AAH Pharmaceuticals Ltd & Anor v Jhoots Healthcare Ltd & Anor

Neutral Citation Number: [2020] EWHC 2524 (Comm) IN THE HIGH COURT OF JUSTICE BUSINESS AND PROPERTY COURTS IN BIRMINGHAM CIRCUIT COMMERCIAL COURT (QBD) Case No: CC-2020-BHM-000023 The Birmingham Civil Justice Centre Priory Courts, 33, Bull Street, Birmingham B4 6DS Date: 23 September 2020 Before : HHJ WORSTER - - - - - - - - - - - - - - - - - - - - - Between : (1) AAH Pharmaceuticals Limited (2) Barclay Pharmaceuticals Limited - and - (1) Jhoots Healthcare Limited (2) Pasab Limited Claimants Defendants - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Thomas Croxford QC (instructed by Addleshaw Goddard) for the Claimants Mark Grant (instructed by Harrison Clark Rickerbys) for the Defendants Hearing date: 16 September 2020 - - - - - - - - - - - - - - - - - - - - - Approved Judgment I direct that pursuant to CPR PD 39A para 6.1 no official shorthand note shall be taken of this Judgment and that copies of this version as handed down may be treated as authentic. ............................. HHJ WORSTER HHJ WORSTER Approved Judgment HHJ WORSTER: AAH v Jhoots and anor 1. 2. 3. This is an application by the Defendants made by notice dated 25 August 2020. The draft order with the application seeks the variation of the disclosure order made in this case. The application is supported by a witness statement from Mr Morgan, the Defendants’ solicitor of 25 August 2020, and opposed in a witness statement from Mr Kamstra, the Claimants’ solicitor of 14 September 2020. Much of their evidence reflects the correspondence between their respective firms over the last few months. I heard the application last week, but a point of principle arose in reply which needed some further consideration. The parties filed supplemental written submissions on Friday. References to page numbers in square brackets are to the pages of the application bundle. The Claimants are wholesalers of pharmaceutical products and have been selling pharmaceuticals to the Defendants (who dispense them from their stores) since 2009. In May 2019 a dispute arose between the parties which has led to this claim. The Claimants seek payment for goods ordered and supplied in the sum of c£1.66M plus interest. The Defendants counterclaim for c£1.3M after setting off the sums due to the Claimants. The major part of the counterclaim alleges that the Claimants breached an agreement to afford the Defendants discounts and rebates pursuant to its “Market Watch” scheme. The agreed case summary puts it this way: … the Claimants allege that they give a rebate to customers if the price of a typical basket of pharmaceutical products was available from the Claimants competitors at a lower price in the market at the time of purchase, in accordance with express contractual definitions. The Defendants contend that the contractual obligation of the Claimants was instead to price match each individual item purchased by the Defendants against the lowest price for that precise product anywhere in the market at the relevant time, causing loss. The other major issue is whether (or not) the Claimants’ various standard terms and conditions were incorporated into the parties course of dealing over the years. The case is subject to the requirements of PD51U (the Disclosure Pilot). In the run up to the CCMC on 15 May 2020, the parties engaged in the preparation of a Disclosure Review Document. There were 27 issues for disclosure. Issues 1-6 relate to the question of incorporation, and issues 7-27 relate to the discount and rebate issues. There are 8 custodians for the Claimants and 7 for the Defendants, and following the court’s orders to date and the parties agreement, some 47 keywords. There are potentially millions of documents. 4. At the CCMC on 15 May 2020, Waksman J made the following order about disclosure: 7. The Court approved the Single Joint Disclosure Review Document (“DRD”) in the terms agreed and subject to the following directions: HHJ WORSTER Approved Judgment AAH v Jhoots and anor He then made a number of more detailed orders. 7(a) dealt with the disclosure models to be adopted in relation to certain issues. In some cases this was to be in accordance with the DRD, and in some cases was by reference to Model D (with or without limitations). 7(b) required the Defendants to search the private email accounts and (in certain circumstances) the mobile phones of 2 custodians. It is the order at 7(c) which is the important one for the purposes of this application: The parties are to agree keywords, date ranges, custodians and such other matters of fine detail. In the event that no such agreement can be reached, the parties are to apply to the Court on paper. 7(d) and (e) related to the provision of native electronic documents, and narrative documents. 