Infinity Reliance Limited v Heath Crawford Limited

The claimant (“Infinity”) is an online retailer. As “My 1st Years”, it sells personalised gifts for babies and children—clothes, toys, blankets and the like. When an order is placed, the goods need to be brought from storage, and then personalised for the buyer, such as by embroidering the child’s name. Because of this, Infinity cannot simply use a third-party logistics company to pick, pack, and post its orders. It needs to arrange space in the warehouse where its staff can personalise the goods before they are despatched.

In 2021, it was using space in a warehouse in Northampton. The warehouse was owned and operated by a logistics company called Cygnia. Infinity paid Cygnia for much of the logistics work. It had space in the warehouse where its staff could add the personal touches to the goods before they were sent out.

Infinity had thrived during Covid and its lockdowns. As the country came out of lockdown in April 2021, it remained to be seen how far it would be able to maintain that growth.

In May 2021, there was a fire at Cygnia’s warehouse. Infinity could not use the warehouse any more. Its business was interrupted and it lost sales until it was able to find suitable alternative premises, and fit them out. That took time, and it cost money. Infinity lost many millions of pounds of sales, and spent more than £2 million to fit out new premises.

Infinity had insurance cover, including cover for business interruption (“BI”). Its insurance broker, since late 2018, had been the defendant, Heath Crawford Ltd (“Heath Crawford”). Heath Crawford advised Infinity to buy a commercial combined policy from Aviva plc. The policy included BI cover. When the fire damaged the warehouse, the most recent policy had been renewed at the start of November 2020.

Unfortunately for Infinity, the cover was not enough. Infinity’s BI insurance was based on a forecast gross profit of £24.9 million over two years. But the right figure, given the policy’s terms, would have been higher—more like £33 million. So Infinity was underinsured. When it settled the claim, Aviva applied the doctrine of average. Since Infinity’s insured exposure was 26 percent less than its full exposure, it only recovered 74 percent of its adjusted loss, £9.25 million instead of £12.17 million. That was agreed between the loss adjuster appointed by Aviva and the claims assessor appointed by Infinity.

There was another disappointment. There was equipment at Cygnia’s warehouse to enable it to pick and despatch orders. That was owned by Cygnia. When Infinity had to find a new warehouse, it could not easily find similar premises, and had to spend heavily to equip new premises with equivalent equipment, and to do some other things such as begin work on installing sprinklers which insurers of the new premises required. These costs were only partly insured.

Nobody suggests that Aviva did not pay what was properly due under the policy it had underwritten. Infinity says that Heath Crawford, as its broker, was at fault. If it had given proper advice, Infinity would have been fully insured for the loss. It therefore claims the shortfall. Its specific complaints are:

For the reasons given below, I have decided that Heath Crawford was at fault in the way it advised Infinity, but that Infinity was also careless. I assess damages, after a deduction of 20 percent for contributory negligence, in the sum of £2,336,842.

Heath Crawford accepts that it owed a contractual duty to Infinity. It concedes breach by providing a misleading explanation of how to calculate the sum insured. It almost (though not quite) concedes breach in failing to recommend that Infinity purchase declaration linked cover. It denies the third alleged breach, which concerns assessing Infinity’s need for specific cover for fit-out costs at alternative premises.

When it comes to causation, Heath Crawford accepts (as I understand the submissions made on its behalf by Mr Parker) that the misleading guidance it provided on BI cover was at least a cause of loss, but subject to the contributory negligence argument discussed below. It does not accept that the absence of a recommendation of declaration linked cover caused loss, because it does not accept that Infinity would have accepted the recommendation. It does not accept that Infinity has proved that any failure to consider fit out costs has led to any loss. That is because it does not accept that anything more than cover for the additional increased cost of working would have been obtained, and it does not accept that such cover would have made any difference to Infinity.

As to contributory negligence, Heath Crawford focuses on how the sum insured was calculated by Infinity. It says that Infinity did not follow the advice it had been given properly, and that if it had followed that advice the error that Heath Crawford had made would not have mattered. It says that this amounts to contributory negligence which should greatly reduce, or even eliminate, the amount that Infinity recovers.

There were two minor issues about the quantum of Infinity’s claim in relation to fit out costs, though those were largely agreed.

I heard evidence from four witnesses of fact: Mr Cress, Mr Sitton, and Mr Andrews for Infinity; and Mr Leyens for Heath Crawford. Mrs Ward was originally going to give evidence for Heath Crawford too. But she was unable to attend the trial for family reasons. Her statement was not relied on, and I have not taken its contents into account.

Mr Cress had a background in corporate finance, as an internal consultant for AOL UK, as CFO of a company that operated an e-commerce marketing platform, and as finance director for a company that sold luxury designer goods and jewellery on social media. He joined Infinity in April 2019 as Finance Director, replacing Ms Kenner who moved to another role. He is evidently intelligent. I did not get the impression that he was deliberately seeking to mislead me.

Nevertheless, I have two reservations about aspects of his evidence. First, one theme of his evidence to me was that he had not, at the time, expected Infinity to continue to experience strong growth in revenues or gross profits. He was asked to comment on various financial projections which painted a different picture of his expectations at the time. I discuss these in more detail later in this judgment. But the general point was that these projections consistently showed a much more bullish view of Infinity’s growth prospects than Mr Cress now espouses. For example, one such projection given to the Board at the end of 2019 was (Mr Cress agreed) based on an assumption of 20 percent growth. Another was a set of “internal forecasts” which used higher numbers. In those and similar cases, Mr Cress’s evidence was that he regarded those projections as punchy to the point of being unrealistic: Board forecasts, he suggested, had been inflated in the hope of encouraging third party financiers, and the internal forecasts had served as “stretch” targets. He had never truly set much store by them, he said.

However, when the time came to claim on the BI cover, the loss adjusters had been told that Board estimates were “conservative”, and that internal estimates were “key”. Those sentiments, as Mr Cress largely accepted (though he cavilled about the origin of certain phrases) must have come from him. But, if his evidence at the trial was true, they were varnished. Mr Cress resisted describing them as “lies”. He preferred to call them “advocacy positions” put forward because of the high importance Infinity attached to maximising its recovery on the BI cover, and in the expectation that the loss adjusters would take them with a grain of salt and make further inquiries. He explained away other projections that were at odds with the view he was expressing to me as motivated by ad hoc strategic purposes.

There is not much purpose debating the appropriate epithet. In any case, however, Mr Cress’s evidence is that he is prepared to advocate for forecasts for strategic aims: to encourage investment, to spur employees, to maximise recovery on an insurance claim. He will do that, he says, even if he personally believes they are not realistic and accurate.

In these proceedings, Mr Cress espouses different forecasts, including forecasts which left no trace in the contemporaneous record. I cannot place much implicit confidence in those when they, too, have an obvious strategic purpose—to assist Infinity in the claim that it is now making, and to rebut suggestions that Mr Cress made mistakes. That is not to say they are dishonest. Forecasts are imprecise even in stable times. And late 2020, in the shadow of Covid, was far from a stable time. Arguments for varying forecasts can, as arguments, be honestly devised. But I am not concerned with what arguments might be defensible as advocacy positions, but with what Mr Cress in fact did, or in fact would or should have done, at the time. On that issue, Mr Cress’s evidence, given his attitude to forecasts as malleable and manipulable for strategic reasons, must be tested sceptically.

The second difficulty is that Mr Cress was often being asked about hypothetical, counter-factual, questions: what would he have done, if things had been different? And even where he was being asked to remember what factually happened, it was obvious that Mr Cress often relied on reconstruction. Again and again, he answered questions about what he had done or thought with answers which described what he would have done or thought. Gaps in memory are to be expected. Mr Cress did not, at the time, have BI insurance high on his list of priorities. He had come as Finance Director to a company which turned out to face serious financial challenges. It was then dealing with all the uncertainty and chaos of Covid. Mr Cress is now asked about events which seemed quite unimportant when they happened. So it is not surprising that much of his evidence is retrospective reconstruction; but as such it carries all the dangers of hindsight. I accept that Mr Cress now believes what he says. But that does not make it reliable. The objectively known facts and inherent probabilities are more important.

