Executive summary
- Time leaks twice on its way to the bank: once as a write-down before the invoice goes out, and once as lock-up after it does. Firms measure the second, argue about the first, and rarely quantify either by matter.
- The UK median lock-up is 134 days (144 including unbilled disbursements), on 121 firms with £1.2bn of fee income. The most widely quoted international figure is 93 days — but that's North American data, so benchmarking against it tells you almost nothing.
- Median chargeable hours are 807 per fee earner against the 1,100-hour target the same survey cites. That gap is recording discipline, not laziness, and it's invisible in any annual firm-level benchmark.
- In the worked example below, £7.87m of recorded time becomes £6.40m of banked fee income — 81p in the pound. Three points of realisation on exactly the same hours would be worth £236k.
Legal & professional at DataHexis
"Matter profitability, utilisation rates, WIP tracking, and client reporting." This article is about the one number that connects all four, and why no annual benchmark will ever show it to you.
— from the Industries section of the DataHexis websiteAsk a managing partner about cash and you'll get a lock-up figure. Ask about productivity and you'll get chargeable hours. Ask what happened between the two — how much recorded time turned into an invoice — and the room usually goes quiet.
That silence is the expensive part. Hours and lock-up sit at either end of the same pipe, and the biggest loss happens in the middle, when a bill goes out for less than the time recorded against it. It rarely gets escalated, because nobody has to approve it. It is simply how the invoice was always going to be raised.
The money leaks twice
Realisation = value billed ÷ value of time recorded. The first leak. Recovery = cash received ÷ value of time recorded. The second leak, and the only one that pays salaries.
A write-down reduces recorded time before the bill is raised: a quoted fee, a fixed fee, a courtesy discount, supervision time nobody was ever going to charge for. A write-off cancels part of an invoice after it has been raised.
Every firm has a process for write-offs. They're visible, they need authority, and somebody has to explain them. Write-downs are different: they happen at the moment of billing, one matter at a time, and they're recorded as "that's just what the file was worth". Individually each one is defensible. In aggregate they are the single largest difference between a good year and an average one.
A write-off is an argument. A write-down is a habit. Only one of them gets escalated.
The reason this stays hidden is structural, not cultural. Recorded time lives in the practice management system, billed value lives in the billing ledger, cash lives in the office account, and the cost of the person who did the work lives in payroll. Getting realisation by department, by fee earner and by matter type means joining all four, and no single system you already own does that.
Benchmarking against the wrong country
Here's where it gets worse. Search for law firm benchmarks and the top results give you a realisation rate of 88%, a collection rate of 93% and total lock-up of 93 days. Those figures come from tens of thousands of firms and they're genuinely useful — in North America, which is where the data is from.
A UK firm judging itself against 93 days is comparing against a different billing culture, different payment norms and different work mix. Crowe's 2025 UK benchmarking puts lock-up at 138 days, improved from 145. The Law Society's 2026 survey says 134. Those are the numbers to be measured against.
Two more traps in that comparison. "Realisation" doesn't mean the same thing on both sides of the Atlantic: UK practice usually means billed over recorded, while US usage often folds in collection as well, so an 88% figure may not be measuring what you think. And medians don't add up — in the same US dataset the two lock-up components are 43 and 32 days while total lock-up is 93, because the median of a total isn't the sum of the medians. If your report adds component medians together, it's wrong before anyone reads it.
There is also a gap nobody has filled. Search as long as you like: no UK survey publishes a realisation or write-off benchmark at all. The Law Society and Crowe both report lock-up as a single combined figure, and neither splits WIP days from debtor days. So the middle of the pipe has no UK benchmark, which is precisely why it needs an internal one.
What a point of realisation is worth
This is the arithmetic that gets a partnership's attention, because it doesn't require anybody to work longer hours.
Take a firm recording £7.87m of time a year at standard charge-out rates. One point of realisation is £79k of fee income. Three points — the difference between a firm that reviews write-downs weekly and one that reviews them never — is £236k, on precisely the same recorded hours, with no new clients and no rate rise.