5. At paragraph 8 of the order of 15 May 2020 Waksman J directed that: The parties shall, by 5pm on 4 September 2020: a. b. make and serve on each other a Disclosure Certificate and an Extended Disclosure List of documents pursuant to paragraphs 12.1(1) and (2) PD 51U; and produce the documents which are disclosed over which no claim is made to withhold production pursuant to paragraph 12.1(3) PD51U, or (if the party cannot produce a particular document comply with paragraph 12.3 PD51U. The timetable which followed provided for the exchange of witness statements by 23 October 2020, the exchange of experts reports by 20 November 2020, an experts meeting and joint report by 18 December 2020, and a trial not before 8 March 2021. No trial has yet been fixed, but as I indicated to the parties at the hearing last week, the Court can accommodate a trial in May 2021. 6. The parties were able to agree many things about disclosure, but some significant differences remained. On 23 June 2020 [135] the Claimants wrote to the Defendants saying this: In common with our exchanges in respect of the pretrial timetable, the parties appear unable at present to agree various aspects of disclosure, and we do not consider that it is a viable investment of time and costs to engage in further protracted correspondence following a hearing over a month ago when disclosure harvesting and searches should already be underway to meet the suggested trial timetable. This is regrettable, as the parties should have been able to agree the approach to disclosure, and further delay has occurred which could have been avoided. We do not, therefore, propose to expend further time and costs by engaging in detail with repeated communications regarding disclosure. Rather we have set out below the parties’ final positions in respect of the aspects of disclosure HHJ WORSTER Approved Judgment AAH v Jhoots and anor that have not yet been agreed. In line with the approach of saving time and costs as far as possible, and reaching a determination quickly, we propose that the parties write to the Court requesting that Waksman J makes a decision in respect of the remaining issues outlined below. The letter then sets out the issues as to date ranges, the use of different keywords for different issues, and various issues about keyword search terms. The Claimants agreed that the matter should be referred back to the Court. 7. By his order of 15 July 2020 Waksman J determined the issues about date range, and “resolved” the keyword search term disputes. In respect of the use of different keywords, he gave the following directions: 2. Different sets of keywords should in principle be adopted for different issues as opposed to using one generic set of keywords across all issues save 1-6. 3-4 … 5. There is permission to apply once the initial harvesting of electronic date has been completed and the extent of responsiveness considered. 8. On 17 July 2020 the Defendants wrote with a list of proposed keywords by issue [139]. On 5 August 2020 the Claimants replied [147] The majority of the list was agreed, and some modest changes suggested. In their reply of 6 August 2020 [152] the Defendants agreed two of those three changes. The Claimants letter of 5 August 2020 then went on to consider how to apply the keywords for issues 7-27 by reference to paragraph 2 of Waksman J’s order of 15 July 2020. There were three matters raised: groups; (i) splitting issues 7-27 into sub- (ii) refining keywords; and (iii) using artificial intelligence (“continuous active learning”). 9. The use of AI was agreed. The other two matters were not. Firstly sub-groups. The Claimants proposed to split the issues into groups which had similar search criteria and dealt with similar subject matter. Different search terms would then be applied to these different groups to reduce duplication. The Defendants did not think this would work in practice; see paragraph 5 of their letter of 6 August 2020 [153]. They asked for a detailed account of how search terms had been applied to the Claimants dataset to date and the methodology used. A telephone or video call was offered. The Defendants position was that running a single set of keywords over issues 7-27 was consistent with the Disclosure Pilot and the order of 15 July 2020 “as much as is practically possible”. 10. The Claimants replied on 13 August 2020. At paragraph 4 the letter identifies the subgroups and explains that the keywords applicable to the issues are then applied to the whole of the Claimants’ email dataset for the date ranges applicable to that group. HHJ WORSTER Approved Judgment AAH v Jhoots and anor Responsive documents are then moved into the pool for review. At paragraph 5 the writer says this: We do not intend to correspond on this issue with you further at present, given the impending deadline for disclosure. The Court has made an order that the different sets of keywords should be adopted for different issues and that is what the Claimants are doing. Should the Defendants remained concerned, we suggest that they consider whether to make an application to the Court at a later date. In his skeleton argument Mr Croxford QC indicated that this meant an application for specific disclosure following the completion of extended disclosure. 