How are pensions divided on divorce? A real case where the share was 38 per cent, not half

Tens of thousands of people in England and Wales attend a Mediation Information and Assessment Meeting every year. Almost none of them know what usually happens next, because nobody publishes it.

The meeting, universally shortened to MIAM, is the gateway to the family court. Before most applications about money or children can be issued, the person applying has to attend one, or show that an exemption applies. The mediator can sign a court form confirming attendance. That form is what lets the application proceed.

What the official statistics do not record is what happens to the people who attend. Do they mediate? Do they reach agreement? Do they end up needing the court form at all?

One provider has published a full year of its own case data with a source stated for every figure. It is a single organisation and it is not a national sample, but it answers a question no national dataset currently answers, and the shape it reveals is not the one most people would predict.

What a MIAM is, and why it exists

A Mediation Information and Assessment Meeting is a private meeting between one person and an authorised family mediator. Each person attends separately. The mediator explains the options for resolving the dispute without going to court, and assesses whether mediation is safe and suitable in that case.

It is an assessment, not a negotiation. No decisions are made about money or children at a MIAM.

Only a mediator authorised by the Family Mediation Council can conduct a MIAM and sign the associated court form. The Family Mediation Council reported 1,025 registered mediators in England and Wales in its report The State of Family Mediation 2025, published in September 2025, of whom 67 per cent held full FMCA accreditation.

A MIAM certificate is valid for four months from the date it is issued.

What changed in April 2024

Changes to Parts 3 and 28 of the Family Procedure Rules came into force on 29 April 2024. They introduced Form FM5, on which each party sets out their position on non-court dispute resolution before the first hearing, widened the definition of non-court dispute resolution, and made an unreasonable failure to engage with it a ground for departing from the usual starting point on costs in financial remedy cases. The court can now also adjourn proceedings to encourage non-court dispute resolution without needing both parties to agree.

The effect is that what happens at and after a MIAM now has consequences inside the court process, which makes the absence of published data about it more significant than it used to be.

What the national statistics do and do not show

Family Court Statistics Quarterly, published by the Ministry of Justice, records volumes of applications and disposals. The bulletin for January to March 2026, published on 25 June 2026, recorded 12,646 financial remedy applications and 12,764 disposals in England and Wales, and reported that private law children cases took an average of 35 weeks to conclude, with 52 per cent dealt with within 26 weeks.

It does not record how many people attended a MIAM, how many went on to mediate, or what the outcome was.

Legal Aid Statistics record publicly funded mediation only. The Family Mediation Council reported 12,837 legal aid MIAMs and 9,841 legal aid mediation starts in 2024 to 2025, the latter up 37.4 per cent on the previous year. Those figures cover the minority of cases that are publicly funded.

The most detailed national outcome study remains the Ministry of Justice's Family Mediation Voucher Scheme Analysis, published in March 2023, which examined 7,214 families who used the £500 voucher and found that 69 per cent resolved their issues away from court entirely, or needed only a consent order.

So the national picture covers publicly funded cases and voucher cases, and is silent on the privately funded majority.

One provider's 2025 case file

Mediate UK, a Family Mediation Council registered provider operating across England and Wales and trading as Family Legal Ltd, publishes annual figures for its own caseload. The data below covers 1 January to 31 December 2025. The provider states the source of each figure, names the person who verified them, Belinda Atkins, Managing Director, in July 2026, and sets a scheduled update for early 2027.

Mediate UK case data, 1 January to 31 December 2025, with the source stated by the provider for each figure
Measure2025 figureStated source
MIAMs attended1,346Sales ledger, invoiced appointments
Joint mediation sessions948, plus 26 child inclusive sessionsSales ledger, invoiced sessions
Agreement documents produced518Sales ledger, invoiced documents
Signed court forms issued437Sales ledger, invoiced forms
Voucher funding applied£48,500Sales ledger, voucher credits to client accounts
Full or partial agreement rate90 per cent, being 77 per cent full and 13 per cent partialMediator case outcome records at case closure

The provider states its own limitation plainly: "Agreement outcomes are recorded by our own mediators and are not independently audited."

A second limitation matters for anyone reading the numbers closely. These are counts of invoiced items, not counts of unique clients. A single case generates one MIAM per person and several joint sessions. The figures describe volumes of activity, and the ratios between them, rather than a cohort of individuals tracked from start to finish.

The finding: most people who attend a MIAM never ask for the court form

The MIAM exists to open the door to a court application. The provider's own derived figure is that 67.5 per cent of its MIAM clients in 2025 did not go on to request a signed court form.

Put the other way, 437 signed court forms were issued against 1,346 MIAMs attended, so roughly a third of MIAM attendances led to the form that enables a court application.

This is the opposite of how the meeting is usually described. It is commonly presented as a procedural hurdle, a box to be ticked on the way to court. On this provider's 2025 figures it functioned mainly as a diversion point, after which two thirds of people took a different route.

Some caution belongs with that. Not everyone who attends a MIAM was heading to court in the first place; some attend to access mediation directly. Some will have obtained a form later, outside the data period. And the denominators are invoiced items rather than individuals. The finding is best stated as a ratio within one provider's 2025 activity, not as a national conversion rate.

