Key takeaways
- Benchmarks are only useful if the firms in the sample look like yours; 78% of UK legal businesses employ 0 to 4 people[1], and most surveys over-represent larger firms.
- Check definitions before comparing: what counts as a fee earner, how profit is measured and whether owners' notional salaries are deducted.
- Region matters: London holds about 41% of UK legal activities enterprises[2], with very different costs and rates.
- Compare against a range, such as the middle half of similar firms, not against a single average.
- Your own trend over three to five years is the most reliable benchmark you have.
Why benchmarking tempts and misleads
Every partner wants to know whether their firm is doing well compared with others. Benchmarking surveys from professional bodies, accountancy firms and banks promise an answer: average fees per fee earner, profit per partner, overhead ratios, lock-up days. Used carefully, they can show where a firm is out of line and prompt useful questions. Used carelessly, they produce false comfort or needless alarm, and sometimes bad decisions.
The problem is rarely the survey. It is the comparison. A figure from a group of firms that differ from yours in size, region, work mix or accounting method tells you very little about your own performance.
There is also a timing problem. Most surveys are published months after the period they describe, and the legal market can move quickly when property transactions, interest rates or legal aid fees change. A benchmark that was accurate for last year may already be out of date for the decision you are making now.
Know what the legal market looks like
Public statistics give a reality check before you open any survey. The ONS counted 34,255 VAT and/or PAYE-registered legal activities enterprises in the UK in March 2026[3]. 78% of them employed between 0 and 4 people[1], and 58% had annual turnover below £250,000[4]. Total turnover for the sector was £50.7 billion in 2024[5], concentrated in a relatively small number of large firms.
Location skews the picture further. London had 13,885 legal activities enterprises in 2026, about 41% of the UK total[2], while the North West (3,880) and South East (3,145) had the most outside London[6]. Costs, salaries, rents and charge-out rates vary widely between these markets.
If a survey's respondents are mainly firms with twenty or more fee earners, or mainly London firms, its averages describe a minority of the market. That does not make the survey useless, but it does mean a small regional practice should not expect to match its figures.
Questions to ask of any survey
- Who took part? How many firms, of what size, in which regions, and how were they recruited?
- Is the sample self-selected? Firms that choose to take part are often better organised than average.
- Which year does it cover, and are the figures final or provisional?
- Are results shown as medians and ranges, or only as averages that a few large firms can distort?
- How are fee earners counted: by headcount or full-time equivalent, and are trainees and paralegals included?
- How is profit defined: before or after a notional salary for owners, and before or after interest and tax?
- How is work in progress valued, and is lock-up measured in the same way you measure it?
- Can results be filtered by firm size, region and work type?
Even official data needs care. The ONS released the 2026 provisional earnings tables on 22 October 2026[7]; provisional figures are revised later, so a comparison made on provisional pay data may shift when final figures appear.
Making your numbers comparable
Restate your own figures
Before comparing, recalculate your measures using the survey's definitions. If the survey deducts notional partner salaries, do the same.
Choose the right peer group
Filter for firms of similar size, region and work mix. If the filtered group is very small, treat its results as indicative only.
Compare with a range
Look at where you sit within the middle half of the peer group, not just against the median. Being in the range is normal; being well outside it deserves a question.
Look for patterns, not single numbers
A high overhead ratio combined with low fees per fee earner tells a different story from a high overhead ratio with high fees.
Ask why before acting
Every gap should prompt an explanation before a decision. Some gaps reflect deliberate choices, such as higher support levels to protect service.
Measures that benchmark well and badly
| Measure | Benchmarks well when | Benchmarks badly when |
|---|---|---|
| Lock-up days | Peers have similar work mix and billing practices | Comparing conveyancing-heavy firms with litigation firms |
| Overhead ratio | Peers are similar in size and region | Comparing a London office with a market-town firm |
| Fees per fee earner | Fee earners are counted the same way | Some surveys count headcount and others full-time equivalents |
| Profit per equity partner | Notional salaries and partner numbers are treated alike | Different leverage and partner definitions |
| Staff cost ratio | Peers have similar support models | Some firms outsource cashiering, IT or reception |
Build a simple internal benchmark pack
Before looking outwards, create a one-page pack that compares your own teams and offices with each other on the same definitions. Internal comparisons remove most of the problems that make external surveys hard to use: the accounting is the same, the region is the same and the data is current.