11. Secondly the refinement of keywords. Paragraph 7 of the Claimants letter of 5 August 2020 [148] indicates that the Claimants had extracted their data and begun to apply the keywords. This process had identified some 240,000 responsive documents. That is not a proportionate number of documents to review, and consequently the Claimants decided to refine their search terms. What they decided to do is explained at paragraph 9: (a) (b) We have linked all the Claimants keywords to the words Jhoots and/or Pasab in order to eliminate as many irrelevant documents as possible, including those which relate to other customers of the Claimants; and We have removed the words Jhoots and Pasab in singularity from the Claimants list to avoid any duplicative irrelevant data following the application of the terms referred to at 9(a). 12. In their letter of 6 August 2020 the Defendants asked for some information about the Claimants 240,000 responsive documents, including a breakdown of the hits in relation to each keyword and custodian. Their position was that this number of documents was the result of over extensive date ranges and widely drafted keywords. They asked for this information expressly with a view to amending the initial list of keywords, and expressed concern at the linking of keywords without prior agreement; see paragraph 9 [154]. 13. In this letter the Defendants referred back to the Claimants letter of 9 June 2020 [130] which said this at paragraph 17: … the parties should proceed with their respective data scrapes and apply the keywords set out in our letter. The parties can then confirm the volume of data after the scrape and de-duplication, and after running initial keyword searches. Should the volumes of responsive documents be considered disproportionate, the parties can test alternative/amended keywords, communicate how this changes the responsive hits, and seek to agree the same. At the moment the parties are discussing several possible scenarios, without knowing the actual results, or which searches may lead to the most responses. HHJ WORSTER Approved Judgment AAH v Jhoots and anor The Defendants agreed with that approach in their letter of 15 June 2020, and in the letter of 6 August 2020 said that in principle they continued to do so; see paragraph 8 [154]. But they took the view that amendments to keywords should be agreed before they were tested. Without agreement the danger was that the parties would adopt different disclosure exercises, which were harder to compare and verify. The Defendants indicated that they were in the process of finalising collection and ingestion of the Defendants dataset and would run the initial keyword search in the week of 10 August 2020, and provide a breakdown of hits for each keyword (effectively the information sought from the Claimants). They suggested that once this information was shared there could then be a collaborative attempt to agree to refine keywords. 14. In their reply of 13 August 2020 on this aspect, the Claimants: (i) (ii) (iii) (iv) noted what they described as the Defendants continued uncooperative approach … in particular in respect of the use of different keywords for different issues; paragraph 2; said that they had adopted a collaborative approach and provided details of what they had done; paragraph 14 - a list of the Claimants revised keywords is at [162]-[163]; confirmed that the result of applying the Claimants revised keywords to the Enterprise Vault and OneDrive was to reduce the number of responsive documents from 240,000 to 17,026 documents, a number which increased to 38,911 when applied to Office 365 ; see paragraph 7 [158]; defended their decision to begin the review of documents. Their concern was to meet the disclosure deadline which was less than a month away; paragraph 15; (v) did not understand the need for a “hit” report; paragraph 16(d), ending that sub-paragraph with this: The Claimants have carried out a reasonable and proportionate search for documents and shall fully explain their approach to the refinement of keywords in their Disclosure Statement, at which point the Defendants can make an application to the Court for specific disclosure if they consider any documents to be missing from the Claimants’ disclosure list. (vi) referred to the possible need to revise its budget because of the high volume of responsive data and the Defendants’ approach to the disclosure exercise. 15. Having been told in effect that the Claimants were not prepared to change their approach, the Defendants decided to make this application. The application notice is HHJ WORSTER Approved Judgment AAH v Jhoots and anor framed in terms of revising the parameters of the disclosure exercise as previously ordered and agreed. Mr Morgan’s witness statement puts it on the basis of an application to restore the CCMC and pursuant to paragraph 5 of the order of 15 July 2020. 16. Paragraph 18 of the Pilot provides for the variation of an order for Extended Disclosure. 