Lock-up works the same way, but in cash rather than profit. At 117 days, that firm has £2.05m of its own money sitting in WIP and debtors. Getting to 90 days releases £473k once, permanently — more than most firms' overdraft facility, and considerably cheaper.
Both numbers are available today, in systems the firm already pays for. They are simply in four different places.
A worked example: the department view
Here's what the finished reporting looks like for a mid-sized private practice: about £6.4m of annual fee income, roughly 40 fee earners and four equity partners, across residential conveyancing, private client, family, and commercial and dispute resolution. It's a composite of typical firm structures rather than a single client, and the figures are illustrative — but they're built to sit inside the benchmarks cited throughout, and every number reconciles to the others.
The conventions matter as much as the numbers, so they're stated rather than assumed:
| Line | % of fee income | Note |
|---|---|---|
| Fee-earner salary cost | 32% | Salaries and on-costs of fee earners, excluding equity partners |
| Support staff and management | 8% | Total staff cost of 40%, against 38.9% in Armstrong Watson's 2024/25 review |
| PII and professional costs | 5% | 4.8% in the same review; small firms run higher |
| Premises, IT and compliance | 25% | Non-salary overheads of 30% here, against 28.4% in the Law Society's 2026 survey |
| Operating profit before partner drawings | 30% | Stated before drawings; Armstrong Watson reports a 27.7% net margin, 27% for firms with 1–7 equity partners |
Matter margin below therefore means margin after the fee earner's own time cost, not after overheads. Four systems feed the model:
Department heads open the same view each Monday. Private client is shown here because it's the department with the slowest cash and the most fixed-fee work — the combination that hides the most.
Department view: private client
Month to date, refreshed overnightRealisation sits first, because it's the number the department can change this month. Colour never travels alone: each status has a word and an icon too.
Where the department's cash is sitting
£390k of WIP and £251k of debtors. The £86k that is more than 90 days old is the part to act on: WIP that old is usually an unmade decision about a bill, not work in progress.
Live matters running over their fee
| Matter | Basis | Fee or estimate | Time value | Over by | |
|---|---|---|---|---|---|
| Estate administration · PC-41207 | Fixed fee | £4,500 | £6,180 | +37% | ⚠ Review |
| Estate administration · PC-40988 | Hourly, estimate given | £7,500 | £9,240 | +23% | ⚠ Review |
| Trust and tax planning · PC-41455 | Fixed fee | £2,750 | £3,190 | +16% | ! Watch |
| Will and LPA package · PC-41502 | Fixed fee | £1,200 | £1,340 | +12% | ! Watch |
| Deputyship application · PC-41180 | Fixed fee | £3,400 | £3,620 | +6% | ✓ On track |
On an hourly matter with an estimate, going over means a conversation with the client. On a fixed fee it means a conversation with yourselves, about how the next one gets priced.
The partner view: four charts
Department heads need this month. Partners need to know where the firm's money goes between the timesheet and the bank, which departments are carrying the cash, and which work is worth taking at the price it's currently quoted.
From recorded time to banked cash
£7.87m recorded, £6.60m billed (realisation 83.9%), £6.40m banked (recovery 81.3%). The write-down is six times the size of the write-off, and only the write-off ever reaches a partners' meeting.
Lock-up by department
The firm-wide figure is 117 days — better than the 134-day UK median, and still £2.05m of its own cash. Conveyancing collects on completion; probate and family don't, and averaging the four hides both facts. This is also why WIP days and debtor days belong on separate axes: one is a billing decision, the other a collection one.
Fixed-fee matters: margin after the fee earner's time
One matter in eight lost money before a single overhead was allocated. The median is healthy; the tail is where the fee scale needs rewriting. An average would have shown neither.
Margin by work type
Conveyancing isn't the problem: its spread is. A median of 58% with a 10th percentile below zero is a pricing and scoping question, not a performance one.
Realisation by department
The firm-wide 83.9% is a blend of four very different billing cultures. Conveyancing's 74% is mostly fixed fees meeting unfixed scope.
Under the hood (for the data team)
The reporting is the easy half. Getting four systems to agree on what a matter, an hour and a pound are is the work.
