The fastest way to reduce time recording admin in a UK law firm is to capture time where the work happens instead of reconstructing it afterwards - and Writford is built around exactly that: research, document analysis and drafting are logged against the matter as they happen, so time entries draft themselves and fee earners approve rather than remember.
This is not a marginal efficiency point. It is one of the largest recoverable revenue leaks in most firms, and unlike winning new work, recovering it requires no new clients, no new hires and no rate rise. This post sets out the scale of the leak, where the time actually goes, what contemporaneous capture changes, and the maths for a typical small firm.
The scale of the leak
An April 2026 Censuswide survey of 200 UK legal professionals, commissioned by Access Legal, found that fee earners lose an average of 4.16 billable hours per week to administrative workarounds, manual processes and time-capture tools that do not work well enough. At the £194 hourly rate used in the study, across a 50-fee-earner firm working 48 weeks, that is just under £2 million a year in unrecorded billing.
Two other findings from the same survey deserve attention. First, the breakdown of where the time goes: time recording and billing itself (28%), switching between disconnected applications (26%), document drafting (22%), and client onboarding and AML checks (20%). Second, the perception gap: 71% of fee earners said their firm's software hinders accurate time capture - while 71% of firm leaders were confident their systems capture everything. Someone is wrong, and it is usually not the people filling in the timesheets.
| Leak source | What it costs you | The fix |
|---|---|---|
| Retrospective entry (end of day / week) | Short units vanish first: calls, quick emails, five-minute reviews | Contemporaneous capture at the point of work |
| App-switching | Work done in research tools, Word and email never reaches the time ledger | Research, drafting and documents in the same workspace as time recording |
| Vague descriptions | "Reviewing documents" gets challenged and written off | Entries drafted from the activity log, so descriptions are specific |
| No WIP review before billing | Unbilled time is silently written off, matter by matter | A WIP report run before every invoice cycle |
| Junior under-recording | Trainees shave time they fear looks slow | Suggested entries from actual activity, reviewed by the fee earner |
Why time leaks
The pattern is consistent across firms, and it is a design problem before it is a discipline problem.
Reconstruction, not recording. Most time recording is retrospective. The fee earner does the work, then at 5pm - or Friday, or month-end - tries to rebuild the day from sent emails and memory. Long blocks survive that process. The six-minute units do not: the call that interrupted drafting, the quick review of an incoming letter, the email answering a client's follow-up question. Individually trivial; across a year, they are the leak.
App-switching. Research happens in one tool, drafting in Word, correspondence in Outlook, and time recording in the practice system. Every switch is a boundary the time entry has to be carried across by hand, and plenty never make the crossing. It is telling that disconnected applications were the second-biggest frustration in the survey, barely behind time recording itself.
The confidence trap. That 71%/71% perception gap explains why the problem persists: the people who could fix it believe it is already fixed. The WIP report says what was recorded; it cannot say what was never recorded, so the leak is invisible in the very reports management relies on.
What contemporaneous, matter-linked capture changes
The principle is simple: if the work happens inside the same workspace as the time ledger, the record of the work already exists - nobody has to remember it.
On Writford, a research session runs against the matter: the queries, the sources retrieved from live UK sources, the time elapsed are all logged. A document analysis runs against the matter's own files. A draft is produced on the matter. Each of these generates a suggested time entry with a specific description drawn from the activity itself - "Reviewed warranty carve-outs in draft SPA clause 12; noted exposure for client email" rather than "Document review, 0.5". The fee earner reviews, edits and approves. See Time Recording for the workflow.
Three things follow. The short units stop vanishing, because they are captured at the moment they happen. The descriptions get specific, which means fewer client challenges and fewer write-offs at billing. And the WIP report finally reflects reality, so the partner reviewing it before the invoice run is clearing genuine unbilled time rather than admiring an undercount. We covered the description and write-off side of this in more depth in AI and legal billing.
The maths for a ten-fee-earner firm
Worked example, clearly labelled: these figures are illustrative, using the survey's averages (4.16 hours per fee earner per week; £194 per hour; 48 working weeks). Your rates and leak will differ - substitute your own numbers.
- Leak per fee earner per week: 4.16 hrs × £194 = £807
- Leak per fee earner per year: £807 × 48 = about £38,700
- Across 10 fee earners: roughly £387,000 a year in unrecorded work
Nobody recovers all of it - some of that lost time is genuinely unavoidable admin. So be conservative. If better capture recovers just one hour per fee earner per week, that is 10 × 1 × £194 × 48 = £93,120 a year. A quarter of the survey-average leak lands in a similar place, around £96,800.
Against that, the software cost: ten Writford Standard seats at £59 per seat per month (billed annually) is £7,080 a year - and that buys the whole workspace (matters, research, drafting, document analysis, billing, the Outlook add-in), not just a timer. Even if the recovery is half the conservative estimate above, the arithmetic is not close.
Where Writford fits
Writford puts research, document analysis, drafting, matters, time recording and billing in one workspace, so the activity log is the time record. Suggested entries with specific, defensible descriptions; a single WIP screen across all matters before every invoice run; UK 20% VAT, disbursements and aged debtors handled natively. It is on every plan from £59 per seat per month - AI capture is not a premium tier. If your fee earners are doing their real work in three other tools and reconstructing timesheets on Friday afternoon, this is the specific problem Writford was built to remove.
What to check before buying time recording software
If you are evaluating tools for this problem, these are the questions that separate a fix from a new timer:
- Is capture contemporaneous or retrospective? A prettier end-of-day entry screen does not fix reconstruction. Ask where the time entry originates.
- Does the work happen in the same place as the ledger? If research and drafting still happen elsewhere, the app-switching leak survives intact.
- Do suggested entries carry real descriptions? "Suggested: 1.2 hrs" saves nothing at billing time. The description is where write-offs are won or lost.
- Is there a usable WIP view? Unbilled time you cannot see is unbilled time you will write off. Check the report exists and that someone will actually run it.
- Does the fee earner stay in control? Auto-capture that posts entries without review creates a compliance problem instead of solving a revenue one. Suggest, review, approve.
- What does it cost per seat, all-in? Add up the modules. A cheap timer plus a research subscription plus a document tool usually costs more than one workspace.
The honest caveat
Software will not fix a firm that will not record. If partners treat timesheets as optional, no capture mechanism - AI-assisted or otherwise - recovers the leak, because the review-and-approve step is where the entry becomes real. The tooling removes the friction and the memory problem; it cannot supply the intent. The firms that recover six figures from this are the ones where management stops assuming the systems already capture everything (remember the 71%) and makes the WIP review before every invoice cycle a fixed habit. Do that with better capture underneath it and the numbers above are realistic. Do neither and no vendor, including us, will save you.
Try Writford free for 14 days, no credit card required, or see a feature overview.