18.1 18.2 18.3 The court may at any stage make an order that varies an order for Extended Disclosure. This includes making an additional order for disclosure of specific documents or narrow classes of documents relating to a particular Issue for Disclosure. The party applying for an order under paragraph 18.1 must satisfy the court that varying the original order for Extended Disclosure is necessary for the just disposal of the proceedings and is reasonable and proportionate (as defined in paragraph 6.4). An application for an order under paragraph 18.1 must be supported by a witness statement explaining the circumstances in which the original order for Extended Disclosure was made and why it is considered that order should be varied. 17. Mr Croxford QC referred to Vannin Capital PCC v Rhos Shareholders Action Group Ltd and anor [2019] EWHC 1617 (Ch); Joanna Smith QC. He accepted that this was not an application to vary to which Tibbles v SIG applies, but emphasised the need for the Defendant to show that the order sought was necessary for the just disposal of the proceedings. 18. Before I turn to the arguments in this case, I note that at paragraph 4 of her judgment in Vannin the learned Deputy Judge said this: Before addressing the detail of the applications, I note that there has been no attempt by either party in this case to seek guidance from the court in accordance with the procedure identified in CPR PD 51U, paragraph 11, in advance of making formal applications to the court. Whilst applications to vary an order for Extended Disclosure do not appear to be contemplated as suitable for Disclosure Guidance Hearings, applications concerning the scope of Extended Disclosure expressly fall within that provision. Lengthy skeleton arguments have been filed on both sides in respect of these applications and detailed submissions have been made which took more than half a day of court time. This approach seems to me to be both undesirable and contrary to the spirit of the Disclosure Pilot which requires the parties to cooperate so as to promote the reliable, efficient and cost-effective conduct of disclosure. Whilst the differing positions of the parties appear to have been amply explored in inter partes correspondence and, it seems, were not capable of resolution without further intervention from the court, this seems to me to be just the sort of situation in which guidance could have been sought from the court under paragraph 11 (at least) on the issue of whether the Claimant's HHJ WORSTER Approved Judgment AAH v Jhoots and anor 19. 20. 21. 22. application fell within the scope of the existing Disclosure Order. Had such guidance been sought and obtained, some of the issues arising on this hearing might well have fallen away, thereby saving time and costs. Many of the same considerations apply to this case. As Mr Grant submits, the Pilot requires that the parties take a collaborative approach throughout. Paragraph 2.3 refers to the expectation that the parties “cooperate with each other” and paragraph 3.2(3) to a legal representative’s duty to the court to “liaise and cooperate” with the legal representative of the other party. Paragraph 20.2 provides for sanctions if these duties are not complied with. In his judgment in McPartland and Partners v Whitehead [2020] EWHC 298 (Ch) at [51] the Chancellor emphasised that need for cooperation: [51] The Disclosure Pilot is built on cooperation as its terms make clear (see paragraphs 2.3, 3.2(3), and 20.2(3) of PD51U). This is not intended to be mere exhortation. [58] Cooperation between legal advisers is imperative. The Disclosure Pilot must not be used as an opportunity for litigation advantage. If that is attempted, the parties responsible will face serious adverse costs consequences. The working through of the process of disclosure in a case like this is bound to throw up problems and genuine differences of view. But it does not assist the process of resolving those problems for one party to say (as the Claimants have here) we are not going to correspond with you on this issue anymore. Nor is it helpful to engage in confrontational or point scoring correspondence. The Claimants see some deliberate attempt on the part of the Defendants to delay the case. I have not reached that conclusion. The Claimants letters are responded to quickly, a good number of issues are agreed, and there are offers of video or telephone calls. The Defendants may not agree with the Claimants views, and they may be wrong about that. But instead of ending the dialogue, the Claimants should attempt to persuade the Defendants that what they propose is worth trying, and offer them some comfort about the areas of relevant disclosure the Defendants say will be missed. If that comes to nothing, then the next step is to ask for the Court’s guidance at a Disclosure Guidance hearing. If the Defendants are taking an unreasonable view, and deliberately delaying matters, that will rebound on them. I now turn to the arguments on this application. In his witness statement of 25 August 2020, Mr Morgan recognises that something needs to be done to reduce the numbers of documents the parties will have to review. He says that the data collected and ingested by the Defendants’ e-disclosure provider amounts to some 