SHORENET
← Shorenet Guides Family

Social Media Evidence in the Family Court

The court controls what comes in, authenticity is admitted unless somebody objects, and the expert test is stricter

By Dr Anthony Rose-Edwards

Published , last reviewed

Yes, social media evidence is used in family proceedings, and the family court has wide powers to control it. The court decides what evidence it needs and how it is to be put before it, and it can exclude evidence that would otherwise be admissible. Authenticity of a disclosed document is admitted by default unless a party serves notice requiring it to be proved. And in family proceedings other than children proceedings, the test for permitting expert evidence is stricter than the civil one; children proceedings are governed separately.

This guide covers family proceedings in England and Wales. It is general information, not legal advice. Scotland and Northern Ireland have their own rules, and the civil and criminal courts are different again. For the wider picture, see social media evidence in UK courts.

Can social media evidence be used in family proceedings?

In principle yes. Posts, messages, profiles and public online conduct are ordinary documentary evidence, and they turn up across the range of family work.

The practical constraint is rarely admissibility in the abstract. It is relevance, proportionality and the court’s control over what it is willing to receive, which FPR 22.1 sets out in terms.

How does the family court control what evidence comes in?

By direction. FPR 22.1 provides that the court “may control the evidence by giving directions as to” the issues on which it requires evidence, “the nature of the evidence which it requires to decide those issues” and “the way in which the evidence is to be placed before the court”. It adds that the court “may use its power under this rule to exclude evidence that would otherwise be admissible”, that it “may permit a party to adduce evidence, or to seek to rely on a document, in respect of which that party has failed to comply with the requirements of this Part”, and that it “may limit cross-examination”.

Two things follow for anybody planning to rely on social media material.

First, a bundle of screenshots nobody asked for is not self-evidently going in. The question is what issue it goes to, and whether the court wants evidence on that issue in that form.

Second, the power cuts both ways. Non-compliance is not automatically fatal, because the court may permit reliance on a document anyway. That is a discretion to ask for, not a position to rely on.

How is authenticity challenged?

By a notice to prove, in the same structure as the civil courts use.

FPR 22.16 provides that “a party to whom a document is disclosed is deemed to admit the authenticity of that document unless notice is served by that party that the party wishes the document to be proved at the final hearing”. The deadline is the latest date for serving witness statements, or within 7 days beginning with the date of service of the document, whichever is later.

So authenticity is conceded by silence. That is why current practice works most of the time, and why it fails suddenly when it fails: the party who wanted to rely on a screenshot discovers the need to prove it at a point when the post has long gone and the account of how the capture was made is a memory.

What that means in practice is that the quality of your evidence was fixed at capture time. How to preserve social media evidence before it is deleted sets out what to do while you still can.

Is it different from the civil courts?

On authenticity, barely. On experts, materially.

The notice-to-prove mechanism is near-identical: compare FPR 22.16 with CPR 32.19, which deems a party “to admit the authenticity of a document disclosed to him under Part 31 ... unless he serves notice that he wishes the document to be proved at trial”. Both use a deadline of the witness-statement date or 7 days, whichever is later, but the two sevens are not identical: the civil one runs “within 7 days of disclosure of the document”, and the family one “within 7 days beginning with the date of service of the document”.

One further difference, if you also work in the civil courts: the civil rule reaches only documents disclosed under Part 31, and it does not apply to claims on the small claims track, because CPR 27.2(1) disapplies Part 31 and Part 32 other than rule 32.1 there. FPR 22.16 has no equivalent carve-out.

The expert position is where they part company, and the difference is in the test itself rather than in how it is applied.

Permission is needed in both. CPR 35.4 provides that “no party may call an expert or put in evidence an expert’s report without the court’s permission”, and CPR 35.1 restricts civil expert evidence to “that which is reasonably required to resolve the proceedings”.

