Scope & Billing Guardian
Reviews each draft invoice against the engagement letter scope and agreed rate card before it goes out — flagging out-of-scope work, rate variances and non-billable time, and drafting the adjustment note for the engagement owner to clear.
The live demo, running on fabricated data. Open it to step through the full flow — every output is shown for a person to approve before anything happens.
Reads each line on a draft bill, checks it against the costs agreement scope and the agreed rate card, drafts the exceptions with the clause they breach, and holds the bill for the responsible principal to approve before it reaches the client.
It does the line-by-line reconciliation against the costs agreement every time, in full, before the bill leaves the practice.
- On a fixed-fee matter the value is buried in the exceptions: a Phase 4 workshop billed under a fee that only covers Phases 1–3, an interview billed at $480/hr against an agreed $420, non-billable travel that slipped onto the bill.
- Best for a mid-sized practice or in-house team running enough concurrent matters that no one can hold every scope and rate in their head — on a mix of fixed-fee, estimate-capped and time-billed work.
- The fee-earner or costs lawyer who reconciles a stack of pre-bills by eye gets that reconciliation as a draft, and redirects the recaptured hours to the judgement calls — whether to raise a variation, whether to write off, whether to honour a rate.
It is confidently wrong when the scope or rate it checks against is stale — and an adjusted bill looks more authoritative than the costs agreement behind it.
- Weak where scope lives in a partner's recollection rather than a written costs agreement — with nothing to compare a line to, it will either flag everything or flag nothing.
- Weak on genuinely judgement-heavy billing — value-based pricing set per matter, blended rates negotiated verbally, "we'll sort it out at the end" — where almost every line is an exception to a rule no one wrote down, so the exception list becomes noise.
- An out-of-date rate card or an unrecorded variation produces stale flags with the same certainty as real ones, so the held list is never a substitute for the principal reading the bill.
Pull your last ten draft bills and try to point, for each questionable line, at the clause in the costs agreement that governs it. If you can't find the clause quickly, neither can the tool — and that's a billing-discipline problem before it's an AI one.
It flags the exception and drafts the note. The responsible principal decides and bills. That boundary is not optional.
What is charged and disclosed to a client is regulated work in Australia. Under the Legal Profession Uniform Law a law practice must give costs disclosure where costs are likely to exceed $750 ex GST and disbursements, legal costs must be fair and reasonable and proportionate, and trust money can only be withdrawn for costs after a bill is given — and overcharging is a matter for the Legal Services Commissioner. A wrong bill lands on the responsible principal, not the tool, so the person accountable for the costs stays on the decision.
A written, current costs agreement scope and rate card, and time entries structured enough to match against them.
The scope and the variation trail are almost always the weak point — priced in email threads, costs disclosures that don't tie to time entries, and a partner's memory of what was agreed. Getting the scope, rate card and billable rules into a consistent, machine-readable shape — usually by changing how the costs agreement and time entry are captured, not by buying a tool — is the real first job, and it is bigger and more valuable than the AI layer that sits on top.
The worried-buyer questions, answered straight
Fixed scope, fixed price, fixed dates.
Considering this for your practice?
The honest place to start is the costs agreement — getting your scope, rate card and non-billable rules written down and owned. Tell us where the bill review hurts; we'll play it back, scope it, and show you what's possible.