3.8M documents, and that following the application of the 47 agreed and ordered keywords there were about 1.5M responsive hits. That compares with the 240,000 documents the Claimants had identified by the same process from a total of over 900,000. He expresses some concern as to why there is this discrepancy, but the real issue is what to do about the disproportionate number of documents. Mr Morgan’s position is that the current keywords are too broad; indeed the Defendants have long complained that the HHJ WORSTER Approved Judgment AAH v Jhoots and anor keywords “account”, “cost”, “credit” and “saving” are particularly responsive (paragraph 39). He also considers that the date ranges need reviewing. He is concerned that the Claimants have not provided the information about “hits” his firm has requested to enable the Defendants to analyse the Claimants disclosure exercise, but have proceeded to adopt a unilateral approach. As to that, he regards the Claimants approach to the use of different keywords to different issues as illogical (paragraphs 51-55), and the linking of keywords as having the potential to exclude some highly relevant evidence (paragraphs 42-44). 23. Mr Kamstra’s witness statement runs to some 22 pages and reviews the correspondence in detail. I note the following matters in particular: (i) At paragraph 36 he says this: The Defendants have not acted in accordance with the 15 July Order to tailor the scope of the disclosure process, rather they have carried out searches in the precise manner that the 15 July Order said the searches should not be undertaken (a reference to using one set of keywords for issues 7-27). Nor have the Defendants engaged with this process at an early juncture, to assess the volume and nature of initially responsive documents and refine any approach (if needed) in good time. He provides further details of the Claimants approach. These appear to have been sufficient to largely allay the Defendants concerns as to the data sources harvested. He explains why there will be a discrepancy between the numbers of documents the parties will have, and that the Claimants have only applied keywords relevant to the issues which a custodian can speak to. At paragraph 46 he suggests that the Defendants could reduce the numbers of documents for review by adopting a parallel approach to the one the Claimants have tried, namely by linking their keyword searches to names relating to the Claimants; “AAH”, “Celesio” or an “AAH” domain email address. (ii) (iii) (iv) 24. 25. At paragraph 56 and following Mr Kamstra defends the Claimants’ approach to the refinement of keywords and the grouping of issues. As he points out, in the submissions made to Waksman J in June 2020 the Claimants proposed that different keywords be used for different issues. It is apparent that Waksman J agreed that this was the right approach “in principle”, and directed accordingly. The Claimants say that they are seeking to comply with that order, whereas the Defendants still want to apply one generic set of keywords to issues 7-27. That will produce an unworkably large number of documents. A central issue for the Defendants is that the way the Claimants have refined their searches will exclude documents which will show whether the Market Watch scheme was a “one size fits all” offer to all of the Claimants customers, as the Claimants say, or was bespoke, as the Defendants say. Mr Croxford QC was able to offer some HHJ WORSTER Approved Judgment AAH v Jhoots and anor comfort on that matter in the course of his submissions. There were 750 other customers, and to disclose all the documents in relation to those accounts would be disproportionate and unnecessary, but he acknowledged that the Claimants were conscious of their duty to disclose adverse documents, and expected there to be documents dealing with that aspect of the matter. 26. Attached to Mr Grant’s skeleton argument was a spreadsheet of “Keyword Hits and Revisions” that had been compiled for the purposes of the hearing before me. Mr Croxford QC noted that it came not in evidence or in correspondence but from Counsel, and that it included the Defendants apparently adopting the suggestion Mr Kamstra made about pairing keywords with “AAH” and “Celesio”. On the one hand these are all perfectly valid points to make. On the other, this does at least show a willingness on the Defendants part to make some constructive suggestions. 