Data model and the joins that breakStar schema, cost rates, and the disbursement question
One small warehouse sits in front of Power BI and everything is conformed on the way in. Whatever the practice management system — LEAP, Actionstep, Osprey, Proclaim, Partner for Windows, Clio — the model conforms it to one shape:
| Table | Grain | Why it matters |
|---|---|---|
| Fact Time Entry | Fee earner × matter × day | Units, charge-out rate and cost rate as at the date of the entry — the denominator for everything |
| Fact Bill | Bill line, linked to matter | Billed value against time value gives realisation, and isolates the write-down |
| Fact Receipt | Receipt, with allocation date | Recovery and debtor days; part-payments allocated to the right bill |
| Fact Disbursement | Item, flagged billed or unbilled | The 134-vs-144-day question: unbilled disbursements in or out of WIP |
| Dim Matter | Matter, with fee basis and work type | Fixed fee, hourly with estimate, or conditional; the fee basis changes the maths |
| Dim Fee Earner | Person, grade, department, cost-rate history | A pay rise shouldn't rewrite last year's matter margins |
Two joins break in almost every firm. The first is time recorded after a bill has been raised, which quietly reopens a closed matter's realisation unless bills and time entries are reconciled on billing date rather than matter. The second is part-paid bills across multiple matters for one client, which wrecks debtor days by matter if receipts aren't allocated properly.
Decide the disbursement convention once and publish it. The Law Society reports lock-up both ways precisely because firms do it differently, and the two answers are ten days apart.
The measuresSample DAX for realisation, recovery, lock-up and fixed-fee margin
Time Value Recorded =
SUMX ( 'Time Entry', 'Time Entry'[Units] / 10 * 'Time Entry'[Charge-out Rate at Entry] )
Realisation % =
DIVIDE ( [Billed Value], [Time Value Recorded] )
Recovery % =
DIVIDE ( [Receipts Allocated], [Time Value Recorded] )
Lock-up Days =
VAR WIPDays = DIVIDE ( [Unbilled Time Value] + [Unbilled Disbursements], [Fee Income R12] ) * 365
VAR DebtorDays = DIVIDE ( [Debtors], [Fee Income R12] ) * 365
RETURN WIPDays + DebtorDays
Fixed-fee Matter Margin % =
DIVIDE ( [Agreed Fee] - [Time Cost at Cost Rate], [Agreed Fee] )The detail that decides whether anyone trusts the output is Charge-out Rate at Entry and its twin, the cost rate as at the date of the entry. Look either up live and last year's realisation changes every time rates are reviewed, which is the fastest way to lose a partnership's confidence in a dashboard.
For fixed-fee margin, use the fee earner's cost rate, not their charge-out rate. Charge-out rate measures ambition; cost rate measures what the work actually consumed.
Already have a legal BI product, or good PMS reports?Where this adds something, and where it doesn't
Some practice management systems ship genuinely good reporting, and a couple now include certified Power BI connectors with pre-built report packs. If you're getting activity, billing and cash out of one of those, you already have the basics, and a specialist legal BI product bolted onto your PMS will cover them too.
What those tools can't easily do is reach outside the system that owns them: payroll cost rates, the accounts ledger, several years of history from before your last system change, and the private client spreadsheet that never made it into the PMS. Matter profitability needs cost, and cost lives in payroll. That's the join we build, and it's also the reason we'd tell you not to bother if your data already lives in one system and your reports already answer the question.
What we never need to see
Every conversation with a firm reaches this point, usually within ten minutes, and it should. So to be direct about it: this reporting needs financial metadata, not the substance of anyone's matter. Matter references, work types, fee bases, time units, rates, bill and receipt values, dates. Not attendance notes, not correspondence, not advice, not the file itself.
Practically, that means we're a processor under UK GDPR and work under a data processing agreement with a named sub-processor list and UK hosting. It also means the firm's COLP and COFA keep their obligations: the SRA is explicit that a firm remains accountable for work carried out through others, so the sensible order is due diligence first, data second. Cyber Essentials is the realistic floor for this kind of engagement, ISO 27001 is what larger firms and panel questionnaires tend to ask for, and a DPIA is worth supporting even where one isn't strictly required. A read-only reporting layer over your ledgers is not a third-party managed account and doesn't touch client money — worth saying out loud, because it's the first thing a cautious COFA will want to rule out.
