We have adopted a specialist team approach to our practices for many years. We feel that this is the way our clients want us to work, and that specialisation leads to the provision of a better service.
We have adopted a specialist team approach to our practices for many years. We feel that this is the way our clients want us to work, and that specialisation leads to the provision of a better service.
Insights
05/11/2025
By Oliver Manley and Emily Girvan-Dutton
On 9 October 2025, His Honour Judge Grimshaw handed down judgment in Perrin v Walsh [2025] EWHC 2536 (KB). This judgment provides both an overview of the balancing exercise a court will undertake when exercising their discretion whether to admit or exclude surveillance evidence under CPR 32.1 and a stern warning to defendants as to the consequences that may follow when their agents act in an “extremely poor” manner when obtaining such evidence [79]. Perrin v Walsh also highlights the importance the court places on the probative nature of surveillance evidence, even where there are concerns of tainting and manipulation.
The following article provides a helpful guide to the caselaw surrounding surveillance evidence and the factors a court will consider in its discretion, examines defendant and claimant arguments for and against its admission, and provides some tips for making and responding to applications to rely on surveillance in practice.
In February 2020, the Claimant was knocked off her motorcycle in a road traffic accident where she sustained multiple injuries, allegedly including, amongst other injuries, mild traumatic brain injury, PTSD, Somatic Symptom Disorder and Functional Neurological Disorder. Liability was not in issue.
For two years, between June 2022 and June 2024, the Defendant instructed a surveillance company, The Surveillance Group (“TSG”) to carry out covert surveillance on the Claimant. Footage from this surveillance was then served on the Claimant and an application was made to rely on it as well as amend the defence to plead fundamental dishonesty.
On review of the footage served, the Claimant made criticisms which largely fell into four categories [5]:
These criticisms formed the basis of the Claimant’s objection to the Defendant’s application which alleged that the Defendant’s had engaged in a “deliberate and cynical” manipulation of the footage making it tainted and manifestly unfair [6]. In view of this prejudice, the Claimant’s position was that the surveillance evidence should be excluded.
Surveillance evidence has long been used as “legitimate weapon” for Defendants, as HHJ Collender QC put it in Douglas v O’Neill [2011] EWHC 601, to prevent “the successful advancement of a case, which is based on untruth” [42].
The court has a discretion under CPR 32.1 to prevent the Defendant from relying on such evidence. This discretion must be exercised in accordance with the overriding objective to deal with cases justly (Grobbelaar v Sun Newspapers Ltd, The Times, 12 August 1999, CA; O’Brien v Chief Constable of South Wales [2005] UKHL 26).
Issues the court may consider in exercising this discretion include: fairness, public interest, the extent of Article 8 breach, and the limitations of the surveillance footage. The court will undertake a balancing exercise between these considerations, and any prejudice that may arise from them, and the probative value of the surveillance footage. The crucial issue is whether the Defendant can show that the evidence undermines the case of the claimant to an extent that would substantially reduce the award of damages to which she is entitled (per Potter LJ in Rall v. Hume [2001] EWCA Civ 146 at 19).
When considering the fairness of surveillance evidence being admitted, the court will often look at whether its admission would amount to a ‘trial by ambush’. The guidance on this point also comes from Rall v Hume, where Potter LJ stated that once the Defendant has established that the evidence will likely substantially reduce the amount of damages then: “
“it will usually be in the overall interests of justice to require that the defendant should be permitted to cross-examine the plaintiff and her medical advisors upon it, so long as this does not amount to trial by ambush.” [19]
HHJ Collender QC, in Douglas v O’Neill, set out what is meant by an “ambush”:
“In my judgment the issue of ambush comes to this – are the circumstances in which the evidence is disclosed such that the Claimant has a fair opportunity to deal with it, or was the time or circumstances of disclosure such that the court should use its case management powers to prevent the Defendant from relying upon it.” [46].
HHJ Callender QC continued, explaining how a defendant should disclose surveillance evidence to avoid an ambush:
“The Defendant accepts that a defendant in possession of surveillance evidence should make the decision to rely upon it and disclose it as soon as reasonably possible after receiving sufficient material setting out the Claimant’s case, which has been endorsed with a statement of truth so as to enable the surveillance material to be used effectively. If a defendant fails to do so, and the failure to do so, has unacceptable case management implications, then that defendant risks being unable to rely upon the material.” [56]
In Stuart v Kelly [2016] EWHC 3263 (QB), Blake J that there was a duty on the Defendant to act with expedition, once the Claimant has pinned his colours to the mast [28].
