as Fulcrum’s witnesses attested, Markit did not drive the process, as it was enjoined to do, and instead left Fulcrum to do the running. 92. It is common cause that ultimately Fulcrum started drafting detailed sub-BRDs.  Its witnesses told the court that this was because Markit wasn’t doing anything to advance the process.  Mr Berridge submitted that by doing so, and consistent with what Mr Schreuder recorded in the PID, Fulcrum accepted that the BRD’s were its responsibility.  Mr Schreuder explained why this was not so: the ‘owner’ of the BRD being Fulcrum, as he had recorded it in the PID, simply meant that ultimately Fulcrum had to take responsibility to ensure that its business requirements were accurately recorded in the BRD.  Mr Schreuder’s explanation is consistent with the evidence discussed above.  It does not support the conclusion that the parties understood and agreed that Fulcrum was responsible for producing the BRD.  Such a conclusion would be contrary to what the agreement reflects and contrary to the weight of evidence adduced at the trial. 93. The only remaining point to consider on this issue is Markit’s complaint that Fulcrum did not comply with its responsibility to cooperate with Markit by providing it with the necessary information to enable it to analyse and document Fulcrum’s business requirements.  Much of the evidence I have already discussed above does not support Markit’s complaint.  The evidence shows that Markit did not appreciate the significance of reaching consensus on the BRD and that, contrary to its contractual obligations, it did not take the driver’s seat in progressing the process. 94. Fulcrum provided Markit with access to its Flexibroker system in early May.  There was some complaint by Mr Mclean in his evidence that Markit was not given access to the ‘data base layer’ of the Flexibroker system and did not have much access to subject experts.  However, he could point to no evidence indicating that Markit had pressed for better access.  As for as the subject experts were concerned, they were available at the August/September workshop. As to the Flexibroker access, Markit didn’t explain whether it gave its developers access to the system, when this was done and if so, what use the developers made of it.   It appears that whatever access and use they had made of it was minimal because Ms Oakley-Brown testified that at the August/September workshop she was surprised to find that the developers appeared to have very little knowledge of the Flexibroker system. 95. Once again, the evidence does not support Markit’s case that it was Fulcrum who held back on co-operating with Markit.  The point is worth repeating that it was Markit that had the primary responsibility to drive the information-gathering and documenting process: this is what they were paid to do.  Fulcrum’s responsibility was to assist which, according to the evidence, it did. From all of this I conclude that Markit failed to establish on a balance of probabilities that Fulcrum breached its obligations under the agreement in this respect too.  The court a quo cannot be faulted for reaching the same conclusion. 96. Consequently, the appeal in respect of Markit’s claim must fail. COUNTERCLAIM 97. The remaining issue is whether the court a quo erred in upholding Fulcrum’s counterclaim.  The claim identified two amounts: the first was the amount of R1,5 million, representing the payment of the second part of the first instalment; and the second was the amount of R3 million, representing the second instalment that was paid after the 24 August 2017 meeting. 98. Markit’s case on appeal is that the counterclaim was in the nature of a damages claim.  However, Fulcrum did not plead nor press a case of breach on the part of Markit.  Instead, it relied on clause 16(c)(v).  This being the case, the court a quo ought properly to have dismissed the counterclaim as no basis was laid for the awarding of damages.  Similarly, Fulcrum did not make out a case of misrepresentation which, according to Markit, was the basis for its claim in respect of the amount of R3 million. 99. The court a quo upheld the counterclaim, awarding a globular amount of R4,5 million.  The court a quo found that because Markit had failed to show that Fulcrum was not entitled to cancel the agreement under clause 16(c)(v), it followed that Fulcrum was entitled to a refund of the amounts paid thereunder in terms of clause 19(c)(iii). 100. Fulcrum submits that the court a quo was correct in upholding its counterclaim albeit that it may not have accurately described the nature of the claim.  Fulcrum says that in essence, it claimed restitutionary damages, based, not on a breach of the contract by Markit, but on the cancellation by agreement.  Fulcrum contended that where a contract is discharged by cancellation based on an agreed cancellation clause, each party’s performance should be restored to the pre-contractual position.  In support of its case, Fulcrum referred to the authors Christie and Bradfield, and their exposition of the general principles applicable on the discharge of a contract. [2] Where, as in this case, a contract is discharged after there has been part-performance, difficulties can occur.  The authors refer to such a situation as a ‘half-cooked omelette’.  To avoid problems arising from the undercooked product, the parties should make express provision for what the consequences should be.  Failing this, and failing any questions that are left open despite an attempt to regulate these consequences, resort must be had to general principles. 101. The authors go on to explain that the first of these principles is that an agreement to discharge a contract is presumed to include a tacit agreement to restore what has been delivered in part performance.  This would include payments made for goods or services, with the corollary being that any goods received against payment should also be restored.  The second principle is that either party can recover from the other the contra presentation for those portions of the contract that she has performed.  So, for example, they say, on cancellation of a lease agreement, rental would remain payable for the period of occupation. 102. On these principles, the court a quo correctly understood the nature of the counterclaim not to be for damages arising out of a breach of contract.  The true nature of the counterclaim was for restitution, consequent on the cancellation of the agreement, of what had been performed by Fulcrum.  Where the court a quo did err in its approach, however, is that it did not consider each claim separately.  It ought to have done so.  As is apparent from what appears below, however, this error is not material to the outcome of the appeal. 