Mr Sitton is Infinity’s CEO. He was tangentially involved in the insurance placement because he introduced Mr Leyens, participated in one early meeting, and was copied on a few emails. He was a straightforward and credible witness who had, in truth, very little of relevance to say. He played no part in any of the calculations of turnover which are directly relevant to the case.

In so far as Heath Crawford suggested that Mr Sitton was intent on economising on insurance, even to the point of preferring underinsurance, I do not agree. The documents suggest that he was cost-conscious. But there is nothing surprising about that. They show no case where he rejected a soundly reasoned recommendation about insurance simply to save money in the short term. Although Infinity’s cash position was sometimes tight, its objectives were long term success, and there is no reason to think that it was unable or unwilling to pay for appropriate insurance.

Mr Andrews worked with Mr Cress. He assisted in some aspects of the insurance placement. His evidence was not seriously challenged, and I accept it; but it barely touches on anything of importance.

Mr Leyens is a director of Heath Crawford. It is a family firm, set up by Mr Leyens’ father, originally specialising in health insurance. Since 2008, Mr Leyens has developed the “general insurance” side of the business, covering a broad field and specialising in SME clients and individuals.

I found Mr Leyens to be a guileless, even suggestible witness. On some points, he had previously signed documents (such as statements of truth on the defence) which contained statements he did not, in his oral evidence, support. But in those cases his oral evidence was less helpful to Heath Crawford than the original account, and I am sure that his oral evidence was candid. It was consistent with other evidence and the inherent probabilities. If anything, despite entirely proper cross-examination, he appeared punch drunk: inclined to concede points without much thought, even when they were—put alongside the documents—questionable. He did not make excuses or pin the blame on others. I thought him an honest and honourable man. I did not, however, get the impression of a technically adept broker, nor of a person who considers documents carefully. He has a client-focused and pragmatic approach. But he is not someone who thinks things through in a logical and rigorous way, from the bottom up. He sometimes answered questions with the first thing that came into his head. I have therefore been cautious even about what seemed to be admissions on his part.

I heard expert evidence from two broking experts, Mr Finnegan (for Infinity) and Mr McGrath (for Heath Crawford). Both were brokers of long experience, well qualified to give evidence. Both sometimes trespassed beyond the legitimate realm of their expertise, and stooped to taking small swipes at the facts where it would help the party instructing them. But I did not regard either expert as egregiously partisan, and I am sure both were doing their best to help me. In the end, there was little difference between them about the relevant underlying principles.

The parties adduced written evidence from surveying experts on quantum. But in the end this was entirely agreed, apart from two matters which were left for submission.

Paragraphs 29 to 76 set out my findings of fact. Where a relevant fact is in dispute, I explain the reasons for the conclusion I have reached, on a balance of probability. Some of the issues are hypothetical questions. I save those for later, preferring to start with primary facts.

Infinity was established in 2010. By 2018, it had an annual turnover of just over £11m, and gross profit of just over £4m. It was still making an operating loss, however, and was not yet financially stable.

Infinity placed its insurance programme annually, incepting in November each year. It is convenient therefore to refer to the 2018/19 policy year (and so forth) to cover the 12 months from November 2018, and to the 2018/19 renewal to refer to the placement and renewal of insurance for that period, which would occur in the preceding month or so.

Until the 2018/19 renewal, Infinity placed insurance through a firm of brokers called Bridge Insurance Brokers Ltd (“Bridge”). In 2017, Mr Sitton encountered Mr Leyens in connection with his household insurance. He asked Mr Leyens to quote for the 2017/18 renewal. It was too late to do anything that year, but as the 2018/19 renewal came around, Mr Leyens asked to be allowed to quote. He already had Bridge’s summary of the expiring programme, and Mr Sitton’s assistant sent him Infinity’s existing insurances. These covered a variety of risks including BI.

BI cover can be placed on one of two different bases. The traditional form of cover, which I shall call “sum insured BI cover”, requires the insured to forecast its insured profit for the indemnity period. The premium is fixed in advance. But if the insured has underestimated the risk or grows faster than expected, it will be underinsured. If that happens, the insurer will apply average to any claim. It will reduce, pro rata, the indemnity that is paid by the proportion of underinsurance. So if the insured gross profit is £1m, but the actual gross profit is £2m, the insurer will only pay 50 percent of the loss. That is true even if the loss is far below £1m. The insured is treated as having chosen to insure only part of its revenue, and to have self-insured the rest, so that insurer and insured contribute rateably to any loss.

That makes it important for an insured that wants full coverage to estimate its future gross profit correctly. That may be no easy task. If the BI cover has a two-year indemnity period following the insured peril occurring, as Infinity’s did (i.e. it may compensate the insured for up to two years’ lost revenue after a loss begins) one must reckon with the possibility that a loss may start on the last day of the insured period, a year after the renewal, and continue for two years thereafter. So the insured must estimate not only its revenue for the year of insurance but for the indemnity period immediately following—in my example three years from renewal. If the estimate is too low, the insured will be exposed to average, however carefully it made its forecasts.

The alternative basis is a declaration linked cover, originally developed during the 1970s to cope with the problems that insureds faced forecasting revenue during inflationary periods. When covered on this basis, the insured must still declare its turnover and profit in advance, so that the initial premium can be assessed. But at the end of each period of insurance, the actual performance may be considered. Within broad limits, if it is higher than forecast, the insured pays an additional premium; if it is lower, premium is returned. The insurer agrees not to apply average. Effectively, the actual risk is retrospectively rated by adjusting the premium. It is only if the insured’s original estimate was very badly wrong that a claim may be imperilled.

The policy that Bridge had placed with AXA was declaration linked cover.

On 17 October 2018 there was a meeting to discuss Infinity’s insurance. It was held at the WeWork offices in King’s Cross that Infinity then used. Mr Sitton and Ms Kenner, the Finance Director, attended for Infinity. Mr Leyens and a colleague, Mrs Ward (then Ms Clifft), attended for Heath Crawford. Nobody who was there has preserved any notes.

Mr Sitton thinks he will have explained the company’s operations, especially at the Northampton warehouse. He was taken to task on the fact that his first witness statement did not mention that. I do not think that is a weighty point. Mr Leyens needed to know about Infinity’s business. Mr Sitton was the obvious person to provide that information, and it is likely he did. On the other hand, it is unlikely that he went into details such as the difficulty of finding alternative logistics suppliers who would replace Cygnia. That is now an obviously salient point—but it would not have been high on the agenda of anyone at an introductory meeting.

Heath Crawford’s mission was to see if it could obtain essentially the same or better cover as Infinity already had, but at a lower premium. Whether or not Mr Sitton said so in terms, that is (as Mr Leyens accepts) how Heath Crawford understood its task. The 2017/18 programme provided at least the starting point, and Heath Crawford built upon it.

Mr Leyens’ evidence is that at some point during the renewal process Ms Kenner made it clear that Infinity did not want BI cover on a declaration linked basis. He could not say whether it was in this meeting or in conversations later in the process. He remembered that this was because Infinity had been hit by a premium adjustment request in relation to its existing cover, and that it wanted premiums fixed in advance.

Mr Leyens was firmly challenged on this. It was put to him that this attitude would have been irrational. The premium rate for a declaration linked policy is not higher than under a traditional cover, as Mr Leyens and the experts agreed. If additional premium becomes due, that will be because the business has performed better than originally estimated. And that will always mean, it was put to Mr Leyens, that if it had been insured under a sums-insured policy on the original estimate, it would have been underinsured, and liable (if a claim occurred) to see it reduced by average. So it would be senseless not to prefer declaration linked cover.

Mr Leyens stuck to his guns. He insisted that in his experience, whatever the theoretical advantages of declaration linked policies, many of his clients prefer to avoid them, and that Infinity (as Ms Kenner told him) was one of those clients. Ms Kenner did not give evidence. I accept Mr Leyens’ evidence. Whether sensibly or foolishly, a business may prefer to take the risk of underinsurance and have stable premiums. That may be particularly likely if the insured does not understand the full implications of under-insurance, and sees declaration linked cover as a device for insurers to profit from their insured’s financial success at a time when (nearly always) the cover will have expired loss-free. Or it may be because the insured prefers to self-insure part of its business interruption risk. Those may not be very sensible preferences, but it does not mean that they are not real preferences. Sum-insured BI cover is, after all, still available in the market, so it must have its fans and uses.

