The billable work your firm is doing but not charging for

Most law firm partners don’t have a utilisation problem. They have a capture problem.

The work gets done. An associate reads a client email at 7:15am, joins a short call with opposing counsel, marks up a draft agreement, then moves into a hearing prep session. By the end of the day, their time entry is incomplete, generic, or missing altogether.

The usual reason is simple. Time recording happens after the work, often at the end of a long day or at the end of a week. At that point, memory has faded and the attorney is trying to reconstruct activity from email, calendar entries, documents, and call logs.

That process creates three predictable outcomes:

• Small tasks disappear because they don’t feel worth entering individually.
• Larger tasks get entered late and described poorly.
• Partners cut vague entries before invoices go out, reducing realisation.

For firms in the USD 1 million to USD 25 million range, this can create annual leakage of roughly $80,000 to $250,000. The exact number depends on practice mix, billing rates, and headcount. The pattern is consistent. A few missed minutes across a day become several unbilled hours per attorney each week.

We regularly see firms estimate that 4 to 6 hours per attorney per week go unbilled. Not all of that is recoverable, and not all work should be billed. But much of it is client matter work that was necessary, useful, and simply never captured cleanly enough to make the invoice.

AI time entry capture is not about squeezing every minute from lawyers. It is about recording legitimate work while the evidence is still there.

If you want a broader view of where AI fits across the firm, See Omni for law firms. The time capture workflow is often one of the fastest places to find recoverable revenue because it sits close to work that is already happening.

Why manual time entry breaks down in real legal work

Lawyers don’t work in tidy one-hour blocks. A day can contain 18 separate pieces of client work, many lasting five to 20 minutes.

Consider a commercial litigation associate on a normal Tuesday:

• A client sends an urgent email about a new document request.
• The associate spends 12 minutes reviewing the request and replying.
• They call a paralegal for seven minutes to confirm document custodians.
• They review a 26-page production set for 35 minutes.
• They message a partner about a procedural point.
• They make edits to a discovery letter between other meetings.
• They join a five-minute call with the client to clarify dates.

Each activity can be billable, subject to the engagement terms and the firm’s billing judgment. Yet the associate may only remember the production review and the letter when entering time later.

The problem is worse in practices where work arrives constantly. Family law, employment, immigration, personal injury, conveyancing, corporate transactions, and litigation all involve a steady flow of short communications and matter administration.

The attorney knows time entry matters. The managing partner has reminded the team. The practice management system is available on mobile. None of that changes the core friction. Asking a lawyer to reconstruct their work after the fact is asking them to do another administrative task when they are already tired.

Then there is the narrative issue.

An entry like “Emails re matter” may reflect real work, but it gives a billing partner little confidence and a client little context. It is more likely to be written down than an entry that clearly states the purpose, matter, and result.

A useful entry might read:

Review and analyse claimant’s supplemental document request, confer with client regarding relevant custodians, and advise on proposed scope of response.

The work may have taken 0.6 hours. The difference is not cosmetic. A specific narrative helps the partner assess the charge, helps finance prepare the invoice, and gives the client a defensible explanation of what the firm did.

What an AI time capture agent actually does

A time capture agent does not replace your billing rules or decide what the firm should charge. It creates a timely, reviewable draft based on work signals already generated in the course of practice.

Those signals may include calendar events, matter-labelled emails, document activity, call records, meeting notes, task management tools, and practice management platforms. The agent connects activity to the right person and, where confidence is high, to the likely client and matter.

The workflow looks like this.

First, the agent monitors approved data sources throughout the day. It sees that an associate attended a 42-minute conference call booked against a matter. It identifies related emails sent before and after the meeting. It notices a document comparison completed in the same matter workspace.

Second, it groups related events into a work unit. Instead of proposing three disconnected entries, it may draft one entry for the conference and follow-up analysis. It doesn’t need to bill every system event. It needs to identify the legal work behind those events.

