4 of 9 need a decision. The other 5 do not.
Marking every difference between a counterparty's draft and your template produces a document nobody reads. Marking it against a playbook — preferred position, acceptable fallback, the line you do not cross — produces a short list, and a short list gets acted on.
62 pages against 9 rules
Ordered by consequence. Toggle to see the rules the draft already satisfies, which is the part you do not need to read.
Counterparty's standard MSA
62 pages · 187 clauses · 9 playbook rules applied
Liability cap
Below your walk-away linecrossed in 12% of past deals100% of fees paid
Up to 200% of fees
Uncapped
Received. Supplier liability uncapped for all heads of loss.
The firm's own bank has crossed this line before, which is why the escalation carries the rate rather than just the flag.
Data residency
Below your walk-away linecrossed in 5% of past dealsProcessing in the Kingdom
Transfer on notice
Unrestricted transfer
Received. Permits transfer to any jurisdiction at the counterparty's discretion.
Termination for convenience
Clause not present90 days, mutual
120 days, mutual
Supplier-only right
nothing in the draft lands anywhere on this ladderReceived. No termination-for-convenience provision anywhere in the draft.
The most expensive category on this page. A redline against your template shows a deletion; a skim shows nothing at all.
Audit rights
Clause not presentAnnual, on notice
On reasonable cause
No audit right
nothing in the draft lands anywhere on this ladderReceived. No audit provision. The compliance annex references one that does not exist.
An internal cross-reference to a clause that was never drafted. This is what a document assembled from three precedents looks like.
2 clauses sit below the firm's own walk-away line and 2 are not in the document at all. Those 2 are the ones a redline cannot show you, because a redline compares text that exists.
The two clauses that are not there
No termination for convenience anywhere in 62 pages, and no audit provision — although the compliance annex cross-references one, which is what a draft assembled from three precedents looks like.
A reviewer reading the counterparty's document sees nothing wrong, because there is nothing there to see. Absence is only visible from the playbook side, as a rule that found nothing to land on.
And two findings that combine
The IP indemnity sits comfortably inside fallback: one-way, capped. It is capped by reference to the liability cap — which is the clause sitting below the line as uncapped.
Individually: one acceptable finding and one escalation. Together: an uncapped indemnity. Nothing that reviews clause by clause will tell you that, and it is the single most valuable thing a human reviewer does.
12%
of past deals crossed the liability line anyway
A rule you override is not a rule
The playbook says never accept uncapped liability. The firm's own precedent bank says it has, in 12% of agreements. Flagging the clause without that number invites exactly the same override, from a partner who has no reason to think this instance is different.
So the escalation carries the rate. 4 of the 9 rules here have been crossed before, and the 12% on liability cap is the one the firm should either enforce or stop pretending to have.
Where that number comes fromFour things that decide whether anyone reads it
None of them are about finding more. All of them are about returning less.
- Findings interact
- The IP indemnity is inside fallback and points at a liability cap that is not there. Each is a reasonable finding on its own; together they are an uncapped indemnity, and no per-clause review will say so.
- Absence is the expensive category
- Two clauses in this draft simply are not present. A redline shows a deletion against your template and a reviewer reading the counterparty's document sees nothing amiss, because nothing is there to read.
- A rule you override is not a rule
- The liability walk-away has been crossed in the firm's own bank. Flagging it without that number invites the same override again; showing the number makes it a decision.
- Order by consequence, not by count
- A sixty-page draft produces dozens of differences. Ranking them by how far they sit below your own line is the only ordering that survives a partner with twenty minutes.
Where the playbook runs out
The third one is the quiet failure: a review is only as complete as the playbook behind it, and it will not tell you what the playbook forgot.
- It reviews against your playbook, not the law
- A clause can sit at your preferred position and still be unenforceable. The playbook encodes commercial appetite; enforceability is a separate question and this does not answer it.
- A playbook has to exist first
- Most firms have one in three partners' heads and a template. The first month is turning that into positions with fallbacks, and it is the part clients underestimate.
- Absence detection needs the playbook to be complete
- It can only tell you a clause is missing if the playbook says one should be there. A gap in the playbook is a gap in the review, silently.
- It does not negotiate
- It tells you where the draft sits and what you have accepted before. What you do with an uncapped liability clause depends on how much you want the deal, which is not in any document.
The rest of the platform
Assistant
Ask the file a question. Every answer cites its source.
Research
Nizam, regulations and ministerial decisions.
Drafting
From your own precedents, not a generic bank.
Arabic and English
Two versions that say the same thing.
Document sets
Thousands of pages, ranked by what matters.
Matters
Every document, deadline and hour against the file.
Time and billing
Captured as you work, not reconstructed on Thursday.
Deadlines
Counted from the trigger, not from memory.