It is not incredible that Ms Kenner, working for a company which was growing, cash-sensitive, and still loss-making, would have seen superficial advantage in a policy which enabled her to know premium in advance. That becomes more probable if she did not understand the downside. In the absence of evidence from her, and given Mr Leyens’ clear evidence on this point when he was ready to concede many others, I accept his evidence.

Mr Leyens accepted, however, that he had not explained to Ms Kenner that declaration linked cover had real potential benefits for a company with a growing or unpredictable turnover. He had not checked that she understood the significance of average in cases of underinsurance, or the significance of declaration linked cover as a way of removing that risk. He had accepted at face value her preference to avoid a form of cover which involved retrospective premium adjustments, without making sure that she knew what Infinity was giving up, or knowing whether its preference for a traditional BI cover represented a genuine willingness to retain part of its BI risk.

Heath Crawford provided a quotation on 30 October 2018. It stressed that it offered a saving, compared to the expiring cover placed by Bridge, of just over £16,000. The BI cover proposed was with Aviva. It had an insured sum of £17m, with an indemnity period of 24 months. In line with Ms Kenner’s wishes, it was not declaration linked cover. The quotation did not draw attention to that fact or explain the difference it might make. The figures seem simply to have been taken from the expiring Axa policy, and Heath Crawford gave Infinity no guidance about how it should be calculated.

There were further discussions over the terms of the cover which slightly reduced the saving. Bridge also provided an improved quotation. In the end, however, Infinity opted for Heath Crawford’s quotation.

In the middle of 2019, Infinity had a problem caused by a cyber attack which put Cygnia’s systems out of action. Infinity notified Heath Crawford about this. It does not matter directly. But it should (as Mr Leyens accepts) have alerted Heath Crawford to the fact that Infinity’s ability to trade was dependent upon Cygnia’s equipment, if that was not already appreciated.

By the time the 2019/20 renewal came round, Ms Kenner had been replaced as Finance Director by Mr Cress. Again, there was a meeting, on 17 October 2019. Mr Cress attended for Infinity. Mr Leyens and Mrs Ward attended for Heath Crawford. Nobody kept detailed notes, but the participants worked through a document which listed some information known and required. Later that day, Mrs Ward sent that document to Mr Cress. There is a version with added notes in red, which I infer from their phraseology were added by Mrs Ward to reflect discussion at the meeting. Mrs Ward’s covering email shows that there had been more than superficial discussion of Infinity’s business and its development.

One item in the document raised the question “Could you provide more detailed information on the factory/warehouse and the work processes involved in the Blackmills premises”. The answer recorded was:

“Warehouse is run and owned by Cygnia Logistics, who have their own staff. Cygnia staff will pick the items and then hand them over to a MyFirstYears staff member. The staff member will customise the goods by embroidering or printing the name. Once the goods are customised they are then handed back to a Cygnia Logistics employee who then packs the items for sending”.

That information, whoever recorded it, must have come from Mr Cress.

There was discussion at the meeting about how to calculate the sum insured for BI purposes. In the end, Mr Cress and Mr Leyens largely agreed on what happened. Mr Cress asked how this should be done, and Mr Leyens said he would send him a document explaining it. Such a document, entitled “How to Calculate Gross Profit” was among the documents Mrs Ward sent after the meeting.

In its defence, Heath Crawford alleged that Mr Cress had said he already knew how to calculate gross profit for BI purposes. This turned out to be quite wrong. He said no such thing. Neither he nor Mr Leyens now suggests he did. It would not have been true, and if it had been true he would not have needed the guidance which Heath Crawford went on to give. The furthest Mr Leyens went in his oral evidence was that Mr Cress already understood that “gross profit” for BI purposes was different from “gross profit” in its ordinary accountancy sense. I think it probable that Mr Cress did seem to understand that. It would be a good reason for a person who was intimately familiar with accounts to be looking for advice from his broker about the meaning of “gross profit”. What matters is that Mr Cress asked for, and was given, guidance on this critical point.

Heath Crawford did not, at this meeting, or ever, raise the possibility of declaration linked BI cover with Mr Cress. Mr Leyens said that this was because, from his point of view, it had been made clear the previous year that Infinity’s firm preference was not to buy cover on that basis. He called it a “policy”. This may well have been a reason Mr Leyens did not raise the question. It does not follow that it was a good reason for leaving the topic untouched, but that is a question for later.

Heath Crawford’s guide to “How to Calculate Gross Profit” was a two page document. Although Mr Leyens seemed willing to accept that it might not be obvious, it was clear from its face that it was a generic document, not prepared specifically with Infinity in mind. I am sure that Mr Cress understood that. But it was presented as guidance that would apply to any sort of BI cover, and without any suggestion to check the definitions in Infinity’s policy. Heath Crawford did not draw Mr Cress’s attention to those definitions, and Mr Cress reasonably supposed that the document was reliable for his purposes.

The document contained the following statements under the heading “Sum Insured”:

“For many businesses, the basis of the Sum Insured will be the annual Gross Profit figure, which for BI purposes represents:

Annual turnover plus closing stock and work-in-progress less

Opening stock and work-in-progress plus variable expenses.

Variable expenses are those expenses which would reduce, or disappear entirely, in the event of a stoppage to the business.

Once an accurate and current (BI) Gross Profit figure has been calculated, it must be adjusted upwards to allow for anticipated growth in the business during the period of insurance itself and the Indemnity Period selected, bearing in mind that it is possible for a ‘worst case scenario’ loss to occur on the last day of the period of insurance.

As an example, if the current annual Gross Profit figure is £100,000, the period of insurance is 12 months, the Indemnity Period selected is 24 months and business growth is estimated at 10% per annum, the correct Gross Profit sum insured is £254,100:

Gross Profit during 12 month period of insurance = £100,000 + 10% = £110,000

Gross Profit during 1st year of Indemnity Period = £110,000 + 10% = £121,000

Gross Profit during 2nd year of Indemnity Period = £121,000 + 10% = £133,100

Gross Profit sum insured = £121,000 + £133,100 = £254,100

In the event that the Gross Profit sum insured is not calculated correctly, there is likely to be underinsurance and Average would apply to the settlement of ANY claim.”

As a guide to calculating the sum to be insured for the Aviva policy, this was incorrect. The Aviva policy provided:

“We will cover You only for loss of Insured Profit due to (1) reduction in Turnover, and (2) increase in cost of working.

We will pay … (1) Turnover, the sum produced by applying the Rate of Insured Profit to the amount by which, due to the Damage, the Standard Turnover exceeds the Turnover during the Indemnity Period. (2) increase in cost of working, any additional expense You necessarily and reasonably incur solely to prevent or limit a reduction in Turnover during the Indemnity Period which but for such additional expenses would have taken place due to the Damage.

We will not pay, in respect of (2) above, more than the amount produced by applying the Rate of Insured Profit to the reduction in Turnover avoided by the expenditure.

If at the time of the Damage the Sum Insured is less than the sum produced by applying the Rate of Insured Profit to the Annual Turnover, proportionately increased where the Maximum Indemnity Period exceeds 12 months, You will be Your own insurer for the difference and bear a rateable share of the loss.”

“Rate of Insured Profit” was defined as the “Insured Profit earned on and expressed as a percentage of Turnover, during the financial year immediately before the date of the Damage”. “Insured Profit” was defined as “The combined value of the Turnover, closing stock and work in progress, less the combined value of opening stock and work in progress and Uninsured Working Expenses”. And “Uninsured Working Expenses” was defined as “(1) Purchases of materials for production or re-sale less any discounts received (2) discounts allowed, and (3) any additional uninsured working expenses stated in The Schedule.”

The key error in Heath Crawford’s guidance document, therefore, was that for the purposes of the Aviva policy the insured profit did not constitute the difference between turnover and “those expenses which would reduce, or disappear entirely, in the event of a stoppage of the business”. It constituted simply the difference between turnover and the cost of material for production and discounts received (unless there were other “uninsured working expenses” listed in the Schedule, which there were not).