What differs is the test. In family proceedings other than children proceedings, FPR 25.4 provides that “a person may not without the permission of the court put expert evidence (in any form) before the court”, and that permission may be given “only if the court is of the opinion that the expert evidence is necessary to assist the court to resolve the proceedings”. FPR 25.5 lists further factors the court is to have regard to, including the issues involved, the timetable and cost. Children proceedings are controlled separately, by section 13 of the Children and Families Act 2014 (the test there, in section 13(6), is also that the evidence is “necessary to assist the court to resolve the proceedings justly”), and are otherwise outside this guide.

“Reasonably required” and “necessary” are different thresholds. Whether a particular application meets either is a matter for advice on the case, not something a web page can tell you. The general point is that an expert examination of a disputed screenshot is a step that needs permission, and permission is not a formality.

Would an expert settle a disputed screenshot anyway?

Often not, which is the uncomfortable part.

Examination of an image can sometimes identify signs of editing. It is much weaker at the proposition people actually need, which is that an image was not edited. And one of the two easy routes to a misleading screenshot leaves no trace in the image at all: altering a page in the browser before the shutter goes produces a genuine, unedited capture of an altered page. Every forensic property of that file is authentic, because it is authentic. It is simply authentic to something that was changed first.

It is also worth remembering what an expert is for. FPR 25.3 puts it directly: “It is the duty of experts to help the court on matters within their expertise”, and that duty “overrides any obligation to the person from whom experts have received instructions or by whom they are paid”. The civil rule, CPR 35.3, is in the same words. An expert is not there to support your case, and an instruction that assumes otherwise tends to produce a report that helps nobody.

You can see why detection is the wrong place to put your weight by trying it: screenshots as evidence puts three captures of one conversation in front of you, two of them faked in under a minute.

Points commonly covered before filing

  • Capture now, not later. Everything else is downstream of this.
  • Capture context, not fragments. The thread, the profile, the URL, the timestamps. Taken-out-of-context is the easiest objection in family work and often a fair one.
  • Keep the originals untouched and work on copies. Do not annotate, crop or re-save the original file.
  • Write a contemporaneous note of how, when and by whom the capture was made. This is the part that holds up.
  • Think about what issue it goes to, before volume. The court is deciding what evidence it needs; a proportionate, targeted set is more use than a bundle.
  • Assume you may be asked to prove it, and ask yourself what you would say.

Where Shorenet fits

Shorenet addresses the authenticity question, and only that question.

A capture’s evidence is the web traffic itself, not a picture of the screen; a screen recording can run alongside as a guide to what the user saw, but it is the traffic that is sealed and checkable. The traffic crosses a relay we operate which is independent of whoever is doing the capturing; the relay never decrypts and we never see the content. The result is sealed so that any later change shows, including a change by the person who captured it, and it is built to be checked by anyone holding it, including the other side’s expert, without a Shorenet server, account or cooperation.

Being exact about the tense, because it matters: that is the design, and the validator it depends on is not published yet. Shorenet is pre-deployment. What exists today is a demonstration of the decryption step that runs in your own browser on a sample baked into the page; it is not the validator and it does not accept a bundle of your own.

The limits, stated because they are the point:

  • It shows what the platform sent. It does not show the content is true, and it says nothing about what anybody meant.
  • It does not detect deepfakes or synthetic media, and we do not claim to.
  • It cannot reach anything deleted before capture.
  • It captures what you open in a capture browser, not native mobile apps. Telegram on the web works today; WhatsApp is in development. See the FAQ for the current position.
  • Admissibility, relevance and weight remain matters for the court. Nothing here changes what FPR 22.1 lets the court decide about evidence it is willing to receive.

How it works sets out the mechanism and what you have to trust us for.

Sources

Primary sources, each fetched and checked on 12 October 2026, and checked against the Ministry of Justice consolidated rules on 12 October 2026. Every rule quoted above is quoted from these.

Checked in the original on 12 October 2026: