Negotiation Strategist

Negotiation Strategist

> Preparation collapses the role of luck. This skill is the preparation.

What this skill is

A pre-negotiation workflow that codifies your interests, your Best Alternative to a Negotiated Agreement (BATNA), your reservation price, the likely Zone of Possible Agreement (ZOPA), the issues available for multi-issue trades, an anchoring plan, a concession schedule, and a simulated counterparty profile. Outputs a play-by-play with the opening move, anticipated responses, and the trigger conditions to walk away.

What it solves

• Walking into negotiations with a target price but no BATNA
• Single-issue bargaining when multi-issue trades create value
• Anchoring badly (or being anchored) because the open wasn't prepared
• Conceding without getting something in return
• Confusing positions with interests - and missing the integrative deal

When to invoke

• Commercial contract negotiations (vendor, customer, partner)
• Mergers and Acquisitions (M&A) or financing term-sheet talks
• Internal negotiations (budget, headcount, scope, deadlines)
• Hiring offer or counter-offer
• Salary or compensation review
• Multi-party coalition deals

Phase 1: Identify interests, not just positions

For each side, list:

• Stated positions - what they say they want (e.g., "$50 per seat per month")
• Underlying interests - why they want it (e.g., "hit Q4 budget; show win to Chief Financial Officer (CFO); reduce vendor sprawl")
• Constraints - what they cannot do (regulatory, internal policy, signed precedent)

Interests outnumber positions and overlap more than positions do. The negotiation moves when you find an interest-level alignment.

Phase 2: Compute BATNA and reservation price

Best Alternative to a Negotiated Agreement (BATNA) is your fallback if this deal falls through:

• List actual alternatives, not hypothetical ones
• Quantify each (revenue, cost, time)
• Pick the best - that's your BATNA

Reservation price is the worst deal you'd still accept (one notch better than BATNA). Below this, you walk.

Never enter a negotiation without a written BATNA and reservation price. Improve your BATNA before negotiating if you can - it's the single highest-leverage preparation move.

Phase 3: Estimate the counterparty's BATNA and reservation

Use whatever you have:

• Their public alternatives (Request for Proposal (RFP) responses, vendor list, public M&A history)
• Internal signals (urgency, end-of-quarter pressure, regulatory timelines)
• Past behavior (have they walked before? at what level?)

If your reservation price ≤ their reservation price, there is a Zone of Possible Agreement (ZOPA). Negotiation happens inside it. If no ZOPA, the deal isn't viable - change the issues.

Phase 4: Expand to multi-issue trades

Single-issue negotiation is always distributive (zero-sum). Multi-issue negotiation can be integrative (value-creating) by trading across issues each side values differently.

For each side, list all issues and rank by importance (1 = most, n = least):

| Issue | Your priority | Counterparty's priority | Trade potential | |-------|--------------:|------------------------:|-----------------| | Price | 1 | 1 | Distributive | | Payment terms | 5 | 2 | High - they care more | | Contract length | 2 | 5 | High - you care more | | Exclusivity | 3 | 4 | Modest | | Service Level Agreement (SLA) | 4 | 3 | Modest |

The high-trade-potential issues are where value is created. Concede on what they value, get what you value.

Phase 5: Anchoring and opening move

The opener defines the range. Anchoring rules:

• Open ambitious but justifiable - well outside the midpoint, with a rationale (precedent, comparables, internal cost)
• The rationale matters more than the number - "industry comparables say $X"
• If they open first, don't counter-anchor immediately - ask for the rationale, then reframe

Decide: do you go first, or wait? Go first if you have better information about ZOPA. Wait if you don't.

Phase 6: Concession plan

Plan concessions before the meeting. Rules:

• Make smaller concessions over time, not larger ones - signals you're approaching your reservation
• Never concede unilaterally - every concession asks for one back
• Use conditional concessions: "If you can do X, I can do Y"
• Bundle, don't unbundle - concede multiple low-priority items together

| Round | Your move | Counter-ask | Walk-away trigger | |-------|-----------|-------------|--------------------| | 1 | Open at $X, 36 months, SLA Y | - | They counter below $W | | 2 | $X-10, 36 months, SLA Y | Add Z | Below reservation | | 3 | $X-15, 24 months, SLA Y | Approval timing | etc. |

Phase 7: Tactical playbook

Common tactics - know them, plan for them:

• Good cop / bad cop - name it to neutralize
• Higher authority - "I need to check with my boss" delays; counter by asking for the decision-maker
• Deadline pressure - verify the deadline is real; if artificial, ignore
• Take-it-or-leave-it - call the bluff if BATNA supports it
• Nibble - small late asks; refuse all nibbles or reopen the package
• Silence - let the silence sit; the next speaker often concedes

Phase 8: Simulate the counterparty

Pre-game with a colleague playing the counterparty. Specifically:

• Run the opening move and three rounds
• Pressure-test your concession plan
• Identify the one question that would derail you and prepare an answer
• Identify your tells (impatience, over-justification) and plan around them

Output

• Interests and positions for both sides
• BATNA and reservation price for both sides
• ZOPA verdict (exists, doesn't, or depends on issue expansion)
• Multi-issue trade matrix
• Anchoring plan with justified opener
• Concession schedule with conditional asks
• Tactical playbook with anticipated tactics named
• Walk-away triggers - explicit
• Post-negotiation review template

Operating rules

Always

• Define BATNA and reservation price before the meeting
• Convert positions into interests
• Expand to multi-issue when possible
• Make conditional concessions (never unilateral)
• Simulate the counterparty before the real conversation

Never

• Open with a midpoint number
• Negotiate against yourself with successive concessions
• Reveal your reservation price
• Treat their first offer as their reservation price
• Confuse a tactic ("higher authority") with a real constraint

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: varunk130
• Source: varunk130/claude-code-skills
• License: MIT

Install and usage instructions live in the source repository linked above.