How to Build a Contract Negotiation Playbook

A negotiation playbook is the document that turns "ask legal" into "check the playbook." It takes the positions your company has already settled, ideal, fallback, walk-away, and approved wording, and organises them by clause so that whoever is redlining a contract, lawyer or not, can move without waiting on a reply. (Adira, which publishes this guide, builds contract review and CLM software that runs on a playbook exactly like this. Nothing here requires you to use it.)

This is a companion piece to What Is a Contract Playbook, which explains what a playbook contains. This guide is the build steps: which contract to start with, how to write each clause position, what to bake in as fixed Indian rules rather than judgment calls, how to set approval thresholds, and how to keep the document current.

Step 1: Pick your highest-volume contract type

A playbook pays for itself on repetition. Build it for the contract you sign most often, your standard MSA, your vendor NDA, your customer order form, not for a one-off joint venture agreement that will never repeat. If you genuinely cannot tell which type is highest-volume, pull the last 20 to 30 signed contracts from your repository and sort by type. Whichever type has the most instances is where a playbook saves the most negotiating hours.

Step 2: List the clauses that actually get negotiated

Most contracts have 30 to 60 clauses. Only 8 to 15 get redlined in a typical negotiation. Building positions for all 60 wastes effort on clauses nobody argues about (definitions, notices, counterparts) and delays the playbook's usefulness. Start with the clauses that show up in almost every real negotiation:

• Indemnity: who pays for third-party claims, and whether it is capped
• Liability cap: the ceiling on damages, and what sits outside it
• Termination: for convenience, for cause, and the notice period for each
• IP ownership: who owns what is created during the engagement
• Auto-renewal: whether the contract rolls over silently and how much notice stops it
• Payment terms: days to pay, late interest, and currency
• Confidentiality: scope and survival period after the contract ends

For each, link to a clause explainer covering the mechanics in depth rather than re-explaining it inside the playbook itself: Indemnity Clauses Explained, Limitation of Liability Clauses, Termination for Convenience, Termination for Cause vs Convenience, IP Assignment Clauses, and Auto-Renewal (Evergreen) Clauses. The playbook entry states the position; the explainer carries the reasoning and the statute behind it.

Step 3: Write ideal, fallback, walk-away, and approved wording for each clause

Every clause entry needs four things, not one number.

• Ideal: what you ask for first, the version you would happily sign as-is.
• Fallback: what you will concede without escalating. This is the line a non-lawyer can accept on their own judgment.
• Walk-away: the point past which the deal needs a real decision from someone with authority to reject the whole contract, not just redline a clause.
• Approved wording: the actual clause text for the ideal and fallback positions, ready to paste in. Not a description of the outcome you want, the sentence itself.

A worked example, for a liability cap in a mid-size SaaS contract:

• Ideal: general cap at 12 months' fees, uncapped for confidentiality and IP indemnification breaches
• Fallback: general cap at 12 months' fees, no carve-outs
• Walk-away: anything below 3 months' fees, or a cap that also swallows confidentiality and IP breaches
• Approved wording (fallback): "Each party's total liability arising out of or in connection with this Agreement shall not exceed the total fees paid or payable in the 12 months preceding the event giving rise to the claim."

A sales rep reading a vendor's counter-offer of "cap at 6 months' fees, no carve-outs" can accept it on the spot, it falls inside the fallback band. A counter of "cap at 2 months' fees" hits the walk-away line and gets escalated, no discretion needed either way.

Step 4: Bake in the Indian non-negotiables as fixed rules, not flexible positions

Three positions should never appear as a spectrum with a fallback and a walk-away, because Indian statute does not leave room to negotiate them the way a liability cap or payment term does.

Non-competes. Section 27 of the Indian Contract Act, 1872 states:

"Every agreement by which any one is restrained from exercising a lawful profession, trade or business of any kind, is to that extent void."

Read it on Indian Kanoon. This is a binary rule, not a reasonableness test the way US non-compete law usually runs. A post-termination restraint is void whether it lasts one month or two years, whatever territory it covers. The playbook's position on any non-compete clause should read "not after termination, full stop," never "up to 6 months, India-wide" or any other duration-based fallback. See Are Non-Compete Clauses Enforceable in India? for the enforceable exceptions.

IP assignment period and territory. Section 19 of the Copyright Act, 1957 sets two default rules that apply the moment a clause goes silent on duration or geography. Section 19(5):

"If the period of assignment is not stated, it shall be deemed to be five years from the date of assignment."