Despite asking for and getting guidance, Mr Cress does not appear to have made much use of it for the 2019/20 renewal. His first witness statement said he had “calculated” the level of BI cover required that year using it. But his second witness statement retracted that, and he says that it was a “mistake”. In fact, he made no calculation. It remained unclear to me what the mistake was—whether Mr Cress had, when he signed his original statement, a false memory of carrying out detailed calculations in 2019, or whether he gave an unduly dignified description of a very “finger in the air” exercise. Except for Mr Cress’s credibility in other respects, it does not matter. What is clear is, as he accepts, that Mr Cress read the document, became convinced, on a broad-brush analysis, that the cover from the previous year was ample, and left it unchanged.

Heath Crawford’s guidance document included the warning that “In the event that the Gross Profit sum insured is not calculated correctly, there is likely to be underinsurance and average would apply to the settlement of ANY claim”. Mr Cress says he read that, but that he did not understand it, because he did not understand what “average” is. He did not ask Heath Crawford to explain it. He assumed that the sum insured was a limit of liability, but that so long as the ultimate loss was below it, the claim would be paid in full. That (which is not what the document says) coloured his approach thereafter. It meant that, as Mr Cress saw things, underinsurance was problematic only if a loss exceeded the insured sum—which seemed highly unlikely, except on the very worst of “worst case” scenarios.

The guidance document also said that:

“If cover is arranged on a Declaration Linked basis this may well offset the need for projection but the selected indemnity period and any exceptional changes to the business will still need to be taken into account”.

Beyond telling him that there is some form of BI cover which is described as having a “Declaration Linked” basis, this could not have told Mr Cress anything useful. It did not say how that cover worked, how it was priced, how or why it would “offset” the need for “projection”. Nor did it explain how it would alter the consequences of underinsurance.

By the time of the 2020/21 renewal, Covid had struck, and everyone was working remotely. Covid had had, as Mr Cress knew, a large positive effect on Infinity’s business. It also made future projections especially uncertain.

The renewal meeting that year was held by Zoom, probably on 7 October 2020. It was certainly attended by Mr Cress, his colleague Mr Andrews, and by Mr Daniel Levy of Heath Crawford. Mr Levy still works for Heath Crawford, but did not give evidence. Mr Leyens is confused about whether he did or did not attend the meeting, and gave inconsistent evidence on it in the space of five minutes. Overnight, he consulted his records and decided that he had, after all, probably attended because they showed he had organised it. It makes no difference, however, because he can remember nothing about it. None of the participants took any contemporaneous note of the meeting that has survived.

There is one contested feature of the meeting. It is common ground, and clear from a document circulated the next day, that a decision was taken to remove £200,000 of cover for additional increased cost of working. Mr Cress and Mr Andrews say that this was removed because Mr Levy thought it was not needed when office staff were working from home. Mr Leyens thinks that is unlikely because it would not make sense to him. But although he supposes that the reasons for the removal would have been discussed, he remembers nothing of the meeting at all (if he was there). In the absence of evidence from Mr Levy, I accept Mr Cress’s account. It does not seem to me to be inconceivable that this sum might have been designed to provide cover if a business interruption required new offices to be found and fitted out for clerical and managerial staff. Such an event might have been thought likely to cause just the sort of expense, not directly related to preserving turnover, that additional increased cost of working would cover. If so, remote working might have been an occasion to remove it.

Up to this point there never had been any systematic discussion of what Infinity would have to do if major damage prevented it from using the Northampton warehouse. Nor did such discussion take place on this occasion. Nor did anyone raise declaration linked BI cover.

The meeting left open the sum to be insured for BI cover for 2020/21. On 8 October, the morning after the meeting, Mr Levy sent a document in which the level of BI cover was shown as “TBC - Please see attached GP [Gross Profit] calculation document”. That was accompanied by a further copy of “How to Calculate Gross Profit”, unchanged from the previous year.

On 26 October, Mr Andrews prepared a first calculation for Mr Cress’s review. It is apparent that he had based (or tried to base) the calculation on “How to Calculate Gross Profit”, because he presented text from that document alongside his calculation. He offered two different models, based on different sources for turnover. But he made what appeared to be a mistake if one was following the method set out in Heath Crawford’s guidance. He deducted (as “variable expenses”) only the cost of materials and packaging.

When Mr Cress spotted this, he made his own revised calculation. It too was placed alongside an extract from “How to Calculate Gross Profit”. It proceeded as follows:

The general pattern of this calculation was in line with the recommended approach in Heath Crawford’s guidance and the example given. Indeed, it adhered slavishly to it, because it used the 10 percent upwards adjustment given “as an example”, despite it having no application to Infinity. Mr Cress’s first witness statement suggested that in doing so he was following the guidance. If so, he was not doing so with any thought. The ten percent figure is expressly given “as an example”. One would not need to be a Harvard educated economist (though Mr Cress was) to understand that what was called for was adjustment to reflect the expected future performance of the insured’s business, which might be higher or lower than 10 percent. Mr Cress has never suggested that he expected consistent 10 percent annual growth.

So why did Mr Cress use 10 percent? It is possible that he did not read the guidance carefully, did not spot that it was an example, and did not understand that he needed to make his own forecast. But the most probable answer in my view is that he took a short cut because in his mind the figure was of limited importance, since he did not understand average. Since the figure he arrived at was much higher than the previous year (from £17m to £24.9 million), and comfortably above the sort of gross profit that Mr Cress thought Infinity would make, it felt “good enough”. The result was that Mr Cress partly followed Heath Crawford’s guidance (in deducting variable expenses) and partly departed from it (in making upwards adjustments based on 10 percent rather than any actual forecast).

The remaining questions are about what Mr Cress would or should have done if he had either not been working from Heath Crawford’s guidance document, or had in fact been following it correctly. I consider those counter-factual questions when I analyse causation and contributory negligence.

On 1 May 2021, a fire broke out at Cygnia’s warehouse. It became unusable. Infinity took prompt and effective steps to find alternative premises and equip them in time to restart operations for its busy pre-Christmas trading period. It nevertheless suffered extensive loss of turnover, and equipping alternative premises came at a heavy cost. In due course, assisted by an assessor, it negotiated a settlement of its claim with Aviva. Aviva appointed Sedgwick as its loss adjuster.

Sedgwick recommended a settlement. Its final report makes clear that the settlement was commercial (in the sense that it proposed pragmatic compromises in the face of uncertainty, rather than being a definitive assessment of Aviva’s total liability). It is not easy, therefore, to quantify precisely what deductions it made, or on what basis. Mr Cress, however, has given some evidence about the basis of the settlement, on which he was not challenged. For present purposes, there are two key points.

First, in relation to that part of the claim which was made up of the fit-out costs of alternative premises, Sedgwick allowed some of the claim but not in full. There were many costs for physical equipment—racking, lockers, sprinklers, and the like—which Sedgwick reduced to 25 percent of the total sum spent, to reflect the residual value of the equipment at the end of the indemnity period. In total, prior to average, Sedgwick was willing to allow £1,335,830 in relation to the fit-out costs (in addition to a payment for tenant’s improvements of £130,000). That left Infinity bearing, Mr Cress says, £1,303,936.

The evidence does not, however, suggest that any deduction was made at this stage for anything other than residual value. Sedgwick did not refuse or reduce any part of the expenditure on the ground that it had not been economically worthwhile.

Second, the entire BI claim was reduced by 24 percent, for average, based on the difference between the sum insured (£24.9m) and the full exposure under Aviva’s policy (£32.8m). It is not in dispute that is the correct calculation. After making that deduction, Aviva offered £9.25m. Infinity accepted that offer.

In most cases, the broker’s contractual (and common law) duty is correctly and fully stated as a duty to use “reasonable skill and care in and about obtaining insurance on [the client’s] behalf”: JW Bollom & Co Ltd v Byas Mosley & Co Ltd [2000] Lloyd’s Rep IR 136, 140 (Moore-Bick J). Heath Crawford’s terms of business did not alter that duty. Whether something the broker has done, or not done, demonstrates a lack of reasonable skill and care is a factual question. That is why expert evidence is likely to be admissible. It will always be dependent on the facts and circumstances of each case.

