LegalFree
Legal Billing Narrative Review and Rewrite
River rewrites each entry against the client's own guidelines and the codes on the line, so the narrative carries what the fields do not.
River reads the raw time and the client's billing guidelines together, and rewrites every entry to the standard that client actually applies. Each line comes back with what changed and why, so a timekeeper reviewing forty entries sees the rule behind each one rather than a tidied paragraph. Nothing is invented: where an entry does not say what was done, it is returned as a question to the person who recorded it rather than filled in from context.
The rewrite happens against the codes on the line, not just against the prose. A line item carries a task code, an activity code, a timekeeper classification and a unit count, and the client's rules run on those fields before anybody reads a word. So an entry whose narrative describes discovery under a pleadings code is flagged by the system regardless of how well it is written, and improving the sentence changes nothing at all.
Every entry at risk lands on a sheet with the guideline it breaches, the amount exposed and whether the fix is a rewrite, a recode, a split or a write-off before the invoice goes out. The totals reconcile against the phase budget the matter was quoted from, so a phase running hot is visible in the same week rather than at the next revision. The terms it is billed under come from the engagement letter the matter opened on.
The published standard has numbers in it
There are eleven activity codes in the standard set and every one of them is a verb. They run from plan and prepare for, research and draft or revise through four flavours of communicate to manage data and files. That field is already on the line. So the near-universal advice to open a narrative with a past-tense verb spends the first word restating a code the reviewer can see, and gets the entry flagged outright wherever the verb and the code disagree.
One task per entry is the rule everybody quotes and it is not quite the published one. The federal fee guidelines ask whether entries are recorded in increments of point one of an hour and whether discrete tasks are recorded separately, and state that the trustee will object to block billing or lumping. Then they carve out one daily entry combining tasks on a single project where the day's total for them stays at or under half an hour.
The same document names the phrases. It says entries have to carry enough information to identify the purpose of the work, and that attention to or review file, without greater specificity, are generally insufficient. It also names two categories of work that get objected to whatever the narrative says: clerical tasks and word processing as overhead, and non-working travel billed at the full rate. Those are recode or write-off decisions, and no rewrite reaches them.
How it works
Hand over the time
The raw entries as exported, plus the client's billing guidelines in whatever form they exist.
Rules get extracted
The guidelines become a checklist of testable requirements rather than fourteen pages of prose.
Lines get checked
Each entry runs against the codes on it and every rule that governs the line.
Fixes come back
A rewrite, a recode, a split or a write-off, each with the amount at stake.
What you get
- Every entry rewritten against the guideline that governs it, with the rule on the row
- Codes checked against the narrative, because a mismatch flags before a human reads it
- Block-billed lines split by task, or kept where the published exception actually allows it
- Overhead, clerical and non-working travel separated out, because no rewrite saves those
- Amount at risk on every flagged line, so the write-down is known before invoicing
- Entries too thin to rewrite returned as a question, never filled in from context
Common questions
Does it rewrite entries into something we did not do?
Never. A rewrite only reorganises and specifies what the entry and the file already support. Where the recorded words do not establish what happened, the line comes back as a question for the timekeeper with the missing element named, rather than a confident sentence somebody has to defend on a fee application later.
What if the client has no written guidelines?
Then it runs against the published standard and against what that client has actually cut before, which is usually more informative. A pattern of reductions is a guideline the client never wrote down, and reconstructing it from three months of adjustments gives you rules you can hand a team.
Why does it care about the codes at all?
Because most reductions are decided by the codes before a person reads a word. A line carries a task code, an activity code, a timekeeper classification and a unit count, and the client's rules test those fields first. An entry with perfect prose and the wrong activity code is flagged and a beautifully written one never gets read.
Can it handle a fee application rather than an invoice?
Yes, and the standard is stricter. Fee applications get read line by line by somebody whose job is objecting, so the purpose and benefit of each entry has to be on the face of it. The same checks run, with the project category structure and the increment rules enforced rather than treated as preferences.
How much time does a prebill review take?
The pass is fast and what it saves is the argument. The value is not the minutes on the review, it is a write-off decided now rather than an adjustment arriving in month three with no explanation attached. Recurring causes surface as patterns, so the same lines stop appearing.
Where do the discovery hours belong?
Split by purpose, which is where most miscoding happens. Reading to learn the facts, reading for privilege before production, and building the review database are three separate codes, so proving each privilege claim to its elements and validating the load file that arrived bill to different lines entirely.
Legal Billing Narrative Review and Rewrite
Fill in the form and your workspace opens with the work already underway.