There is no hard and fast rule about what length of delay will amount to an ambush, rather the court’s approach has been:
“… an objective one based upon the real effect of the late application to rely upon this kind of evidence on the preparations for the trial and, most importantly, on the trial date itself, particularly if fixed. The time estimate for the trial may change which can affect other cases in the list.” (Hayden v Maidstone & Tunbridge Well NHS Trust [2016] EWHC 1121, Foskett J [31]).
What is clear is that the court will be alive to how much additional work surveillance evidence, if admitted, would require of a claimant. In O’Leary v Tunnelcraft [2009] EWHC 3438 (QB), Swift J considered what work would be required of the claimant:
“Once all the material of the best quality available has been obtained, it will be necessary to obtain statements from the claimant and his witnesses. In addition, the experts (i.e. care, employment, psychiatrist, urologist and orthopaedic experts) will need to see and comment on the footage. … In my judgment, it would only be right for the expert witnesses to see all the additional material together. Quite apart from anything else, this would mean that they would only have to deal with the papers on one occasion rather than on two separate occasions. Addendum reports would then have to be obtained, any discussions would have to take place and any necessary amendments to joint reports” [88].
Swift J was particularly concerned with, not just the impact this amount of work may have on the trial date, but also the impact it may have as a distraction from the “ordinary preparations from trial and from considerations of the Part 36 offers and possible settlement of the case” [89].
What the above authorities show is the courts take delay and concern for ambushing the claimant seriously. Accordingly, this factor is given substantial weight in consideration of whether to admit surveillance evidence. In Douglas v O’Neill, HHJ Callender went so far as to say that:
“At the heart of my consideration when exercising my discretion, must be a determination as to whether or not the Defendant by his advisers has been guilty of delay in producing this DVD film” [72].
Jones v University of Warwick [2003] EWCA Civ 151 concerned surveillance evidence which had been obtained by gaining access to the claimant’s home by deception. Ultimately, despite the concern about the breach of the claimant’s right to privacy, Lord Woolf permitted the surveillance evidence. Lord Woolf identified “two conflicting public interests” in the balancing exercise he undertook to reach this decision:
“Clearly, there is a public interest that unfair, tortious and illegal methods should not be used in general and where they are unnecessary, but the conflicting considerations are on the one side the claimant’s privacy and on the other the legitimate need and public interest that defendants or their insurers should be able to prevent and uncover unjustified, dishonest and fraudulent claims. In the instant case I have no doubt that the latter considerations do and should outweigh the former.” [15]
Lord Woolf continued to explain that the weight to be attached to these two public interests will vary depending on the extent of the breach of the claimant’s right to privacy:
“The weight to be attached to each will vary according to the circumstances. The significance of the evidence will differ as will the gravity of the breach of Article 8, according to the facts of the particular case. The decision will depend on all the circumstances a case in which surveillance evidence was obtained by gaining access to the claimant’s home by deception.” [28]
However, ultimately, it seems that even where the breach of the claimant’s right to privacy has been significant, as in Jones v University of Warwick, this concern will be trumped where the surveillance evidence has probative value:
“Here, the court cannot ignore the reality of the situation. This is not a case where the conduct of the defendant’s insurers is so outrageous that the defence should be struck out. The case, therefore, has to be tried. It would be artificial and undesirable for the actual evidence, which is relevant and admissible, not to be placed before the judge who has the task of trying the case.” [28]
A central issue for courts considering whether to admit surveillance evidence is the limitations of such footage in that a judge can only consider what is shown, and not what wasn’t filmed or left out.
In Cullen v Henniker-Major [2024] EWHC 2809 (KB), HHJ Ambrose explained the court’s approach to circumventing the limitations of surveillance evidence:
“…I accept that if something is shown on the footage, that is determinative of it happening. However I do not accept that the absence of something on the footage is determinative of it not happening” [107]
This approach was endorsed by HHJ Grimshaw in Perrin v Walsh, where he added the following to specifically address the concerns raised by the claimants that the surveillance footage had been manipulated to undermine her claim:
“I would also add that the value of surveillance evidence is in what is seen, not what might have been seen on another day or at another time or place. One must be particularly cautious when, as is alleged in the present case and in Cullen, the footage obtained appears incomplete and where there are gaps in what was recorded. Surveillance footage cannot definitively show what the subject is feeling or how much pain they are in. It is merely a snapshot in time. I also accept that an individual’s condition may vary throughout the course of a day, week, month or year, and maybe in response to activity that had occurred prior the obtaining of such footage. It is therefore only a piece of evidence in the jigsaw.” [18]
HHJ Grimshaw was also at pains to caution that the limitations of surveillance evidence should be handled by medical expert witnesses with care:
“Medical expert witnesses, like the Court, must understand the utility and limitations of such evidence and thus the weight that it should be given. Questions of honesty, dishonesty and exaggeration are matters for the trial judge, not for medical expert witnesses to decide upon, noting that the role of medical expert witnesses is to assist the Court.” [18]
After undertaking a balancing exercise of the above factors, HHJ Grimshaw allowed the defendants to rely on the surveillance evidence. This was despite having “very real concerns about how TSG have approached their assignment in this case” [74].