Sole practitioners and small firms
Of the 8,900-odd firms the SRA regulates, about one in seven is a sole practitioner, and firms with one to seven equity partners run a net margin of around 27%. For a firm that size the arithmetic is starker, not softer: one badly priced fixed-fee retainer or one matter sitting in WIP for six months is a measurable share of the year.
The reporting is correspondingly simpler. One page: realisation this month, WIP and debtors over 90 days, live matters over their fee, and margin by work type. If you bill on fixed fees for conveyancing, probate or immigration — where the SRA's transparency rules already have you publishing prices — then knowing the margin behind each published price is not optional for long.
Frequently asked questions
What is a good lock-up figure for a UK law firm?
The Law Society's 2026 Financial Benchmarking Survey puts the median at 134 days excluding unbilled disbursements, or 144 including them, down from 146 and 156 the previous year. Crowe's 2025 UK benchmarking reports 138 days, improved from 145. Anything below about 120 days is genuinely good in the UK; the 93-day figure you'll find at the top of search results is North American and not a fair comparison.
What's the difference between realisation and recovery?
Realisation is the value billed divided by the value of the time recorded — it measures the write-down at billing. Recovery is cash received divided by the value of time recorded, so it also captures write-offs and bad debt. UK and US usage of "realisation" differs, which is why any dashboard should state the formula next to the number.
How do I work out profitability on a fixed-fee matter?
Take the agreed fee, subtract the time recorded on the matter valued at each fee earner's cost rate (not their charge-out rate), and express the result as a percentage of the fee. The catch is recording discipline: unrecorded time on fixed-fee work doesn't just cost margin, it destroys the measurement. Look at the distribution of matters rather than the average — the loss-making tail is the part worth pricing for.
Why is my WIP so high?
Usually one of four things: work finished but not billed, matters with no billing milestone, time recorded on files that will never be billed, or genuine delays in court and transaction timelines. Ageing WIP into 30-day buckets by department separates them quickly. Anything over 90 days is typically a decision nobody has made rather than work still in progress.
Can I get this from LEAP, Actionstep, Osprey or Proclaim without changing systems?
Partly, and that's the point. Their built-in reports and Power BI connectors cover activity, billing and cash inside the system. What they can't reach is payroll cost rates, the accounts ledger and pre-migration history, which is what matter-level profitability needs. We read your existing systems as they are — no migration, no rip-and-replace, and a first working dashboard typically live in around two weeks.
Sources
- The Law Society / Hazlewoods, Financial Benchmarking Survey 2026 (121 firms, £1.2bn fee income). communities.lawsociety.org.uk
- The Law Society, Financial Benchmarking Survey 2025 (145 firms; 773 chargeable hours against the 1,100-hour target). communities.lawsociety.org.uk
- Law Society Gazette, in-depth coverage of the 2026 survey (client account interest, IT spend above 4% of fee income). lawgazette.co.uk
- Crowe UK, Law Firm Benchmarking 2025 (lock-up 145 to 138 days). crowe.com
- Armstrong Watson, Law Firm Benchmarking Review 2024/25, February 2026 (fee income per fee earner £165k; net margin 27.7%; staff costs 38.9%; PII 4.8%; leverage 10.6). armstrongwatson.co.uk
- Clio, Legal Trends benchmarks, updated October 2025 (North American data: realisation 88%, collection 93%, total lock-up 93 days). clio.com
- SRA, regulated population statistics (firm numbers and legal structures). sra.org.uk
- SRA, Transparency Rules (price publication requirements, current version effective April 2025). sra.org.uk
- SRA, innovation and technology guidance (accountability for work carried out through others). sra.org.uk
- ICO, contracts between controllers and processors. ico.org.uk
Related reading
What happens to your recorded hour?
In a free 60-minute discovery call, we'll map how your time recording, billing ledger, accounts and payroll fit together, and tell you what realisation and lock-up by department would take to produce.
Book a free discovery call → Read-only reporting over your existing systems · UK-based team · First dashboard in about two weeks