Writing
The archive, page 7.
587 published, 143 scheduled
- BenchmarkMar 19 2026
Benchmark: chronology construction from unstructured document sets
The short version: chronology construction from unstructured document sets is easier to get wrong than most teams assume, and the failure is usually silent.
4 min read
- Build logMar 18 2026
Build log: table extraction when the table spans four pages
This one comes up in almost every matter review, and the answer is less settled than the confidence around it suggests.
9 min read
- Jurisdiction updateMar 17 2026
the EU: penalty clause doctrine
We keep coming back to the EU: penalty clause doctrine, because the cost of a miss here compounds across a portfolio.
6 min read
- Clause teardownMar 16 2026
Clause teardown: indemnity
Every review team we work with has a version of this problem. Indemnity rarely announces itself.
4 min read
- Long readMar 15 2026
precedent banks and the half-life of a good clause
The short version: precedent banks and the half-life of a good clause is easier to get wrong than most teams assume, and the failure is usually silent.
3 min read
- Field guideMar 14 2026
Field guide: spotting an evergreen clause before it renews
This one comes up in almost every matter review, and the answer is less settled than the confidence around it suggests.
8 min read
- Practice noteMar 13 2026
Practice note: running a playbook that survives contact with sales
We keep coming back to running a playbook that survives contact with sales, because the cost of a miss here compounds across a portfolio.
5 min read
- BenchmarkMar 12 2026
Benchmark: defined-term consistency checking
Every review team we work with has a version of this problem. Defined-term consistency checking rarely announces itself.
3 min read
- Build logMar 11 2026
Build log: why we index definitions separately from operative text
The short version: why we index definitions separately from operative text is easier to get wrong than most teams assume, and the failure is usually silent.
9 min read
- Jurisdiction updateMar 10 2026
California: service out of jurisdiction
This one comes up in almost every matter review, and the answer is less settled than the confidence around it suggests.
7 min read
- Clause teardownMar 9 2026
Clause teardown: notice provisions
Every review team we work with has a version of this problem. Notice provisions rarely announces itself.
8 min read
- Long readMar 8 2026
trust, verification, and the reviewer's burden
We keep coming back to trust, verification, and the reviewer's burden, because the cost of a miss here compounds across a portfolio.
3 min read
- Field guideMar 7 2026
Field guide: checking a limitation calculation with tolling applied
This one comes up in almost every matter review, and the answer is less settled than the confidence around it suggests.
9 min read
- Practice noteMar 6 2026
Practice note: running a playbook that survives contact with sales
The short version: running a playbook that survives contact with sales is easier to get wrong than most teams assume, and the failure is usually silent.
4 min read
- BenchmarkMar 5 2026
Benchmark: obligation extraction on long-form outsourcing contracts
Every review team we work with has a version of this problem. Obligation extraction on long-form outsourcing contracts rarely announces itself.
5 min read
- Build logMar 4 2026
Build log: detecting when a model is confidently wrong about a date
We keep coming back to detecting when a model is confidently wrong about a date, because the cost of a miss here compounds across a portfolio.
7 min read
- Jurisdiction updateMar 3 2026
Hong Kong: electronic execution
This one comes up in almost every matter review, and the answer is less settled than the confidence around it suggests.
3 min read
- Clause teardownMar 2 2026
Clause teardown: governing law and forum
The short version: governing law and forum is easier to get wrong than most teams assume, and the failure is usually silent.
5 min read
- Long readMar 1 2026
what legal AI gets wrong about legal work
Every review team we work with has a version of this problem. What legal AI gets wrong about legal work rarely announces itself.
6 min read
- Field guideFeb 28 2026
Field guide: testing a non-compete for enforceability
Every review team we work with has a version of this problem. Testing a non-compete for enforceability rarely announces itself.
7 min read
- Practice noteFeb 27 2026
Practice note: sequencing consents so closing does not slip
The short version: sequencing consents so closing does not slip is easier to get wrong than most teams assume, and the failure is usually silent.
3 min read
- BenchmarkFeb 26 2026
Benchmark: redline generation against a firm playbook
This one comes up in almost every matter review, and the answer is less settled than the confidence around it suggests.
8 min read
- Build logFeb 25 2026
Build log: citation grounding that survives document renumbering
We keep coming back to citation grounding that survives document renumbering, because the cost of a miss here compounds across a portfolio.
6 min read
- Jurisdiction updateFeb 24 2026
the Netherlands: privilege in internal investigations
Every review team we work with has a version of this problem. The Netherlands: privilege in internal investigations rarely announces itself.
4 min read