In line with the authorities above, it seems the probative value of the footage was given a considerable weighting in the exercise of discretion:
“I am not satisfied that the failings identified by the Claimant, as serious as they are, are such that I should exclude what is otherwise clearly probative evidential material. The conduct of the Defendant’s agents and insurers is not so outrageous that the Defence should be struck out, nor does the Claimant suggest this to be the case. This case will be going to trial, and it would be artificial and undesirable for the evidence, which is relevant and admissible, not to be placed before the judge who has the task of trying the case. In my judgment, it would be manifestly unfair to the Defendant to deprive them of the ability to place the surveillance material before the trial judge. The issue of the weight that the trial judge gives such evidence is an issue for the trial judge. It will be open to the Claimant to cross-examine the surveillance operatives and others form TSG and to submit that the evidence is unreliable for the reasons that have been advanced before me.” [76]
Perrin v Walsh, and the authorities before it, are not claimant friendly decisions. HHJ Grimshaw advised that, whilst the authorities tend towards allowing reliance on the surveillance evidence, there are occasions where this won’t be the case:
“There will be circumstances where the conduct of a defendant, or its agents, is so egregious that probative and otherwise admissible surveillance material should be excluded from consideration utilising the Court’s power pursuant to CPR 32.1.” [79]
Serious failings by the defendant’s agents were found in Perrin v Walsh, including the “providing of patently untrue witness statements to the Court, endorsed with statements of truth” [79]. Despite these failings being of “serious concern to the Court”, HHJ Grimshaw was “not satisfied that the threshold has quite been reached such that the surveillance evidence in this case should be excluded, but it was not far away.” [79].
In light of this, it is difficult to contemplate circumstances where the defendant’s agent’s behaviour would meet this threshold and unclear what type of conduct would be considered “so egregious”. This leaves claimants responding to applications to rely on surveillance evidence in a difficult position where the probative value of evidence trumps any question as to poor conduct by the defendant’s agent.
Conduct that the courts do give considerable weight to is any delay by the defendant in giving notice of intention to rely on and in serving surveillance evidence. When this delay amounts to an ambush on the claimant, the court’s approach is often more sympathetic to the claimant, preventing the defendant from relying on surveillance evidence (as was in the case in O’Leary v Tunnelcraft).
Where there has been a delay in serving or poor conduct in obtaining surveillance evidence, and the defendants are allowed to rely on it, hope is not entirely lost for claimants. HHJ Grimshaw in Perrin v Walsh, after making a finding that the defendants can rely on the footage, quoted Lord Woolf in Jones v University of Warwick in which he set out the cost’s implications of a defendant’s agent’s poor conduct:
“As said in Jones, at [30]:
“Excluding the evidence is not, moreover, the only weapon in the court’s armoury. The court has other steps it can take to discourage conduct of the type of which complaint is made. In particular it can reflect its disapproval in the orders for costs which it makes… In addition, we would indicate to the trial judge that when he comes to deal with the question of costs he should take into account the defendant’s conduct which is the subject of this appeal when deciding the appropriate order for costs. He may consider the costs of the inquiry agent should not be recovered. If he concludes, as the complainant now contends, that there is an innocent explanation for what is shown as to the claimant’s control of her movements then this is a matter which should be reflected in costs, perhaps by ordering the defendants to pay the costs throughout on an indemnity basis. In giving effect to the overriding objective and taking into account the wider interests of the administration of justice, the court must while doing justice between the parties, also deter improper conduct of a party while conducting litigation. We do not pretend that this is a perfect reconciliation of the conflicting public interests. It is not; but at least the solution does not ignore the insurer’s conduct.”” [80]
It is clear that where surveillance evidence shows the claimant acting in a way inconsistent with their claim, this will likely lead to admission of the evidence. However, surveillance evidence, even where it doesn’t show anything inconsistent with the claimant’s claim, can still cause the claimant significant issues. These issues largely fall under two headings:
Covert surveillance is often carried out over an extended period, often for anywhere from 18 months to 2 years. The claimants will be followed and filmed for random periods during that time as they go about their day. Footage will also often be taken of claimant’s children if they are present at the relevant time. Some surveillance operatives have gone to extremes to obtain footage including trespass and deception, as in the case of Jones v University of Warwick where the operatives entered the claimant’s home posing as market researchers and used a hidden camera to film the claimant without her knowledge.
