I am grateful to Aidan Robertson KC for sending me a copy of the decision of  the Competition Appeal Tribunal, Mr Justice Marcus Smith,  in Sportradar AG -v- Football Dataco Limited [2022] CAT 37 a copy of which can be found here. SportsRadar  The decision is important because it provides a helpful and succinct summary of the principles of the electronic “sift” in edisclosure.  It is also significant in relation to subsequent applications for specific disclosure once disclosure has taken place.


“We all must recognise that the disclosure process in this

jurisdiction is one of the foremost tools for getting at the truth and for producing

important documentary evidence for the courts to consider. We must also

recognise that it is one of the most significant cost centres in litigation and that

is a bad thing.”


The Tribunal was being asked for orders for specific disclosure by the claimants.  Disclosure had already taken place and the documents were substantially electronic.  The judge refused the claimants’ application.  The reasons for this are of some significance.


In the past, when the disclosure process was paper or hard copy based – where
documents were contained in filing cabinets or lever arch files – disclosure (or
“discovery”, as it was then known) involved identifying the file storage
practices of the disclosing party, identifying the relevant files within that
universe, and reviewing those files by looking through them, using the
“eyeballs” of a qualified person to ascertain that which was relevant and that
which was not relevant, according to a pre defined standard that was general in
nature. The issues against which a search was to be conducted were not
generally set out. There was no need. The issues were stated in the pleadings,
and there was no purpose of defining precisely how the search was to be
conducted because an in-person “eyeball” review was the only way.
  1. When once the disclosure process was complete, it would not be looked behind,
questioned or re-visited unless good reason could be shown. One such reason
might be the inadequacy of the original process. But generally speaking, the
Tribunal or Court would not look behind the list of documents or disclosure
statement produced by the disclosing party.
  1. A Tribunal or Court might, however, order a further, specific, search for a
particular document or class of document in the appropriate case. That would
not involve a “re-run” of the original process, but a specific and targeted review
for a limited set of documents. That was a process that the nature of paper files
could accommodate.
  1. The position with electronic disclosure is extraordinarily different. There is a
great deal of front loading in terms of the documents that are isolated and then
reviewed. I am not going to go into the detail of the disclosure process
undertaken in this case. I am going to illustrate my point by reference to a
hypothetical process that is (broadly speaking) typical of today’s disclosure
(1) In the first place, it is necessary to isolate the relevant electronic files.
That may be done by reference to the computers operated by various
“custodians” and/or by reference to the servers and/or back up held by
the disclosing party. The universe of documents so produced is generally
speaking vast and an “eyeball” review to differentiate between relevant
(disclosable) documents and irrelevant (non-disclosable documents) is
almost always not practicable or cost-efficient as a first step.
(2) Instead, the volume of material produced for “eyeball” review is
slimmed down or reduced through the use of information technology.
That is possible because electronic disclosure is readable electronically.
Keyword searches or other forms of electronic review are used to effect
this slimming down or reduction (electronic sifting).
(3) The nature of the electronic sifting to be applied is almost always
discussed between the parties and is usually highly contentious. That is
because documents that are discarded as a result of the electronic sift are
never reviewed by a competent (human) professional and – of course –
the computer does not apply its mind to the question of relevance. It
simply carries out the sifting process that it has been instructed to
undertake. As a result, if (for example) the keywords selected do not
cause a highly relevant document to be retained rather than discarded,
then that document will never be considered by a competent
(4) The slimmed down universe of documents produced as a result of the
electronic sift is then subjected to an “eyeball” review, so as to further
exclude irrelevant or non-disclosable (i.e. privileged) material. To
reiterate, for the point is important, the material that is excluded from
disclosure as a result of the electronic sift – which will typically be vast
(hence the electronic sift) – is not further reviewed and will never be
produced on disclosure.
  1. Self-evidently, it is important that the process of review – in particular the
electronic sift – is done appropriately from the beginning. I use the word
“appropriately” advisedly, for it will never be the case that the process will be
perfect. The electronic sift is a binary tool that either excludes or does not
exclude, and it does not (at least, not at the moment) exercise a form of
judgement. Documents that are excluded from professional review will never
see the light of day, and (since no process is perfect) that will include some
relevant documents. That is why it is necessary to consider the process at the
outset and that is why this Tribunal, on this occasion, took considerable time
and effort to set out a series of issues which were debated by the parties,
according to which the parties then conducted their search.
  1. It will take a great deal to persuade this Tribunal to conduct a re-run of such an
exercise. That is because such a re-run will typically require a revised electronic
sift (not, in itself, an issue: such processes are fast and relatively inexpensive)
followed by a further “eyeball” review, which will be expensive and time consuming.
  1. I am not prepared to contemplate such a review at this stage of these
proceedings. Inevitably, that means that certain documents which might exist,
and I underline “might”, and which, had the electronic searches been differently
structured, might have been uncovered will not be uncovered for purposes of
disclosure at trial. We all must recognise that the disclosure process in this
jurisdiction is one of the foremost tools for getting at the truth and for producing
important documentary evidence for the courts to consider. We must also
recognise that it is one of the most significant cost centres in litigation and that
is a bad thing.
  1. The fact is that when a process, in this case done under the close supervision of
the Tribunal and the Court, has been completed, it may well be, and almost
certainly will be, that there are gaps because the search process, with hindsight,
could have been differently done. In such circumstances, absent the exceptional
case, this Tribunal and Court must make do with what has been produced. It is,
of course, completely open to both sides to say that certain inferences can be
drawn from the fact that certain documents might or might not exist, and what
they would say if they did.
  1. That is one of the fact-finding functions of the Tribunal or Court. It is the
function of the Tribunal or Court to decide the issues in dispute, according to
the evidence that is produced before it. That evidence includes not merely that
which is, but that which is not. I want to be very clear that in trying matters
where the documentary record is incomplete, as it almost always will be, this
Tribunal and Court will draw inferences from the totality of the facts.
  1. I am in little doubt that – with the clear benefit of hindsight – internal
communications of Genius regarding Perform might have been better searched
for using a different process. But that is not enough to persuade me to order a
re-run of the process (in whatever form). Indeed, Sportradar’s application falls
well short of what is needed to bring the specific disclosure regime into play.
  1. The disclosure process in this case has been carefully considered and thorough.
I am completely satisfied that it would be disproportionate to revisit matters
now. Sportradar’s application is refused