27. If this were a Disclosure Guidance hearing, I would offer the following views on the material presently before me. (i) (ii) (iii) (iv) (v) The number of responsive documents on both sides means that something significant has to be done to reduce the numbers of documents to manually review. The Defendants will have to move from their position of using one generic set of keywords for all of issues 7-27. The Claimants sub-group proposal should be actively explored. The Claimants proposal of pairing keywords is a sensible way forward, so long as there is some thought given to the documents about the Market Watch scheme as it applied to the Claimants other customers. If a search of other customers’ documents is required, the parties should consider a search of a representative sample of those customers (say 10) over a limited date range to see what that produces. Similarly the proposal Mr Kamstra makes (and which the Defendants now seem to adopt) to pair keywords on their searches with “AAH”, “Celesio” and “AAH” domain names is a good idea. The Claimants should be prepared to look at Mr Grant’s spreadsheet of keyword hits and revisions to see whether there is anything of merit in it – particularly if taken in conjunction with a willingness on the part of the Defendants to use different keywords for different issues or sub-groups of issues. The parties should try to conduct their discussions face to face rather than in correspondence. A weekly telephone call would be a minimum in the current situation. It is important to have some record about what is being done, and for arguments to be put. But set piece letters are not always the best way of promoting a collaborative process (a point which echoes a view Mr Morgan expresses). Speaking to each other is more likely to promote cooperation. It is also quicker than writing a long letter. HHJ WORSTER Approved Judgment AAH v Jhoots and anor (vi) (vii) There needs to be a willingness to share information and to try reasonable proposals made by the other side, even if there are concerns about them. Disclosure is a process. The passage from the Claimants letter I quote at paragraph [13] above shows a realistic approach. The results will not, and cannot be perfect. The timetable to trial will have to be adjusted, and realistically more along the lines the Defendants suggest than the Claimants. Concerns about further delay can be quietened by fixing a provisional trial date now and identifying a date for a further Disclosure Guidance hearing in the next 2-3 weeks to monitor progress. If it is not needed it can be vacated. 28. 29. I set out those views because when I asked him the direct question, Mr Grant was frank enough to accept that in practical terms, the best thing the court could do was require the parties to exchange information and enter into discussions about the way forward. Where that leaves the application is a matter I will come back to, but that did seem to me to be the best hope of moving this disclosure exercise forward. The Claimants’ position was that the application was flawed, and a smokescreen to disguise the Defendants’ “abject failure” to progress its own disclosure exercise. Mr Croxford QC submitted that the Claimants approach had been a model. They had kept the Defendants informed of what they were doing, and if the Defendants did not like the result then they could make an application for specific disclosure after extended disclosure had been provided. The Claimants had come up with some sensible proposals to reduce the number of documents to be reviewed to manageable proportions, but the Defendants had stuck to an approach which Waksman J had rejected. In the circumstances, the Claimants were entitled to take their own approach to this problem. 30. Mr Croxford QC also submitted that a disagreement about what keyword searches to use did not change the fundamental obligations of the parties to provide disclosure in accordance with their duties under the Pilot, and that the matters which remained to be agreed between the parties in relation to keyword searches were not matters the court should look to determine. These were aspects of disclosure which a party was entitled to pursue as they thought fit, subject of course to the overriding duties owed to provide proper disclosure. In that regard he referred to PD51U and the text of the Section 2 Questionnaire at 10. Keyword search terms Please list any keywords identified at this stage that you may use to search the data to identify documents that may need to be disclosed. If a certain keyword is relevant only to a particular Issue for Disclosure, please indicate this if it might allow the scope of the search to be narrowed. Nb: The use of initial keywords may assist the parties to identify the likely volume of data that may need to be tested and refined during the disclosure HHJ WORSTER Approved Judgment AAH v Jhoots and anor process. Accordingly, any keywords proposed at this stage are for the purposes of discussion only. The fact that a party may propose a keyword at this stage should not be taken as an acceptance that the keyword should ultimately be used, particularly if, on testing the keyword against the available data, it provides false positive results. If it is not practicable to provide a list of keywords prior to the CMC, the parties should engage and seek to co-operate following the CMC to identify and agree the key words they propose using and thereafter test those key words against the data to determine whether or not they are appropriate. 