This can cause claimants significant distress in having their privacy, and their loved one’s privacy, invaded to such a degree. This distress is especially acute in claimants that are already vulnerable, whether because of pre-existing vulnerability or as a result of their injuries and disability.
Serving surveillance evidence, especially where it is served late, places an additional burden on the parties to obtain statements from the claimant and other witnesses, pass the footage onto their experts, and obtain addendum reports. This additional work both provides a “distraction” from the proceedings, as Swift J in O’Leary v Tunnelcraft was concerned about [89] and may result in a significant delay in proceedings. Swift J, in the same case, posited that it was “inevitable — and indeed I do not think that this is in dispute — that the granting of the defendants’ application would add to the length of the trial” [90].
The types of high value PI claims where the need arises for surveillance evidence have often already been going on for several years. In O’Leary v Tunnelcraft the claim had been going on for more than six years, and this length of time is becoming increasingly uncommon. Swift J was significantly concerned about how this additional delay would affect the claimant, in particular what the impact would be upon an already vulnerable claimant’s psychiatric health:
“… if the defendants’ application were to succeed, it would not be practicable for the trial to proceed. It would place unfair pressure on the claimant’s advisers and would not, in my judgment, be manageable. An adjournment of the case would have serious implications for the claimant. It is already more than six years since his accident and, according to the defendants’ psychiatric evidence, the proceedings themselves are not assisting his recovery. No doubt the adjournment of the case would cause further distress to him. Thus, an adjournment would cause him prejudice, as would the continuation of the trial for which he or his advisers would be unable properly and fully to prepare.” [91]
On the other hand, where surveillance evidence is obtained that shows the claimant acting inconsistently with their claim, the defendant case is that it should be relied on to prevent the claimant from advancing a fraudulent claim.
In Jones v University of Warwick, Lord Woolf upheld the trial judge’s decision to allow the defendants to rely on surveillance evidence. Lord Woolf cited the trial judge’s judgment which endorsed a view that the need to prevent fraudulent claims is a considerable factor in the court’s discretion:
“The judge continued by saying that:
“The primary question for the court is not whether or not to give approval to the method whereby evidence was obtained. It is whether justice and fairness require that this highly material evidence, which contradicts the evidence which she has given to others, should be put to her before the trial judge to enable him to reach a sound conclusion about the true extent of any disability. True, the claimant was herself deceived but there is strong prima facie evidence that she herself is deceiving or misleading the defendants to enrich herself thereby. It is not easy for the defendants to protect themselves against exaggerated claims. Anyone with much experience of personal injury litigation will know that the defendants and their insurers are frequently faced by claimants who suggest that their disabilities are far greater than they are, and large sums of money may be unjustifiably sought. Though such people are rarely, if ever prosecuted, in many cases what they do or seek to do must amount to the crime of obtaining property or pecuniary advantage by deception. In these circumstances I do not believe that the courts should be too astute to prevent effective investigation by the defendants of claimants against them.”” [15]
HHJ Grimshaw in Perrin v Walsh, citing the above, disagreed that the frequency of fraudulent claimants was at the level the trial judge depicted, however he nonetheless agreed that “there is a clear public interest in discouraging and unearthing cases where such fraudulent claims are advanced.” [15].
For both defendants and claimants, making and responding to applications to rely on surveillance evidence can give rise to issues. Below you can find some helpful practice points to help avoid these issues as far as possible:
What Perrin v Walsh makes clear is that, in the balancing exercise, the court gives the greatest weight to the probative value of the surveillance evidence. From the above review of the authorities, it seems that only in exceptional circumstances, where the defendant’s agents conduct has been egregious or the delay has had a considerable impact on the claimant or the trial date, will a defendant not be allowed to rely on probative surveillance evidence. Defendants should not, however, take this to mean they can act in any way they wish, Perrin v Walsh, despite its overall defendant-friendly tone, provides a stern warning to defendants about the consequences that may follow from the conduct of their agents.
To stay updated, follow us on LinkedIn
Call: 2005
Sign up
To be kept up-to-date with our latest news and future events, please complete the short form.
A member of the clerking team will help you resolve your request.