Third, it creates a plain-English draft narrative using the firm’s preferred billing language. It can apply rules such as avoiding block billing, excluding internal firm administration, or flagging entries that need lawyer review.

Fourth, it sends the draft to the attorney while the day is still fresh. The lawyer can approve, edit, reject, or assign a matter from a simple prompt. The final approval remains with the attorney or billing partner.

Fifth, the approved entry goes to the timekeeping or practice management system with an audit trail that shows what source activity informed the draft.

That last point matters. Legal firms need control. A proper implementation doesn’t create an opaque machine that posts charges without oversight. It gives people a clearer queue, a documented source trail, and rules aligned to how the firm already bills.

The operations architecture behind this is similar to the work we deliver through Omni Ops. The aim is to remove repetitive coordination and drafting work while keeping accountable judgment with the people responsible for the client relationship.

The difference between capture and automated billing

These terms are often mixed together, and they shouldn’t be.

Time capture means identifying probable billable work and preparing a draft entry. Automated billing means issuing a charge or invoice. They are separate controls.

For most firms, the right starting point is conservative. The agent drafts time entries. Attorneys approve them. Billing teams maintain their normal pre-bill and invoice review process.

This gives the firm a chance to improve capture rates without changing its financial controls on day one.

A staged rollout often looks like this:

1. Start with a small group of attorneys who already use consistent matter codes.
2. Connect calendar and email metadata before bringing in document activity.
3. Set a minimum confidence level for matter matching.
4. Keep entries in draft status until the attorney approves them.
5. Review rejected entries to improve rules and prompts.
6. Measure hours proposed, hours approved, time edited, and time rejected.
7. Extend the workflow to another practice group once the first group trusts it.

The approval data becomes valuable in its own right. It shows where the firm loses time, which work types create vague entries, and which matters generate disproportionate unrecorded activity.

It can also expose process issues. If a team spends hours every week classifying incoming documents or chasing basic intake details, time capture will reveal that load. The better move may not be to bill more of it. It may be to automate the work upstream.

Time capture gets stronger when intake is handled properly

The quality of a time entry depends on the quality of the matter data behind it.

If a call comes in after hours, is logged as an unknown contact, and sits untouched until the next morning, the firm has a lost intake risk and a weak record of early activity. If a form submission is manually copied into three systems, staff spend time on administration that can be missed, misclassified, or never tied to a new matter.

This is where the Intake Voice Agent and Matter Triage Agent fit alongside time capture.

The Intake Voice Agent answers calls during lunch, evenings, weekends, and overflow periods. It can gather initial facts, conduct an approved conflict-check process, capture contact details, and book a consultation into the right calendar. It doesn’t provide legal advice. It makes sure a prospective client reaches the firm and the relevant information is recorded.

That matters because 30% to 40% of after-hours intake may never convert when firms are slow to respond. A faster response protects revenue. A structured intake record also gives the firm cleaner data from the first client interaction.

The Matter Triage Agent reviews inbound forms and emails, classifies the practice area, scores fit against firm criteria, and routes the matter to the right partner or team. It can attach a short brief that explains who the prospect is, what has happened, deadlines mentioned, and the recommended next step.

Those two workflows reduce manual handoffs. They also create cleaner matter context for later work. Once an engagement is opened, attorneys and staff are less likely to spend time hunting for basic facts across inboxes and notes.

For a practical way to assess your front-end process, download the AI Client Intake Checklist for Law Firms. It is useful as a working sheet for mapping who answers calls, where information is entered, how conflicts are handled, and where prospective matters stall. You can also access the direct checklist here.

Document review is another source of hidden time

Time leakage is not limited to emails and calls. It appears in document work too.

Junior associates can spend days completing first-pass review of discovery batches, contracts, transaction documents, and historical matter files. At $200 to $400 per hour of associate time, slow review creates a real cost base. Some of that work is billable. Some is written down because the effort exceeds what the matter can bear. Some is performed under fixed fees, where every extra hour reduces margin.