A major part of the broker’s role is to bridge a gap between the client’s knowledge and its own. The ideal client deeply understands its business, its appetite for risk, its capacity to bear loss, and its budget for insurance. The ideal broker deeply understands the cover available in the market, its terms and cost, and how to obtain it. The broker must learn enough about the client’s needs and business to make sensible recommendations. It must tell the client enough about insurance to enable an informed decision (not necessarily one the broker agrees with) and an effective purchase. What is required to do that with reasonable care varies. A multinational corporation with a multi-billion-dollar turnover and a dedicated risk-management department needs different treatment than an individual buying cover for a startup business. And bridging the gap does not mean eliminating it. The client will always know more than the broker about its business. The broker will nearly always know more than the client about insurance. Mr McGrath described what was needed as a “dialogue”, which is a fitting description.

The cases provide instructive general guidance about the expected standard of brokers in a number of relevant respects, though always in terms which leave ample scope for case-by-case analysis:

Those duties apply on renewal as much as on original placement of a risk: Beazley Underwriting Ltd v The Travelers Companies Inc [2011] EWHC 1520 (Comm), [2012] Lloyd’s Rep IR 78, [135]–[136]. I accept Mr McGrath’s evidence (which is common sense, and consistent with Eurokey) that what must be done to discharge them will vary. A broker comes to renewal with an existing fund of knowledge. The broker need not begin each renewal by pretending to forget all that and start again. But it cannot simply assume that renewal is all that is required, even if nothing appears to have changed. It must apply its mind to the client’s present circumstances and the sufficiency of the cover to those circumstances.

If breach of duty is established, the court must decide what loss it caused. In this respect, the burden of proof lies on the claimant: see JW Bollom & Co Ltd v Byas Mosley & Co Ltd [2000] Lloyd’s Rep IR 136, at 142. There has been no dispute between the parties before me about that.

In some cases, it will transpire that a loss was caused both by the defendant’s breach of duty and by some lack of care on the part of the claimant. In such circumstances, the law no longer requires the court to decide between these competing causes and either allow full recovery of damages, or refuse it altogether. Instead, under the Law Reform (Contributory Negligence) Act 1945, the court can apportion responsibility and reduce damages proportionately.

Section 1 of the 1945 Act provides that where a person suffers damage “as the result partly of his own fault and partly of any other person”, his claim is not defeated, but is subject to a reduction “to such extent as the court thinks just and equitable having regard to the claimant’s share in the responsibility for the damage”.

“Fault” in this context does not mean breach of duty, since we do not owe ourselves duties of care. It means simply a “failure to guard against a risk which is reasonably foreseeable” and one that is “unreasonable in all the circumstances”: see Barclays Bank plc v Fairclough Building Ltd [1995] QB 214 (CA), 226. As Philips J put it in Youell v Bland Welch & Co Ltd (No 2) [1990] 2 Lloyd’s Rep 431, 460, in a brokers’ negligence case the question is whether the insured has neglected “what would be prudent in respect of their interests”. Courts are reluctant to expect a claimant to take steps to identify the defendant’s breach of duty (Dunlop Haywards (DHL) Ltd v Barbon Insurance Group Ltd [2009] EWHC 2900 (Comm), [2010] Lloyd’s Rep IR 149, [210]). Numerous cases attest to a reluctance to criticise the insured for not having done the broker’s job, when the broker has not done it. But it is always a question of the particular facts, and such defences may succeed where the insured was sufficiently experienced that it could be expected to have taken that precaution: see the cases cited in Dunlop Haywards at [211]–[212]. Nor do I think that such reluctance will be shown when the thing the insured has not done properly is something that it needed to do itself, and had the experience and knowledge to do, such as providing figures which it had to the broker: see Eurokey at [157].

If the claimant is at “fault”, then in determining what is “just and equitable” the court pays attention to their relative “responsibility” for the damage. That includes the relative significance of their fault (how badly the claimant and the defendant fell below applicable or reasonable standards), the rationale of those duties or standards, and the causal significance of the conduct of each: see e.g. Reeves v Metropolitan Police Commissioner [2000] 1 AC 360 (HL). It follows that both fault and causation may feature twice in any assessment of contributory negligence. They feature first as threshold questions: without at least some want of prudent care that amounts to fault, from which the damage “partly results”, there is no power to reduce damages at all. They feature again as factors in assessing what is “just and equitable”, in which the (relative) degree of fault and (relative) causal significance are important.

Heath Crawford admits that it breached its duty of reasonable care by providing inaccurate information about how to calculate the insured sum for BI purposes. It accepts, rightly, that to supply generic information which was not applicable to the actual policy that it had placed (and intended to place) was a breach of its duty.

At the end of “How to Calculate Gross Profit” was a disclaimer:

“Whilst we are able to provide quotations for varying indemnity periods and information about how to calculate the sum insured, we do not accept any responsibility for the adequacy of your indemnity period and sum insured - and in some instances, you may need to consider the assistance of a suitable professional service.”

Heath Crawford does not suggest that this disclaimer absolves it of responsibility. It was designed to make clear that Heath Crawford could not undertake responsibility for making the calculations (which required a detailed financial understanding of the insured’s business). It did not mean that Heath Crawford was not responsible for providing accurate guidance on what the insurance policy required to be calculated. See Eurokey at [86(1)].

The breach was serious. It went beyond a mere omission to give guidance. It consisted of the positive provision of inaccurate and misleading guidance, on two occasions, accompanied by specific recommendations to rely upon and follow the inaccurate guidance note.

From before the trial began, the experts agreed that a reasonable broker would have recommended declaration linked cover to a client in Infinity’s general position. Heath Crawford kept the embers of a defence alive, however. It suggested that, whatever might have been its duty in a normal case, it was not a breach of duty so far as Infinity was concerned, because Infinity had made it crystal clear that it did not want declaration linked BI cover. By the end of the trial that contention was advanced in a very muted way. If any part of it survived to closing submissions, I reject it.

It is not for a broker to force upon its client a type of cover that the client does not want, even if it disagrees with that preference, and even if it is a foolish preference. I have accepted that Ms Kenner said Infinity preferred not to have declaration linked cover for 2018/19. However, in my view that would not be an adequate answer to Infinity’s complaint, for two reasons.

First, even when Ms Kenner first expressed her preference (and even if it had been described as a “policy”), a reasonable broker would not have accepted it without providing further advice. A broker must respect a client’s informed decision, even if it is a bad one. But it must make sure that it is an informed decision. Heath Crawford needed to be sure that Ms Kenner understood the disadvantageous consequences of her decision. Those included the risk that underinsurance would lead to every claim being reduced by average. That was an important point to hammer home because it is an aspect of insurance that may not be obvious to the typical client, even an otherwise financially literate one.

However, Mr Leyens did not do that. He did not make sure that Ms Kenner understood the implications of underinsurance, or the difficulty and importance of estimating the sum insured. He did not make sure that Ms Kenner understood that the price Infinity would pay for certainty about the premium was uncertainty about recovery if it suffered a loss and had to make a claim.

Second, even when a preference has been expressed, the reasonable broker would check that it remained a genuine and informed preference at renewal, especially as circumstances change. I accept that it is unnecessary to reprise at length, year after year, a topic that has already been thoroughly explored in the recent past. But a reasonable broker will be alert to changes of circumstance, and will at least want to make sure that important decisions (especially if they involve doing something that the broker would not recommend) continue to operate.

Here, however, the topic had never been thoroughly explored, and circumstances were changing. One change was that Mr Cress replaced Ms Kenner. Mr McGrath agreed that was relevant, and the case law supports him: see Eurokey [86(7)]. Covid, too, was something that everyone knew about. The increased unpredictability it caused was obvious. The most basic inquiry into how Infinity’s business was going would have shown how much it had been affected. Because no notes were taken of the renewal meeting that year, I cannot know how far that was explored; but it plainly should have been, at least in outline. Mr Leyens said it would have been. Even looking at the figures for BI cover that Infinity was providing, Heath Crawford could have seen a jump from the £17m insured in 2018/19 and 2019/20 to nearly £25m in 2021. That showed just the sort of variability that might well make declaration linked cover desirable.