31. Mr Croxford QC points to the use of the word “may” in the first sentence. I also note the reference to the use of “initial keywords” and the testing process. But the underlying assumption of this part of the DRD is that (through the process of testing and refinement) the parties will continue to try and agree the keywords they intend to use. Moreover that is what Waksman J directed the parties were to do at the CCMC. Similarly, the order of 15 July 2020 determined the question of whether different keywords were to be applied to different issues “in principle”, and left the detail to be agreed. At (5) the order provided for a return to court later in the process if that was necessary. 32. 33. 34. Consequently, whilst I recognise that the Court may not be anxious to engage in the micro-management of the disclosure process unless that is necessary, and that the parties solicitors are often in a better position to understand the practical issues, in the absence of agreement the Pilot (and the orders made by Waksman J) envisage recourse to the Court on these matters. In the course of argument it became apparent that the parties took a fundamentally different view of how the Pilot worked in this respect. The Claimants approach was that subject to complying with their duties of disclosure, they could proceed with their searches without the agreement of the Defendants. The Defendants disagreed. The Claimants could not proceed unilaterally. In reply Mr Grant referred me to the decision of Marcus Smith J in Agents’ Mutual Limited v Gascoigne Halman Limited [2019] EWHC 3104 (Ch) and in particular to paragraph 15(iii) of his Judgment. Because electronic searches have the effect of determining the scope of subsequent, more intense and probably manual reviews, it is imperative that such electronic searches not be conducted unilaterally, but with the parties engaging with each other with a view to agreeing precisely how the electronic search is to proceed. I have well in mind that framing electronic searches is a difficult process likely to be informed by trial and error. For example, keywords may very well appear to be appropriate, and the expectation will be that an unmanageable universe of documents will be reduced to manageable proportions. But such a search, when actually conducted, may very well HHJ WORSTER Approved Judgment AAH v Jhoots and anor produce too many hits or too few. In which case, in an iterative and cooperative way, the process of re-framing the electronic search process must be done again. It is important that this process be completed before any manual review is undertaken: it is the manual review that costs time and money, not the multiple re-runs of electronic searches. As Mr Croxford QC points out, this is obiter. That said, it is part of a considered judgment, and in my view entirely consistent with the terms of the Pilot and the policy behind it. 35. In his supplemental written argument, Mr Croxford QC submitted that the aim of PD51U was to avoid delay and cost, and that its terms should be interpreted in line with that purpose. When there was such an emphasis on cooperation, one party should not be allowed to delay matters. He put it this way at paragraph 9 of his supplemental submissions: To challenge self-evidently reasonable refinements to an electronic search at the last possible minute to secure delay, both because the client wants to delay a trial and to avoid inevitable default in disclosure obligations would be inimical to the very purpose of the Disclosure Pilot. He submits that whilst Agents’ Mutual is authority (albeit only persuasive) for the proposition that the parties are subject to an obligation to engage with a view to agreeing search parameters, it should not be treated as authority for the proposition that any disagreement, at however late a stage, must stop disclosure in its tracks and secure an automatic delay. When faced with “irrational or profoundly unhelpful intransigence” from the other side a party is entitled to insist that the process continues and that an application be brought pursuant to Paragraphs 17 or 18 of the Pilot after Extended Disclosure is provided. He acknowledges the risk if such an application succeeds, the party who proceeds without agreement may have the increased cost of searches (and I would add, review). 36. The starting point for my approach to this matter is the duties of the parties under the Pilot. In particular: (i) (ii) the duty under paragraph 2.3 … to cooperate with each other and to assist the court so that the scope of disclosure, if any, that is required in proceedings can be agreed or determined by the court in the most efficient way possible . and as the Chancellor emphasised in the closing paragraphs of his judgment in UTB LLC v. Sheffield United Ltd, [2019 ] EWHC 914 (Ch): Legal representatives have continuing obligations under paragraph 3.2(3) of PD51U "to liaise and cooperate with the legal representatives of the other parties … so as to promote the reliable, efficient and cost-effective conduct of disclosure" and indeed the trial of the action more generally. HHJ WORSTER Approved Judgment AAH v Jhoots and anor The Pilot is built on those duties, and provides for sanctions if they are not complied with. Cooperation is “imperative” per the Chancellor in McPartland. If parties choose to ignore those duties, the Court has made it quite clear that there will be serious adverse costs consequences. 