Section 19(6):

"If the territorial extent of assignment of the rights is not specified, it shall be presumed to extend within India."

Read both on Indian Kanoon. This is exactly what happened in Pine Labs Private Limited v Gemalto Terminals India Private Limited (Delhi High Court, Division Bench, 2011): an assignment clause said "assigns" but stated no period or territory, and the court held Sections 19(5) and 19(6) were triggered by that silence, reading the missing period as five years and the territory as India-only. Your approved wording for IP assignment should always state "in perpetuity and throughout the world" explicitly, the fix for a statutory trap, not a fallback to negotiate down from. Full detail in IP Assignment Clauses in India.

Stamping. Section 35 of the Indian Stamp Act, 1899 provides that an instrument chargeable with duty cannot be admitted in evidence or acted on by a public officer "unless such instrument is duly stamped." Read it on Indian Kanoon. An unstamped contract is not void, but it can become unusable in court exactly when you need it, mid-dispute. The playbook should list which contract types require stamping, and on what value, as a pre-signature checklist item.

Put these three at the top of the playbook, marked as fixed rules, before the flexible clause positions. A negotiator reading the document should see immediately that these are not judgment calls.

Step 5: Set approval thresholds

A playbook without stated authority levels either escalates everything (slow) or nothing (risky), with no consistent rule for which. Decide, in writing: who can accept a fallback position alone, usually anyone on the deal team running the negotiation day to day; who can go below the walk-away line, usually legal or a named senior approver, with a documented reason rather than a verbal okay; and what happens with the three fixed Indian rules, which get no "who can override" line at all, if a counterparty insists on a post-termination non-compete, that goes to legal as a legal-risk conversation, not a commercial negotiation with room to concede.

A useful test: pick three real clauses from your last signed contract and ask a non-lawyer on the deal team whether they could have decided each one using only the playbook, without asking anyone. If the answer is no for more than one, the fallback bands are too vague or the approved wording is missing.

Red flags in a negotiation playbook

Normal	Red flag	Why it matters
Separate positions for pilot, mid-size, and enterprise deals	One position applied to every deal size	A liability cap that is safe on a small pilot can be dangerously thin on an enterprise contract, and the reverse leaves money on the table on small deals
A stated walk-away line for every key clause	Fallback position only, no walk-away	Without a stop point, "one more concession" has nowhere to end, and nobody notices until the contract is signed
Non-compete position: "not after termination," period	Non-compete position: "up to 12 months, India only"	Section 27 is binary; a moderate-sounding duration is still void, and treating it as a spectrum approves an unenforceable clause
Approved wording pasted directly into the contract	Only a description of the position, no clause text	A non-lawyer negotiating from a description will draft the clause themselves, reintroducing the inconsistency the playbook exists to remove
Clear approval threshold stated per clause type	No stated authority levels	Without thresholds, everything either gets escalated (slow) or nothing does (risky)
Reviewed after a batch of deals or a relevant court ruling	Written once, never revisited	A playbook is a snapshot of positions at the time it was written, not a permanently correct standard
IP assignment and stamping listed as fixed checklist rules	IP and stamping folded into the flexible clause list next to liability or payment terms	These are statutory defaults, not negotiating room; treating them as flexible invites a fallback that quietly waives a legal protection
Playbook entry cites a clause explainer for reasoning	Playbook entry states a bare number, no source or link	A negotiator who does not understand why a position exists cannot judge a counter-offer just outside it

A position written badly, and well: termination notice

Bad approved wording, generic and undated:

"Either party may terminate this Agreement for convenience upon written notice to the other party."

This gives no notice period at all, so a counterparty can terminate the moment after signing, and it says nothing about fees already paid or data on the platform.

Better approved wording, with the gaps the fallback should close:

"Either party may terminate this Agreement for convenience upon 60 days' prior written notice to the other party. Fees paid for the notice period are non-refundable; fees for any period after the effective date of termination shall not be charged. Each party shall return or destroy the other party's confidential information within 30 days of termination, and the receiving party shall provide reasonable transition assistance for up to 30 days following termination, at the requesting party's cost."

What changed: a stated notice period (60 days here, ideal at 90, walk-away under 30), an explicit fee treatment so termination timing does not become a billing dispute, and a transition assistance clause so nobody is left stranded mid-engagement. A playbook entry should carry wording this complete, not a placeholder rewritten from scratch every time.