It is likely that Heath Crawford should at least have mentioned the topic in 2018/19. I need not decide that. But it is certain that when it came to the 2020/21 renewal, it should have been raised afresh. The combination of the Covid’s effect on Infinity’s business, the upwards jump in the sum insured, and the continuing requests from Mr Cress for information about how to calculate the sum insured for BI cover would have prompted any reasonable broker to advise that declarations linked cover would match Infinity’s requirements manifestly better than the cover it was buying.

The difference between the experts here is difficult to express definitively, because much depends on circumstances. Both agree that the broker must take reasonable care to obtain and maintain a “detailed” understanding of the client’s business. “Detailed” is a chameleon term, whose colour changes depending on whether one has in mind a contrast between the merely superficial, or imagines some sort of microscopic analysis. In Eurokey Blair J said that it was not necessary for a broker to make a “detailed investigation”, and brokers will generally find it sufficient to work from succinct summaries of the key points. I take it that what is required is enough detail to identify the main risks that the client faces which might require insurance, usually an accurate roadmap. Mr McGrath stresses that a broker is not a “detective”. But, again, the dividing line between the reasonably curious and the impertinently sleuth-like is hard to put into words. The broker is not expected to second-guess or audit the information it is given. But it is necessary to follow up reasonably obvious gaps or uncertainties. Intelligent follow-up must be part of the dialogue that Mr McGrath says is necessary.

Heath Crawford knew that Cygnia’s warehouse was critical to Infinity’s operations. It knew that Infinity depended upon it operating. That had become apparent, if it was not apparent before, from the cyber-incident in 2019. It knew that Cygnia owned the warehouse (apart from some machinery and computer equipment that Infinity insured). It knew that Infinity relied on Cygnia’s staff and systems to pick and despatch goods. It knew that it needed its own staff to operate in the same physical space. It knew that the warehouse contained racking and equipment. It should have been apparent that the warehouse was a critical point of failure.

On the other hand, I do not think that Heath Crawford knew, or could have guessed, whether Infinity would need to find an alternative warehouse very quickly, or that its arrangement with Cygnia was difficult to replicate elsewhere, or that it was likely to involve large fit-out costs. For all Heath Crawford knew, there might be no difficulty finding suitably equipped alternative premises easily.

The main reason it did not know, however, was that it had not asked. In agreement with both experts, I find that the question “How will you cope if there is a major problem at the warehouse?” is one that the reasonable broker would have focused on. The possibility that the warehouse might become unavailable was not an obscure detail that needed to be ferreted out. It was an obvious risk. Heath Crawford could not reasonably be expected to guess what would then happen. But it was a question that obviously needed to be addressed, and never was. It was not enough to leave it to Infinity to raise, because its significance depended on appreciating what BI insurance is likely to cover, and its limits. To ask that question would not have been to turn detective, but simply to engage in the dialogue that Mr McGrath rightly identified as central to the broker’s role.

Mr Leyens suggested that it was enough that the BI gross premium document emphasised the need to consider a “worst case scenario”, and that this should have prompted Infinity itself to raise the issue. I do not agree. That comment was made specifically in relation to calculating the sum insured. It was not sufficient to raise the general topic of additional expenses, and it was not enough for Mr Leyens to sit back and wait for Infinity to raise it.

I therefore conclude that Heath Crawford was in breach of duty in this respect, too. In reaching that conclusion, I do not find that Heath Crawford should, based on what it had already been told, have recommended additional cover. It never knew enough to make such a recommendation. It did not get to the point of recommending cover to meet Infinity’s needs, because it had not adequately found out what the needs were. I therefore simply find that a reasonable broker would have raised the question I have identified above. That question, if asked, would have revealed the difficulties, and a potential gap in coverage which Infinity could have looked to plug. What would then have happened is a question of causation, which I consider below.

It is clear from Mr Cress’s evidence (which in this respect I accept) that the misleading guidance in “How to Calculate Gross Profit” did, in fact, lead to under-estimation. It follows that this error was indeed, as Infinity maintains, a cause of the loss it suffered when its claim was reduced by average. But for the mistake, Infinity would have been fully insured. I am satisfied that is so because the documentary evidence shows that, whatever other mistakes he may have made, Mr Cress specifically considered and specifically applied the definition of “variable expenses” by reference to the management accounts. It is probable that if he had been given a different (and accurate) definition, that is what he would have applied.

I also find that if Heath Crawford had, in 2020, raised declaration linked cover with Mr Cress, he would have instructed Heath Crawford to obtain it. By late 2020 he was confronting, every day, the profound uncertainties of estimating future revenue. Everyone agrees that declaration linked cover is not more expensive than sum insured cover. The possibility of an adjustment premium becoming due if profit was better than expected would have been of no moment. Mr Cress would have been confident that if that happened the money would be there to pay it. If cash flow had ever been a serious impediment to declaration linked cover (which I doubt), it certainly was not by 2020. It would have been a completely simple decision, and one that Mr Cress would certainly have taken.

Heath Crawford submitted that Mr Cress’s evidence does not directly say that he would have chosen declaration linked cover if it had been raised. That seems to me to be an unduly pedantic view. Mr Cress says that it would have been exactly what Infinity needed. It would have been. And Mr Cress would obviously have chosen it, for precisely that reason.

I attach no weight to the fact that Mr Cress must have read about declaration-linked cover in the BI guidance document, but did not ask about it. That does not show that if it had been properly presented to him he would have declined it. The document did not describe such cover in detail, or explain its advantages. It would have struck Mr Cress (and any reasonable person) simply as an irrelevant reference in a generic document to a type of cover he did not have. He would have had no good reason to follow it up. An insured cannot be expected to follow breadcrumbs scattered through the documents to discover about suitable insurance products that the broker could, and should, simply have recommended.

I have held that Heath Crawford was in breach of duty by not initiating a dialogue with Mr Cress about what would happen if there was a serious and long-term problem with Cygnia’s warehouse. The more difficult question is where that discussion would have led.

Infinity’s original pleaded case is that “cover for the fit-out costs of alternative premises” should have been obtained. Two possible forms of cover were suggested: inclusion as tenant’s improvements, or insuring them as part of the BI cover, as “additional increased costs of working” (“AICOW”).

By the time the trial started, the experts agreed that coverage as tenant’s improvements was not an option, but that there were a number of possibilities:

I do not accept that the sort of technically complex and undoubtedly unusual cover represented by a first-party property insurance on Cygnia’s warehouse, or by an engineering all risks policy, would have occurred to Mr Leyens. Without knowing its terms, limit, coverage, or likely premium I can have no assurance that it would have been attractive to Infinity either. On the other hand, AICOW cover was ready to hand, and a type of cover that Infinity had itself purchased.

A property-based cover might offer more benefit (because AICOW cover still requires a reduction for residual value after the indemnity period, which a property cover probably would not). But neither party mounted a case, and neither expert suggested, that AICOW cover could not have been reasonably recommended. AICOW was a type of cover that Infinity had bought in the recent past, apparently to cover the costs it would incur if its offices were damaged. It is the only form of cover specifically pleaded by Infinity that the experts consider would have been available. Although reduction for residual value reduces the claim, it does so presumably in part because there is a residual value, which the insured retains, and presumably affects premium. Mr Cress says only that he would have agreed to buy “appropriate cover”; the experts agree that AICOW would have been a form of appropriate cover; and it is probably the cover that Mr Leyens or Mr Levy would, had the dialogue taken place, have recommended.

For those reasons, I conclude on a balance of probabilities that if Heath Crawford had raised the questions I have found should have been raised, it would have led to a dialogue which would have resulted in a sum reasonably sufficient to cover necessary equipment that Cygnia owned being added as AICOW cover under the BI policy. I do not find, on a balance of probabilities, that property cover would have been recommended or agreed.

The first question is whether Mr Cress was at “fault” at all in how he calculated the sum insured. I have no hesitation in saying that he was. He had been provided with a reasonably clear explanation of how to calculate the sum insured. It was plainly prudent, in Infinity’s own interest, to follow it carefully, and reasonable in all the circumstances to do so. That did not require Mr Cress to know or understand anything about insurance. It simply required him to understand what was required to take a gross profit figure based on 2020 figures, and inflate it upwards based on forecast growth for 2021, 2022, and 2023. He certainly had, and Heath Crawford had every reason to believe he would have, the skill and information required to do that.

