← Where AI Pays
legal · Practice ops · AugmentBuild it

AI first-pass contract review

01 — The scenario

A legal business where this lands in Operations.

The friction
document-reviewcontractsturnaroundrisk-review

02 — What is actually going wrong

A senior lawyer spends ninety minutes reading a supply agreement to find the four clauses that matter. Most of that time is locating standard language and confirming it is standard. The expensive judgement is maybe ten minutes of it.

How it works

The document is parsed into clauses, compared against your firm's playbook of accepted positions, and returned as a marked-up summary: what is standard, what deviates and by how much, what is missing entirely. The lawyer starts at the deviations.

What good looks like

Turnaround on routine agreements drops from days to hours, junior time shifts from locating clauses to arguing about them, and your positions become consistent across the firm because the playbook is now written down.

03 — The problem, in money

$16,453$1,044,225/ yr at stake
Modeled
  • Contracts reviewed per week15 contracts100 contractsPlaceholder — not a validated benchmarkStarting point
  • Hours per contract1.5 hours5.1 hoursPlaceholder — not a validated benchmarkStarting point
  • Share automated0.5 ratio0.5 ratioPlaceholder — not a validated benchmarkStarting point
  • Blended hourly rate75 USD210 USDPlaceholder — not a validated benchmarkStarting point
  • Share of freed time redeployed0.5 ratio0.5 ratioPlaceholder — not a validated benchmarkStarting point

04 — The options

Build$15K–$30KA custom system, owned outright.
Bin$0Do nothing — and accept the leak.

05 — The call

Build it

Build it. A custom system is the right move here — the payback justifies owning it rather than renting a compromise.

06 — How this goes wrong

  • Shipping without the playbook. The value is in the comparison; without a documented firm position the tool just summarises, which nobody needed.
  • Trusting a "no issues found" result. Absence of a flag is not clearance, and treating it as such is how a missing indemnity gets signed.
  • Uploading client documents to a general-purpose consumer tool. Privilege does not survive that conversation with your regulator.
  • Counting the saved hours as profit. In an hourly firm, freed time is only money if it is redeployed onto billable work — which is why the model discounts it.

07 — How you would actually do it

  1. 01Pick one contract type you see constantly. Do not start with the bespoke work.
  2. 02Write the playbook: for each key clause, your preferred position, acceptable fallback, and walk-away.
  3. 03Assemble a test set of thirty past agreements with known outcomes.
  4. 04Build the extraction and comparison, then measure it against the test set before anyone relies on it.
  5. 05Deploy as a first pass only, with the reviewing lawyer signing off as they always did.
  6. 06Feed the disagreements back into the playbook monthly.

08 — Tools worth looking at

SpellbookContract review inside Word; the shortest path for a small firm.
Luminance / KiraHeavier extraction platforms, priced for larger volumes.
Custom on Claude or GPT via APIWhere the playbook is genuinely proprietary and worth encoding yourself.

Named for orientation, not endorsement. No affiliate arrangements, and nothing here has been paid for.

09 — Run it on your numbers

Start from industry averages, then drag the inputs to match your business. The number moves live.Drag each input to your number — the starting points are neutral midpoints, not industry data. The number moves live.

Worth to your business
$16,453per year
Modeled — industry averages
Revenue$0Modeled
Cost savings$0Modeled
Hours / week11Modeled
Industry averages
15 contracts
Industry avg: 15 contractsPlaceholder — not a validated benchmark
1.5 hours
Industry avg: 1.5 hoursPlaceholder — not a validated benchmark
0.5 ratio
Industry avg: 0.5 ratioPlaceholder — not a validated benchmark
75 USD
Industry avg: 75 USDPlaceholder — not a validated benchmark
0.5 ratio
Industry avg: 0.5 ratioPlaceholder — not a validated benchmark

10 — The evidence · what others report

No hard numbers reported.

Reported numbers are third-party claims, source-attributed — not verified by me, and never blended into the model above.

11 — Questions owners ask

Every figure is Modeled — computed from industry-average benchmarks against the scenario’s formula. Move the sliders to your own inputs and the numbers recalculate live. Nothing here is a promise; it is a starting estimate you can pressure-test.
The library is organized by department, outcome, effort and readiness so you can find what fits. Add scenarios to your plan, tune the inputs, and see the combined picture on your plan dashboard.
Want this to be a Projected number? That's what the Audit produces — my forecast on your real data, with my name on it.
Build means a custom system is the right move; Buy means an off-the-shelf tool wins; Bin means the math does not justify doing it at all. Binned scenarios can be favorited for reference but never enter a plan.

Is this your business?

Get your free AI Opportunity Score — the same first pass I run at the start of every Audit.

Get Your Free Score