103. As regards the first claimed amount of R1,5 million, although this is described as a ‘damages’ claim in the introductory sentence of paragraph 77 of Fulcrum’s plea and counterclaim, the basis pleaded for the payment of this sum is that it was paid to Markit ‘despite the fact that there was neither agreement in respect of the BRD nor a completed BRD at the time of the payment’.  The court a quo correctly found that Fulcrum was entitled to cancel under clause 16(c)(v) because there was an inability or failure to reach consensus on the BRD.  Clause 19(c)(iii) accordingly came into play.  It provided expressly that: ‘Should this agreement be terminated in accordance with clause 16, due to non-compliance with the BRD or failure to reach agreement on the BRD; the balance of the initial deposit shall thereafter be paid over to the [Fulcrum].’  Whether one calls it a claim for damages as provided for in the agreement, or restitution, as provided for in the agreement, Fulcrum was entitled to repayment of R1,5 million, being the 50% of the first instalment that only fell due on production of the BRD.  Ultimately, the court a quo was correct in upholding its counterclaim in this amount. 104. As regards the claim for the balance of R3 million, this payment was due on commencement of the build.  Fulcrum averred that it was entitled to repayment of this amount because Markit misrepresented that the build had commenced when it had not.  Fulcrum had no automatic right under clause 19(c)(iii) to repayment of this amount, unlike its express right to repayment of 50% of the first instalment.  The parties thus fell into the trap of making provision for one consequence of cancellation but not all: clause 19(c)(iii) is silent on whether repayment of the build-related amount should also be repaid.  It may be that it was assumed by the parties that there would be no build, and hence no justification for payment of the second instalment of R3 million, in the absence of an agreed BRD.  We know, however, that this is not what occurred.  Fulcrum paid the R1,5 million balance of the first instalment and the R3 million second instalment despite there being no BRD.  It paid the latter amount on the basis that the build had commenced because, under the hybrid/agile approach, a finalised BRD was not a necessary precursor to commence the build. 105. In the absence of express provision, what is the fate of the R3 million payment, given the fact that the agreement was cancelled without full performance on both sides?  The general principle supports a default position of restitution.  It may be countered, however, that under clause 16(d) payment of the R3 million was dependent on commencement of build.  Thus, if Markit indeed commenced the build, it had performed under clause 19(d), and it is entitled to retain this amount. 106. I have already discussed at some length the evidence supporting Fulcrum’s suspicions that the build had not commenced despite Markit saying that it had.  If Fulcrum is to establish its claim for repayment of the R3 million, it must satisfy the court that the evidence supports its conclusion that despite Markit’s assurances, the build had not commenced.  I do not intend to traverse, again, the relevant evidence.  I have already concluded that Fulcrum’s concerns that Markit had not commenced the build were reasonable.  It asked for a demonstration at the August/September workshop.  There is no evidence that what Fulcrum was shown was a new build of any sort: it was again shown an existing system that it had seen before.   At the time payment was made, there is thus no evidence that build had commenced. 107. On 13 December, the second requested demonstration was given to Fulcrum.  Ms Oakley-Brown reported on the demonstration in her email, discussed earlier.  Her description there is consistent with her evidence to court: ‘ … what I viewed was a test environment, yes; but its currently set for London market not for us. Product conflict not set up as yet. Currently populated with Lloyds risk codes. Lloyds have got nothing to do with South Africa. The test environment has been branded with Fulcrum; that is fluff. Roger indicated that they could provide us with access and codes, but there was not much there as yet, and as a result was very reluctant to let us play and access the system. By now, based on what they have been telling us, they should be happy to let us have access to that system ... because it should have been a lot further long based on what they had been saying over the past 3 months. ’ 108. Mr Schreuder testified that: ‘ I have seen no evidence that there was any meaningful build, demonstrating that any input from Fulcrum was turned into a viable product. Or not even viable, just any form of demonstrable product, or even a component of product. So without that, we find it difficult to believe that there was any meaningful build done.’ 109. This evidence is sufficient to shift an evidentiary burden onto Markit to show that their assurances were correct and that, when payment of the August invoice was made, the build had indeed commenced.  As noted earlier, Markit was unable to produce any evidence to substantiate its claim.  The probabilities favour Fulcrum’s version: Markit’s assurances that the build had commenced despite the absence of a BRD, and therefore, that payment of the second instalment was due under the contract were not true.  It had not commenced the build.  Consequently, Fulcrum was entitled to restitution of the R3 million it had paid to Markit by way of the second instalment. 110. For these reasons, the appeal against the counterclaim must also fail. COSTS AND ORDER 111. There is no reason to make a costs order different to that usually made when an appeal is dismissed. 112. I make the following order: 1. The appellants appeal is dismissed with costs, including the costs consequent upon the employment of two counsel, one being a Senior Counsel. R M KEIGHTLEY JUDGE OF THE HIGH COURT I agree L WINDELL JUDGE OF THE HIGH COURT I agree SC MIA JUDGE OF THE HIGH COURT Delivered:  This judgment was prepared and authored by the Judge whose name is reflected and is handed down electronically by circulation to the Parties/their legal representatives by email and by uploading it to the electronic file of this matter on Case Lines.  The date for hand-down is deemed to be 05 MAY 2023 APPEARANCES COUNSEL FOR APPELLANT ADVOCATE   B BERRIDGE SC APPELANTS’ ATTORNEYS CLYDE & CO INCORPORATED COUNSEL FOR RESPONDENTS ADVOCATES L MORRISON SC & N MNCUBE RESPONDENTS’ ATTORNEYS NICQUI GALAKTIOU INC DATE OF HEARING: 15 MARCH 2023 DATE OF JUDGMENT: 05 MAY 2023 [1] Judgment court a quo , para 13. [2] R H Christie & G B Bradfield Christie’s The Law of Contract in South Africa (6ed) p 462-3 sino noindex make_database footer start