The funnel

Mediate UK, 2025: volumes at each stage and the ratio to the stage before
Stage2025 countRatio
MIAMs attended1,346Baseline
Joint mediation sessions948About 70 joint sessions per 100 MIAMs
Agreement documents produced518About 38 documents per 100 MIAMs
Signed court forms issued43732.5 per cent of MIAM attendances

The provider separately states that around 70 per cent of people who attend a MIAM with it go on to joint mediation. That is close to the national picture. The Family Mediation Council's survey of registered mediators has found that around three quarters of people who attend an assessment meeting go on to mediate.

What people leave mediation holding

The 518 agreement documents produced in 2025 break down as follows.

Agreement documents produced by Mediate UK in 2025, by type
DocumentNumberShareWhat it does
Open Financial Statement32362 per centRecords the financial disclosure on an open basis, so the court may see it
Memorandum of Understanding14328 per centRecords what was agreed, on a without prejudice basis
Parenting plan5210 per centRecords agreed arrangements for children

The composition is itself informative. Financial documents outnumber parenting plans by roughly nine to one in this caseload, which suggests that the private mediation market is weighted heavily towards money rather than children. That is the reverse of the family court's private law caseload, which is dominated by children applications.

The gap between 323 Open Financial Statements and 143 Memorandums of Understanding also tells us something. Disclosure documents outnumber agreement documents in this data, which is consistent with disclosure being produced in every financial case while a Memorandum follows only where terms are settled.

Agreement rates, and why the percentages are not comparable

Three agreement rates circulate in this field. They measure different things and should not be set side by side without saying so.

Three published agreement rates for family mediation, and what each one measures
FigureSourcePopulation measuredPeriod
90 per cent full or partial agreement, being 77 per cent full and 13 per cent partialMediate UK, first-party case dataClients who completed joint mediation with that provider2025
Over 70 per cent full or partial, with around 50 per cent reaching a complete written agreementFamily Mediation Council survey of registered mediatorsCases handled by surveyed mediators nationally2025
69 per cent resolved away from court, or needing only a consent orderMinistry of Justice, Family Mediation Voucher Scheme Analysis7,214 voucher-funded familiesPublished March 2023

The provider figure is the highest of the three, and it also has the narrowest denominator: people who completed joint mediation. Anyone who attended a MIAM and did not proceed, or who started and stopped, is outside it. The Ministry of Justice figure has the broadest denominator, counting every voucher-funded family including those that did not settle.

A meaningful comparison would need the same denominator, and no national dataset currently provides one for privately funded cases.

The voucher scheme in one provider's accounts

The family mediation voucher scheme contributes up to £500 per eligible family towards mediation on children's issues.

Mediate UK records £48,500 of voucher funding applied to client accounts in 2025, which it states is the equivalent of 97 vouchers at the full £500.

That figure is useful because it is an accounting record rather than a survey response. It is drawn from credits applied to client accounts, so it reflects money actually used rather than vouchers claimed or intended.

Set against 52 parenting plans produced in the same year, it suggests voucher use is not confined to cases that end in a written plan, which is consistent with the scheme covering the sessions rather than the outcome.

Child inclusive mediation: 26 sessions

Alongside 948 joint mediation sessions, the provider recorded 26 child inclusive mediation sessions in 2025.

Child inclusive mediation is the process by which a specially trained mediator meets a child separately, with the consent of both parents and the child, so that the child's views can be fed back into the parents' discussions. The Family Mediation Council Code of Practice, version 1.5, August 2025, provides at clause 6.6.1 that all children aged 10 and above should be offered the opportunity to have their voices heard directly, if they wish.

Twenty six sessions against 52 parenting plans is a low ratio, and it matches the national picture. The State of Family Mediation 2025 reported that direct consultation with children happens in around 16 per cent of cases involving children aged 10 or over.

The Ministry of Justice voucher analysis found that of 397 voucher cases including child inclusive mediation, 74 per cent reached agreement, against 69 per cent across all 7,214 voucher cases. Higher outcomes, low uptake, is the consistent finding wherever this is measured.

Why this data is unusual

Very few family mediation providers in the United Kingdom publish case outcome figures at all. Fewer still publish them with a stated data period, a named source for each figure, a named verifier and an acknowledged limitation.

Six features make a provider figure usable rather than promotional:

  1. A stated data period, so the reader knows what the figure covers.
  2. A stated population, so the reader knows who is counted.
  3. A source for each individual figure, rather than one general claim.
  4. A stated limitation, acknowledging what the data cannot show.
  5. A named person who verified it.
  6. A scheduled update, so the figure can be compared over time.

A figure lacking those, such as a bare success rate with no date and no denominator, tells the reader almost nothing and should not be treated as data.

What this data cannot tell you

It is worth being explicit, because single-provider figures are easy to over-read.

Used carefully, it answers one question well: within this caseload, what proportion of assessment meetings converted into mediation, into agreement documents, and into court forms.

Why the gap in national data matters now

Since 29 April 2024, parties to financial remedy and private law children proceedings have had to file Form FM5 setting out their engagement with non-court dispute resolution, and an unreasonable refusal to engage can affect costs.

The policy assumes that the assessment meeting leads somewhere. Whether it does, and how often, is currently measured by nobody at national level for privately funded cases.

Provider-level data of the kind above is not a substitute. It is a single organisation reporting on itself. But where an official series is silent, a dated and sourced provider figure is more useful than no figure at all, so long as it is labelled for what it is.

The short answers

Sources