UK Law Firms

Law is the best-benchmarked sector in this series, and what the benchmarking shows is uncomfortable. Fee earners record 773 chargeable hours a year against a target of 1,100. That figure is published by the profession’s own representative body, drawn from 145 firms with over £1.1 billion of combined fee income, and it is an improvement on the year before.

What the sector publishes

• Chargeable hours recorded per fee earner773Observed
• Chargeable hours target1,100Observed
• Salary costs as a share of fee income63.5%Observed
• Year-end lockup days146Observed
• Firms reporting year-on-year fee growth71%Observed
• Mid-sized firms approached by private equity in 202570%Observed
• Modelled collection from time worked68%Modelled

Law needs four terms where every other sector needs three, because recording is a separate act from performing the work and from billing for it. The second term is the one the profession under-manages, and the reason is structural rather than cultural: unrecorded time produces no artefact — no write-off to review, no bill to query, no entry in any report. Meanwhile the year’s improved partner profit came substantially from interest on client balances, a rate effect that arrived without effort and reverses when rates fall.

Licensed conveyancers and will-writers sit outside the boundary and inside the competitive picture: they are not the high-street firm’s peers, they are its substitutes, competing for the highest-volume work from a lower cost base under a lighter regime. The volume process practice matters for a different reason. It records the most hours and charges the least because discretion has been removed and time is captured by the system rather than by a fee earner remembering — which is precisely the repeatability private equity entered the sector to buy, and it is available to any firm.

Legal demand is durable and broadly non-discretionary. The constraint in this sector is not the market and has not been for some time — it is entirely internal, and the benchmarking data says so more clearly than any commentary could.

Participation in the survey requires a finance function able to produce the data and the management interest to submit it, and both properties correlate with being better run. Very small firms and unincorporated sole practices are correspondingly under-represented, so 773 chargeable hours against an 1,100 target is, if anything, a flattering picture of the profession rather than a harsh one. The model therefore treats the panel as a reliable description of the observable, managed end of the market and does not extrapolate it to the sector as a whole.

Revenue begins with purchased fee-earner capacity and is lost in four successive stages. The rate card governs only the last of them, which is why rate reviews so rarely produce the improvement the firm expected.

The hours corridor is observed from a 145-firm survey and the 1,100 figure is that survey’s own benchmark target, not a regulatory requirement. Charge-out rates are modelled throughout: no body publishes UK legal rate data.

The rate is the fourth term in the identity and it operates on whatever survives the first three. A ten per cent rate increase applied to 773 recorded hours at 85% realisation produces less additional fee income than closing a fifth of the hours gap at the existing rate — and the second requires no client conversation, no competitive risk and no repositioning. Interim billing is the one middle move that attacks two terms at once: cash arrives during the matter, and a bill raised close to the work is written off less.

The same sector produces very different economics when capture, practice-area mix, leverage and lockup change. Equity partner fee-earning time is normalised to a market replacement rate before economic EBITDA, and interest income on client balances is excluded from all three corridors.

Base central model: 790 chargeable hours per fee earner · 85% realisation · £205 modelled charge-out rate · 63% salary cost share

The specialist boutique carries a modelled 26% central margin against the high-street practice’s 11%, on eleven fee earners rather than nine, and the gap is not explained by effort or client quality. Three mechanisms compound: the rate is roughly double because no substitute provider outside solicitor regulation offers the same work, realisation runs higher because specialist bills are queried less, and lockup is lower. Salary share falls as a consequence rather than through cost control. Specialism takes years to build; capture discipline takes months, which is why the model sequences it first.

Recorded fee-earner hours bind before client demand in every released archetype. At 773 recorded against an 1,100 target, the sector operates at roughly seventy per cent of its own capacity benchmark while 71% of firms report fee growth. The constraint is internal.

The 327-hour gap is not idleness, and treating it as such produces the wrong intervention. Four things consume it: legitimate non-chargeable work, matter work performed and never recorded, time reconstructed from memory and systematically understated, and time deliberately not recorded because it is expected to be written off anyway. The last two are the recoverable portion and share one cause — recording happens away from the work rather than at it. The supervision ceiling then compounds it: adding fee earners without a department head tier converts partner chargeable time into partner supervisory time.

Law does not usually have a demand problem. It has a capture and lockup problem, and both occur inside work the firm has already won and is already performing.

Ask a prospect for these twelve figures. Most firms can produce fee income, debtor days and hours recorded. Hours recorded within 24 hours of the work, write-offs by approver and non-chargeable hours by category are the ones that are missing, and they are the three that bracket the capture loss.

The modelled waterfall loses twelve points to unrecorded time, ten to write-off at billing, eight to lockup ageing and two to bad debt. The largest component happens first and is the only one that leaves no trace. A partner reducing a bill has made a decision that can be reviewed and learned from; work never written down produces no entry, no exception report and no variance. That asymmetry is why firms manage the smaller loss — they have write-off reports because write-offs exist as records — and why the three figures always missing are the ones bracketing capture.

Private equity approached seventy per cent of mid-sized UK law firms during 2025. Deal count is rising while ticket size falls, and investor focus has moved from personal injury and conveyancing into Court of Protection, family and employment work. Capital is buying process repeatability, and the ladder below is what it is testing for.

Deal count is rising while money deployed falls — fifteen transactions in 2026 to date against £250m deployed in 2025, down from £534m in 2024 — which means capital is buying smaller firms more often rather than retreating. The practice-area movement confirms what it is buying: past personal injury and conveyancing, into Court of Protection, healthcare, family and employment work, all repeatable enough to document and supervise at scale. That is why rung three is the hinge. A firm attempting rung five without capture usually finds its chosen specialism less profitable than it believed.

Modelled effect on normalised EBITDA of a one-standard-step move in each driver, regional full-service archetype, widest first. Hours recorded ranks first, which is unusual in this series and is a direct consequence of how large the observed gap against target is.

The sensitivity ordering differs from every other pack in the series, and the reason is arithmetic rather than preference: no other sector has a single measured variable running thirty per cent below its own published benchmark. The rate lever is not worthless — it ranks second — but it operates on whatever survives capture, so a firm raising rates first applies a multiplier to seventy per cent of its own effort. The live downside is behavioural: reading interest-flattered partner profit as evidence the model works defers exactly the capture and lockup work the figures call for.

Technology creates value in law only when it moves one of the drivers on the previous page. This is not a software catalogue and none of these interventions gives legal advice. Each names the driver it moves, the data it requires and the commercial test that decides whether it worked.

Almost every firm can be sold practice management software, and for most that is the wrong first move: a new system with the same recording behaviour produces the same 773 hours. Capture is a change in when recording happens, not in what records it, because time reconstructed on Friday under-states in one direction and always has. None of the six interventions gives legal advice or forms a professional judgement, and all six operate on matters the firm has already won — which is what makes them measurable against a known baseline inside a single quarter.

The sector model becomes commercially useful at the moment a named firm is scored against it. The eight dimensions below convert everything in this model into a diagnostic that can be completed in a single working session with the firm’s own data.

The eight dimensions are scored against the firm’s own figures rather than against an interview, and in law that distinction is especially sharp: partners are confident about utilisation their firm has never actually calculated by department. Where a figure cannot be produced the dimension scores one, which is the finding rather than a penalty. The output is a profile, not a total — a firm at level four on lockup governance and level one on capture discipline needs something quite different from one with the reverse profile, and an aggregate would conceal it.