But in fact he did not. He chose to inflate it not by any actual forecast for Infinity’s growth, but by the 10 percent exemplar figure given in the document. A reasonable person in his position would not have done that. Even though he did not understand the full implications of getting the figures wrong, the document he was working from made it clear that it was important to take care. It clearly explained that it was essential not to arrive at a sum that was too low. The 10 percent figure given as an example was manifestly irrelevant to the exercise.

I therefore conclude that Infinity was indeed “at fault”. How serious that fault was is, of course, a different question.

The second question is whether the loss that Infinity has suffered is “partly the result” of that fault. If Mr Cress had carefully done what he was advised to do, would the outcome have been different? In closing submissions, Mr Smiley (for Infinity) was inclined to approach the question as one about what Mr Cress would have done. I am not sure that is quite right. It would certainly not be enough to say that if Mr Cress had not taken one unreasonable approach, he would have taken a different unreasonable approach. On the other hand, I think Mr Smiley is right to submit that if I conclude that Mr Cress would have taken a particular approach, that it would have been within the range of what is reasonable, and that it would have made no difference to the outcome, then I could not conclude that Infinity’s fault was causally relevant at all.

Mr Cress’s current view, which he admits is entirely retrospective, is that if he had taken a more considered approach it would have made more or less no difference to the sum insured that he would have calculated. He says that he would have used a projected growth rate for profits of 26 percent for 2020, a projected growth rate of zero for 2021, and a projected growth rate of 5 percent for 2021. If he had done that then (on the gross profit figures he was working to, with the wrong deduction for variable expenses) his forward projection to the end of year 2 would have resulted in a sum insured of £25.3 million. In other words, it would have been almost exactly the figure he arrived at using a flat 10 percent year on year uplift.

In late 2020, nobody knew how far Covid signalled a long-term change in shopping habits. On one view, it might herald a permanent “new world”, in which revenue gained during Covid represented a new floor from which growth would continue in line with long-term trends once Covid restrictions ended. In internal documents Mr Cress labelled this assumption, or something like it, “Covid base”. Or it might be a blip, in which case one might expect that over time revenue would revert to roughly where it would have been if the trend growth previously experienced had been allowed to take place. That might mean that for some time after Covid restrictions ended, growth flattened or even became negative when the Covid boom was compared to trend growth. In internal documents, Mr Cress labelled this assumption “Covid cliff edge”.

On top of this there were, for Infinity, additional possible upwards and downwards pressures. The upwards pressures included a slew of initiatives to increase sales and margin. The downwards pressures might include the tendency of growth to slow as a company’s market position matures, and the natural limits on demand for a company that depended on a finite supply of children in need of gifts.

All this goes to show that any forecasting exercise in late 2020 was not clear cut, especially for 2022 and 2023. In my view it is only possible to paint in broad strokes.

Contemporaneous evidence in documents created between June and December 2020 shows two things with reasonable clarity. First, Mr Cress’s view at the time was that the long-term trend growth of Infinity from before Covid was around 20 percent. That is the figure he chose in modelling that he carried out in June 2020. It is, as he pointed out in cross-examination, a little higher than the historic figures show, but not materially so. It seems to have been the figure on which Board forecasting prior to Covid had been based.

Second, there were specific forecasts for 2021 (which was expected to be at least partly affected by Covid), not made at exactly the same time as the renewal, but which are likely to reflect thinking around that time. Despite how Mr Cress subsequently spun them to Sedgwick, I accept that internal forecasts which were designed to spur staff efforts are likely to have been placed deliberately at the upper end of any possible range. But I do not accept that they (much less the Board forecasts) were completely unrealistic. Nor, in fact, does Mr Cress: for he accepts a projected growth of 26 percent for 2021 is what he would have used, and that figure (a little higher, but explicably higher because of an unexpected problem in December 2020 which would flatter 2021’s figures in comparison) was used in a Board forecast for 2021 prepared at the end of the year. I think it is likely that Mr Cress would have taken approximately that figure for 2021.

What is entirely lacking, contemporaneously, is any estimate of the 2022 and 2023 figures, to which Mr Cress had not yet turned his mind. The essential rationale that he presented in his evidence was that he would have expected 2022 to show no or negative growth, by comparison with 2021, as growth reverted to long term trend, and 2023 to show only very modest growth. He prayed in aid a concern, beyond the effect of Covid, that initiatives to increase profits might have fizzled out by 2022, and that Infinity’s maturing business and specialised market would make sustained high growth increasingly challenging.

I have already given my reasons for treating such retrospective assessment, from Mr Cress, as calling for caution. It seems a remarkable coincidence that Mr Cress should alight upon growth rates which just happen to validate the result he reached when he calculated the sum insured on an entirely different and (in this context) arbitrary basis. There is not a single document which points to either of the figures he now relied on for 2022 and 2023 ever being used for any planning purpose. They are pure hindsight.

Nevertheless, it might be said, they could still be valid. If one made some assumption along the lines of the “Covid cliff edge”, and some further pessimistic (but not unthinkable) assumptions that Infinity had already tapped most of the market demand for personalised gifts for babies and children, and exhausted innovative margin-boosting initiatives, it might be a “defensible” scenario (to use a term that Mr Cress often used).

That would have involved taking a highly pessimistic view of the future. The contemporaneous evidence does not support the idea that these were the assumptions that Infinity was making at that time. Although no forecast was ever made for 2022 before the policy was renewed, the forecasts made for 2021 at the end of 2020 assumed that by the second half of the year growth would have resumed. In other words, it was assumed in a budget prepared in November 2020 that the second half of 2021 would see growth (including a 20 percent year on year increase in new customers), even in comparison to the Covid-affected second half of 2020. It makes little sense to suppose that Mr Cress, having assumed when looking at 2021 that growth would be strongly positive by late 2021, would then have decided that it was going to come to a halt in 2022, even though for most of 2022 one would be comparing months unaffected by Covid. Mr Smiley suggested this specific point was not put to Mr Cress. But even if it was not put in exactly that way, Mr Parker repeatedly pressed Mr Cress to explain why his contemporaneous forecasts looked nothing like the arguments he now advances, and beyond his explanation that they were “advocacy” positions for various optimistic purposes, he offers no explanation.

These forecasts, in other words, made something like the “Covid base” assumption, in which after a minor correction growth would resume at something close to pre-Covid trends. Similar assumptions of steady growth (in excess of 20 percent) in 2022 and 2023 were made in June 2021. Those documents post-date the period I am concerned with (and the latter estimate post-dates the fire). But it is not credible that Mr Cress should have assumed in November 2020 that there would be no growth in 2022 and 5 percent growth in 2023, when he never produced any projection which resembles even the general pattern that those figures would require, but always projections which assumed that growth would resume upwards from a higher base established by Covid.

In my view, if Mr Cress had turned his mind to a real forecasting exercise in October 2020, he would prudently have projected growth at 26% for 2021 (which would include some Covid benefit) and at 20% for 2022 and 2023 (consistent with his assumed long term growth trend and the “Covid base” model). I cannot of course be certain about the precise figures, which might have been a little higher or lower. Those figures represent my best estimate, bearing in mind that Mr Cress (had he been acting prudently) would have been alive to the importance of not underestimating the sum insured. I cannot accept that Mr Cress, acting prudently, and aiming to produce a reasonable estimate of projected growth and avoid underinsurance, would have adopted a degree of pessimism which is to be found in none of his other forecasts.

In tabular form, starting from the gross profit assumed (albeit wrongly) by Mr Cress when he carried out his calculation in October 2020, the figures would be as follows:

It follows that Mr Cress’s careless failure to apply a reasonable methodology to inflating the 2020 figure that he had produced into 2022 and 2023 was a cause of the loss. There were, in the end, two mistakes in his calculation. One mistake was that the 2020 gross profit figure was too low. If that mistake, which was due to Heath Crawford’s breach of duty, had not been made, Infinity would have had full insurance, even if the 2020 figure had been inflated by a flat 10 percent. The second mistake was that no real growth forecast had been made, but the figure inflated by an arbitrary 10 percent. If reasonable forecasts had been used instead, Infinity would have had full insurance, even starting from the wrong gross profit figure. Each mistake was a sufficient cause of the loss: avoiding either would have mitigated or eliminated the effect of the other.