How AI-assisted review uses the playbook

Once the playbook exists, it becomes the input for AI-assisted contract review, not a separate document sitting unused next to it. A review tool told generically to "flag risky clauses" applies patterns trained mostly on US and UK contracts, which is exactly where it misses the Indian traps above: it tends to treat a moderate non-compete as reasonable rather than void, and read a silent IP assignment clause as permanent and worldwide rather than five years and India-only.

Feed the same tool your actual playbook positions instead, and the check changes from "does this look risky in general" to "does this clause match, beat, or fall short of the position we already agreed," a check a non-lawyer can act on directly, and one that flags a real deviation, not a generic pattern. The full five-step workflow, clause-by-clause structuring, feeding it your playbook, an exact-sentence citation for every flag, and a human pass on the clauses that can actually hurt you, is covered in How to Review a Contract With AI.

Test the playbook by hand before wiring it into any tool. Paste a real incoming contract into Weave, Adira's free browser-based contract tool, and walk each key clause against your written positions yourself. If your fallback bands are too vague to apply consistently by hand, no tool downstream of the playbook will apply them consistently either.

Keep it living

A playbook is a snapshot of positions at the time it was written. Three triggers should force a review, not a fixed calendar date: a new deal type or contract size band, since positions written for $10,000 pilots do not automatically hold for $500,000 enterprise deals; a relevant court ruling, the Delhi High Court's confirmation in Varun Tyagi v Daffodil Software Private Limited (FAO 167/2025, judgment dated 25 June 2025) that a post-termination restrictive covenant cannot be enforced is exactly the kind of ruling that should prompt a check of your non-compete wording; and a pattern of escalations, if the same clause keeps getting pushed past the walk-away line, the fallback band is probably set wrong, not every counterparty suddenly being unusually aggressive.

Assign one owner, usually in-house counsel or whoever carries legal risk for the company, whose job includes reviewing the document on these triggers, not letting it sit untouched between renewals.

US and global contrast

Most published negotiation-playbook templates originate from US and UK legal teams, and it shows in the default positions they ship with: reasonableness-tested non-compete bands ("6 to 12 months is generally defensible"), "work made for hire" language assuming automatic, permanent, worldwide IP ownership with no separate period or territory line, and governing law defaulting to a US state or English law. None of these transfer as-is. In India, Section 27 asks only whether a restraint operates after the relationship ends, not whether its duration is reasonable, so a US-style "reasonable months" fallback band approves a clause that is void. And where a US template's silence on IP period and territory is harmless, because work-for-hire doctrine fills the gap automatically, the same silence in an Indian contract triggers the five-year, India-only default under Section 19(5) and 19(6). A playbook copied from a US template with the company name swapped in is missing exactly the three fixed positions that matter most here.

FAQ

How is a negotiation playbook different from a contract playbook? Same document, two angles. "Contract playbook" is the umbrella term for the record of your positions; "negotiation playbook" emphasises the build steps that let someone actually negotiate off it day to day. See What Is a Contract Playbook for the underlying concept.

How many clauses should a first playbook cover? Eight to twelve, on your single highest-volume contract type. Ten clauses done well, with real approved wording and clear thresholds, beats forty clauses with vague guidance and no walk-away line.

Can a playbook really let a non-lawyer negotiate safely? Yes, for anything inside the fallback band, because the judgment call was already made by whoever wrote the playbook. A clear walk-away line means anything worse escalates automatically.

What happens if a counterparty pushes past our walk-away line? That is exactly where the playbook stops deciding for you. It becomes a business call made by whoever holds that authority under your approval thresholds, not a redline to keep negotiating on autopilot.

Do the Indian non-negotiables ever have a fallback position? No. A void non-compete does not become "less void" with a shorter duration, and a silent IP assignment clause does not partially avoid the Section 19(5) default. These three are fixed rules with a single correct answer, checked before the flexible positions are discussed.

How often should we update the playbook? On triggers, not a fixed schedule: a new deal type or size band, a relevant court ruling like Varun Tyagi v Daffodil Software, or a pattern of the same clause repeatedly hitting the walk-away line.

---

This guide explains how to build the document and the process around it. It does not tell you whether a specific fallback number, walk-away line, or piece of approved wording in your playbook is legally sound for your business or enforceable as drafted, particularly for the three Indian non-negotiables, where the cost of getting the wording wrong only shows up in a dispute. Get a lawyer to review your first draft before your team starts negotiating off it. This is not legal advice.