The Document Review Agent takes a first pass through defined document sets. It can identify clauses, extract key dates and parties, compare provisions against a playbook, summarise positions, and produce an associate-grade memo for review.

For litigation, it can help organise discovery material around agreed issues and flag potentially relevant communications. For corporate teams, it can identify change-of-control clauses, termination rights, liability caps, governing law, and unusual drafting. For employment teams, it can surface policy provisions or correspondence patterns that need legal attention.

The lawyer remains responsible for the legal analysis. The point is to reduce the hours spent finding the first layer of facts.

This has a direct connection to time capture. When the Document Review Agent completes a defined review task, that activity is already structured. The time capture workflow can use the matter ID, review batch, assigned lawyer, and output as evidence for a proposed entry. The attorney can then confirm the time and refine the narrative instead of recreating the work record from memory.

You can see how these components connect through Omni Apps, where the operating workflow matters as much as the AI model behind it.

What to measure before you buy anything

Don’t start with a generic claim about productivity. Start with your firm’s own data.

Pull four to eight weeks of records and look at:

• Time entered on the same day as work performed.
• Time entered more than three days later.
• Pre-bill write-downs attributed to vague narratives or excess time.
• Calendar hours that have no associated client matter entry.
• Volume of short client emails and calls per fee earner.
• Unbilled matter administration by practice group.
• Fixed-fee matters where associate hours regularly exceed the budget.
• Intake response time for calls and web submissions outside business hours.

You are not trying to prove that every gap is recoverable. You are trying to find the repeatable gaps.

A litigation team may find that late entries are the main issue. A property practice may find that staff manually move information between systems. A corporate team may find that document review is consuming too much associate capacity before the legal work even begins.

The answer may be one agent or a connected set of agents. The right scope comes from the workflow and economics, not from forcing every issue into one software product.

If you want help mapping that opportunity, Book a 60-min Omni Audit. It is a working session, not a sales deck. We identify the manual workflow, estimate the financial exposure, and map the first practical automation path.

A sensible business case for a managing partner

A credible business case uses conservative assumptions.

Say a 10-attorney firm has an average hourly rate of $325. If each attorney is missing four hours per week of legitimate client work, that is 40 hours each week. Over 46 working weeks, the gross value is substantial.

You should not assume every hour becomes revenue. Some time will be non-billable. Some entries will be rejected. Some clients have caps or fixed-fee arrangements. A realistic model applies a recovery rate based on your practice mix and write-off history.

Even a modest recovery of one billable hour per attorney per week can add meaningful revenue without hiring another fee earner. It can also improve time entry discipline, invoice quality, and visibility into matter effort.

The bigger benefit is management clarity. When data arrives late and narratives are weak, partners cannot see the actual cost of delivering work until well after the matter has moved on. Better capture gives you earlier signals about scope creep, staffing pressure, and matters that need intervention.

That is the operating value of AI. It should improve decisions and remove administrative drag. If it only creates another dashboard, it hasn’t solved the problem.

For more examples of operational AI patterns beyond legal work, the Enterprise DNA insights library and practical guides are useful starting points.

Start with the work your lawyers already do

AI time entry capture works best when it respects how law firms operate.

It should not ask attorneys to trust unreviewed charges. It should not force a major practice management replacement. It should not turn privileged client data into a casual experiment.

It should capture approved activity signals, draft useful narratives, match work to matters with clear confidence rules, and keep the attorney in control of approval. Then it should connect with intake and document workflows where the same administrative delays are costing the firm time and margin.

The opportunity is not theoretical. If your firm is leaking $80,000 to $250,000 each year through missed time, late entries, write-downs, and manual coordination, the first workflow is worth examining closely.

See Omni for law firms to understand the audit approach and the types of workflows we assess. When you’re ready to map the numbers against your own systems and practice groups, Book my Omni Audit. In 60 minutes, you will leave with three outputs: a workflow map, an estimate of the economic opportunity, and a practical first-step recommendation.