That is also true in relation to the failure to recommend declaration linked cover, or to investigate AICOW cover (in so far as that would have avoided average). Infinity would still have been fully insured on a sum insured basis, if Mr Cress had inflated the incorrect 2020 gross profit figure appropriately, for whose inadequacy he was not at all responsible, by a reasonable actual forecast of revenue growth, rather than an arbitrarily selected 10 percent figure, for which he was.

It follows that I need to consider whether, and if so how far, it is “just and equitable” to reduce damages.

In terms of causative potency, Heath Crawford’s breaches were the more significant, or at least the more clearly significant. I can be highly confident that, if Heath Crawford had provided accurate guidance about gross profit or had recommend declaration linked cover, the loss would have been entirely avoided. I cannot be quite so confident about Mr Cress’s carelessness. He would not have used the 26 / 0 / 5 percent factors, which would have implied a Covid cliff-edge assumption which he would not have made, and it would not have been reasonable for him to make. But the 26 / 20 / 20 percent factors I think most likely would always have been within a zone of judgment and arguments could certainly be made that a somewhat lower forecast was reasonable. Even they would have left Infinity very slightly underinsured. Anything less than them would have left a bigger gap yet. That unavoidable uncertainty and vagueness about whether care by Mr Cress would have entirely avoided, or only significantly mitigated, the damage seems to me to be relevant to responsibility.

In terms of relative fault, each breach deserves separate consideration.

In relation to Heath Crawford’s provision of information about how to calculate the sum insured, I consider Heath Crawford’s fault to be serious. This was not just a failure to provide guidance: it was the careless provision of misleading guidance, repeatedly, and with specific instructions to follow the misleading guidance on a point which was obviously critical. It was not a close call or a momentary slip, but a fundamental and repeated error.

Mr Cress did not understand the pervasive effect that underinsurance was likely to have, and did not therefore understand the full implications of what he was being asked to do. That is some excuse, but not, in my view, a very strong excuse. Mr Cress was told that something called “average” might apply to “ANY claim”. So it evidently mattered, even if he did not understand its full implications. He was fairly warned of the importance of arriving at a sufficient sum insured. It cannot be suggested that Heath Crawford’s guidance document, whatever its deficiencies, induced Mr Cress’s careless error, which was quite independent of its incorrect description of how to calculate the basic gross profit. Moreover, the “disclaimer” at the end of the document, although it is not an answer to liability is, as Mr Parker submitted, a further fair warning to the recipient that the calculation required care, and that it is for the insured to take it or ask for advice.

In relation to the declaration linked cover, the balance lies more in Infinity’s favour. One of the reasons why declaration linked cover is likely to be superior is that it avoids the common risk of error associated with estimating the sum insured. In other words, one of its very merits is that it may save the insured who makes the sort of careless mistake that Mr Cress made. To my mind that is an important factor.

Nevertheless, this does not seem to me to be a complete answer. As Lord Hoffmann put it in Reeves (at 371) “It is commonly the case that people are held liable in negligence for not taking precautions against the possibility that someone may do something careless and hurt themselves, like diving into a shallow swimming pool, but I do not think it has been suggested that in such cases damages can never be reduced on account of the plaintiff’s contributory negligence”. The cover Heath Crawford was recommending, though sub-optimal, would still have provided full insurance had Infinity carefully followed even the incorrect guidance that it had been given. Although it is just and equitable that Heath Crawford should bear much of the loss, it is also just that Infinity should bear some part of it.

The third breach consists of the failure to make inquiries which would (I have held) have led to AICOW cover being obtained. In so far as one consequence of that would have been that average would not have applied to fit-out costs, it is another factor telling against Heath Crawford, but it affects only a very small part of the loss on that account. In so far as it had any further consequence, it would not be a matter for contributory negligence at all, because Mr Cress’s carelessness or carefulness in making his projections would have made no difference.

Taking all those matters into account, I consider that Heath Crawford has a large part of the responsibility for the loss caused by averaging, but that Infinity (through Mr Cress) also has substantial responsibility. In my judgment it is just and equitable that damages for that aspect of the loss should be reduced by 20 percent. Had I been considering only the provision of the misleading guidance document, I would have made a higher reduction of 40 percent; but my assessment reflects the fact that Heath Crawford was negligent in multiple respects, and that some of its negligence consisted of failure to recommend a type of policy which is in part designed for the very purpose of reducing the risk that careless errors with figures will reduce cover.

There is no difficulty in quantifying the claimant’s loss for the breaches of duty in providing a misleading explanation of how to calculate the sum insured, and in failing to recommend and advise properly about declaration linked cover. It is represented by the amount by which the settlement was reduced by the application of average. That is £2,921,053, before any reduction for contributory negligence. With the appropriate reduction for contributory negligence, the damages in that respect are £2,336,842.

The fit-out costs require a little more thought. In principle, the settlement included a sum for fit-out costs as increased costs of working. That sum was potentially reduced by (a) disallowance of expenditure to the extent that it had not reduced turnover, (b) a reduction for the residual value at the end of the indemnity period of the items that were acquired, and (c) average.

Had coverage been (as I have held it should and would have been) included specifically as AICOW, the second deduction would still have been available to Aviva, but not the others. In principle that would mean that AICOW could have provided additional cover.

There is, however, no evidence that the potential reduction for uneconomic expenditure was ever invoked. As I explain above, so far as can be seen, Aviva agreed to pay for the fit-out expenses as increased cost of working and did not make any reduction because they were not cost-effective. It simply made reductions for residual value (which would have been made anyway), and then applied average.

Had AICOW been in place, therefore, Aviva’s settlement offer for fit-out costs would have been higher only to the extent that they would not have been reduced by average if they had been specifically insured. Since Infinity is already being compensated for the deduction due to average, no further damages should be awarded. Infinity trailed a coat that the existence of additional cover would have had unquantifiable but material positive impacts on Aviva’s generosity. The evidence does not enable me to make such a finding.

I therefore find that Infinity has failed to establish that the Heath Crawford’s failure to instigate more detailed discussions of what would happen in Cygnia’s warehouse became unusable caused any measurable financial loss beyond that already caused by the other breaches, for which it is being compensated.

The parties served evidence from surveyors dealing with various individual items in the re-fit loss. By the end of the trial only two items were in dispute. The first related to the value of software, the second to what deduction (if any) property insurers would have made for betterment in respect of a sprinkler system.

If separate cover had been arranged and if it had covered those costs at all, its terms would probably have been framed to meet the entire costs of the fit-out, including these. “Betterment”, for example, simply makes little sense if one is considering a policy whose starting point is to insure someone not for the loss of value or reinstatement cost of its own property, but for the contingent cost of restarting a business in a new location.

But that demonstrated, in my view, that Infinity has not proved the essential foundation of its “property insurance” case. The questions make no sense—are literally unanswerable—unless one knows what the alternative “property-like” cover would have provided, what it would have covered, and what its bespoke terms for indemnity would have been. A damages claim that is not undergirded by any concrete answer to that sort of question is too speculative. Since the evidence simply does not enable more than a completely speculative answer to the question “what would the property insurance have covered?”, such an award would be an exercise in the arbitrary.

It follows that Infinity is entitled to damages of £2,336,842 in respect of all the breaches.

For the reasons set out above, Heath Crawford acted in breach of duty by (a) providing a misleading and incorrect explanation for how to calculate gross profit, (b) failing to recommend declaration linked cover, and (c) failing to explore what additional coverage Infinity might need to meet its costs if Cygnia’s warehouse became unusable. Those breaches of duty resulted in Infinity being underinsured, so that average was applied to its BI claim. As matters transpired, they probably had no further consequence. Infinity contributed to its loss by carelessly failing to apply even the flawed guidance correctly, and it is just and equitable that its damages should be reduced by 20 percent for that contributory negligence. In the result, Infinity is entitled to damages of £2,336,842.

I am grateful to counsel for their thorough, realistic, and helpful submissions. The parties, having seen this judgment in draft, have agreed an order, and I shall make an order in those terms.