37. I do not regard the Defendants approach to this matter as a tactic to delay the trial. If I did then I would dismiss the application with costs on an indemnity basis to be paid forthwith, and direct that the disclosure exercise proceed as envisaged by the Claimants. The Claimants may be on the right lines by pairing search terms and using sub-groups, but they should not have refused to provide the Defendants with further information or ended the dialogue. 38. Mr Croxford QC’s submissions raise the spectre of the cynical litigant deliberately refusing to agree sensible proposals. That of course would be a breach of the duty to cooperate. The remedy is to use the machinery provided by the Pilot and apply for a Disclosure Guidance hearing or (if the case warrants it) to apply to vary the order. The Claimants concern to keep to the timetable is admirable, but proceeding unilaterally in this case has, in the event, caused both sides to spend time and money on this application which would have been better spent in conference calls sorting out the “nitty gritty”. 39. 40. 41. I agree with the views expressed by Marcus Smith J in Agents’ Mutual. The problems in that case were different, but the approach to electronic searches he outlines at [15iii] make sense for the reasons he gives. If the Pilot allowed for a party to proceed unilaterally on these matters it would serve to weaken the duty to cooperate. That runs counter to the culture the Pilot promotes. Finally I return to the application. I am not minded to make the orders Mr Morgan refers to in paragraph 67 of his witness statement. As matters stand, I am not satisfied that these steps will achieve the desired result. I intend to adopt the approach Mr Grant suggested in his submissions, and require the parties to engage in further discussions about this process. Those discussions can proceed on the basis of the “guidance” I set out at paragraph 27 above. It goes without saying that I expect the parties to adopt a genuinely collaborative approach, to be flexible and open to testing proposals, and to stop fighting old battles or scoring points. In the course of Mr Croxford QC’s submissions I raised with him the question of how to proceed if I were against him on the Agents’ Mutual point. I was concerned to move the process along, and wondered aloud whether he might want the court to retrospectively approve the Claimants’ approach. Mr Croxford took the view that approval was not needed, but that if it was, he sought it. Again I have decided that this would not be the right approach. The onus is on the parties to agree this process. The order I make reflects that. 42. I will list this case for a Disclosure Guidance Hearing in the week commencing 5 October 2020 for 30 minutes. That will be conducted by telephone or Teams. I do not require an application notice, nor should either party make a witness statement. A HHJ WORSTER Approved Judgment AAH v Jhoots and anor short agenda of items for discussion, setting out briefly any competing proposals, should be agreed and filed the day before the hearing, The discussion is to be with the legal representatives with direct responsibility for the conduct of disclosure, and they should provide the Court with their joint availability during that week as soon as possible. In the meantime the parties should identify dates of availability for trial and seek to agree a revised timetable. If there is no agreement by the time of the Disclosure Guidance Hearing I will make some further directions. 43. The costs of this application are no doubt significant. I have not heard submissions, but my provisional view is that the costs should be in the case. I have not varied the orders for disclosure, so in that sense it will be said that the application has failed. But I understand why the Defendants brought the application. They were faced with the Claimants refusal to negotiate further. It might have been better if a Disclosure Review hearing had been sought, but in the event the effect has been similar. Moreover they succeeded on the Agents Mutual point, and the Claimants have been required to re-engage in discussions. No doubt it will be said that this was the underlying purpose of the application. Overall, costs in the case best reflects the justice of the position as between the two sides. If either party wants to make an application for a different order, they may file and exchange written submissions (limited to 4 sides) within the next 7 days, and I will determine the issue on paper. 44. I am distributing this Judgment in an unedited form without the usual confidentiality notice so that the parties can re-start their discussions as soon as possible. I should be grateful for Counsel’s note of any grammatical or other obvious errors as soon as possible so that I can prepare a Judgment in final form. It goes without saying that the approach I have adopted in this case turns upon its particular facts.