- Fees billed and cash collected per full-time equivalent fee earner, by team.
- Recovery: cash collected as a share of time recorded, by team.
- Lock-up in days, split into unbilled work and unpaid bills.
- Direct cost per matter for high-volume teams.
- Enquiries, consultations and instructions per month, by practice area.
- Complaints and write-offs linked to service, per hundred matters closed.
Show each measure for the last four quarters and the same quarter a year earlier. Teams quickly start asking why another team's lock-up is shorter or recovery higher, and those conversations are often more productive than any comparison with firms the partners have never met.
Example: a regional firm reads a national survey
Example (fictional). A six-partner firm in the North West reads a national benchmarking survey and finds its fees per fee earner well below the reported average. One partner proposes raising billing targets across the firm.
- Example: the finance partner checks the survey's sample and finds it is weighted towards larger firms, many in London.
- Example: the survey counts fee earners by full-time equivalent and excludes trainees; the firm had counted heads and included trainees.
- Example: after restating its figures and filtering for regional firms of similar size, the firm sits within the middle of the range.
- Example: the real gap is in lock-up, which is longer than peers, so the partners focus on billing frequency rather than targets.
In this Example, the unadjusted comparison would have led to higher targets and probably more time recording pressure, without addressing the actual problem.
Taking part in a survey
If you contribute to a benchmarking survey, the value you get back depends on the quality of what you submit. Use the survey's own definitions, not yours, even if that means extra work. Ask your accountant or practice manager to prepare the return from the same data each year, so that your submissions are consistent and you can compare your own position across several editions.
Check how your data will be used and protected. Reputable surveys anonymise individual firms and publish only aggregated results with minimum group sizes, so no firm can be identified. Read the terms before submitting partner profit figures or salary data, and make sure sharing them is consistent with any confidentiality obligations to staff.
Finally, separate benchmarks for owners from benchmarks for teams. Profit per partner and return on capital are questions for the partners as investors. Fees per fee earner, recovery and lock-up are questions for team leaders as managers. Mixing them in one league table encourages the wrong conversations.
Your own history is the best benchmark
External surveys are snapshots of other firms. Your own management information is a continuous record of your firm, measured the same way each year. Keep five or six measures consistent over three to five years, review them quarterly and use external surveys to ask questions about them, not to replace them. The measures that matter most to most small firms are fee income per fee earner, lock-up, overhead ratio, recovery and the new business pipeline.
Benchmarks feed many decisions: setting fee-earner targets fairly, pricing residential conveyancing, deciding whether to borrow or fund a premium and preparing for a sale, where buyers will compare your figures with their own, as described in how law firms are valued for sale. In each case, a careful comparison with the right peers is worth far more than a headline average.
Frequently asked questions
Are law firm benchmarking surveys reliable?
They can be useful, but only if the sample resembles your firm and the definitions match yours. Check size, region, sample method and how measures are calculated before comparing.
Should we compare with the average or the median?
The median and the range around it are more useful than an average, which a few large firms can distort. Look at whether you sit within the middle half of similar firms.
Which measures are most worth benchmarking?
Lock-up, overhead ratio, fees per fee earner and recovery are the most useful for small firms, provided the definitions match.
What if our firm is far outside the benchmark range?
Find out why before acting. Some differences reflect deliberate choices or different definitions; others point to real problems such as slow billing or under-pricing.
References
Office for National Statistics (2026). UK business: activity, size and location, 2026 (Tables 2, 4, 9 and 17) (opens in a new tab) ↑
Office for National Statistics (2026). UK business: activity, size and location, 2026 (Tables 2, 4, 9 and 17) (opens in a new tab) ↑
Office for National Statistics (2026). UK business: activity, size and location, 2026 (Tables 2, 4, 9 and 17) (opens in a new tab) ↑
Office for National Statistics (2026). UK business: activity, size and location, 2026 (Tables 2, 4, 9 and 17) (opens in a new tab) ↑
Office for National Statistics (2024). Non-financial business economy, UK: Sections A to S (Annual Business Survey), Table 16 Section M (opens in a new tab) ↑
Office for National Statistics (2026). UK business: activity, size and location, 2026 (Tables 2, 4, 9 and 17) (opens in a new tab) ↑
Office for National Statistics (2026). Earnings and hours worked, occupation by four-digit SOC: ASHE Table 14 (2026 provisional) (opens in a new tab) ↑