Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Commercial policy: the governance framework defining what terms sales can offer and what triggers approval. Use when authoring a policy charter, defining discount/payment/liability rules, auditing deals, or generating a regional policy.
.claude/skills/borghei-commercial-policy/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | 109% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 193% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 199% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 173% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 152% | 0% |
End-to-end commercial-policy authoring and governance: defining the rules that govern what sales can offer, what triggers approval, and what's prohibited. Pairs with our deal-desk (operational enforcement) and pricing-strategy (price-setting) skills — this is the policy that those execute against.
A good commercial policy:
| Situation | Skill applies | |-----------|---------------| | Authoring commercial policy from scratch | Yes — start with policy charter template + scripts/commercial_policy_generator.py | | Refreshing an existing policy (annual) | Yes — see annual policy review workflow | | Auditing recent deals for policy compliance | Yes — scripts/policy_compliance_checker.py | | Analyzing terms-deviation patterns | Yes — scripts/terms_deviation_analyzer.py | | Tailoring policy for new region / vertical | Yes — scripts/commercial_policy_generator.py --region <X> | | Drafting sales training on policy | Yes — see training section | | Setting prices (not policy on deviations) | Use business-growth/pricing-strategy | | Per-deal approval | Use business-growth/deal-desk | | Writing the specific contract | Use business-growth/contract-and-proposal-writer |
Standard scope:
| Domain | Policy area | |--------|-------------| | Pricing | Standard pricing, discount thresholds, MFN, rebates, custom-bundle pricing | | Contract | Standard term length, payment terms, renewal terms, termination, customer audit rights | | Legal | Acceptable MSA modifications, liability cap, indemnification, jurisdiction, IP | | Operational | SLA tiers, custom SLAs, security commitments, dedicated infrastructure | | Customer commitments | Reference / case study / press release obligations | | Channel | Partner discount tiers, channel-conflict rules, deal-registration | | Special terms | Performance-based payment, acceptance criteria, ramp deals |
What it doesn't cover:
The foundational document. Every company that does $5M+ ARR needs one. Use this template:
markdown# Commercial Policy Charter ## Purpose This Commercial Policy defines the rules that govern commercial terms offered to customers. It is binding on all customer-facing functions (Sales, Customer Success, Partner / Channel) and is enforced by Deal Desk. ## Scope Applies to: - All new customer agreements - All renewals (with material change) - All partner-mediated deals - All custom / non-standard agreements Does not apply to: - Self-serve / PLG transactions per standard published terms - Auto-renewals at standard terms ## Owners and Approvers - Policy owner: CRO + CFO + General Counsel (jointly) - Operational enforcement: Deal Desk - Updates: quarterly review by policy owners - Material changes: board awareness ## Pricing Policy ### Standard pricing - All new customers offered at published list pricing - Published price is canonical; deviations require approval per matrix ### Discount approval matrix [Per the deal-desk approval matrix — see business-growth/deal-desk] ### Maximum allowed discount - Standard maximum: 50% - Beyond 50%: CEO + Board awareness required - Discount > 60%: only with explicit strategic-rationale documented and CEO sign-off ### Most Favored Nation (MFN) - Not granted by default - Granted only with: strategic-tier customer + CRO + CFO + GC approval - Always scoped narrowly: same product, same volume, same term length, same geography - Disclosure-only (never automatic price-match) ### Rebates - Performance-based rebates allowed per partner-program tier - Customer-tier rebates: discouraged; if granted, time-bounded and explicit ## Contract Policy ### Standard term - 12-month contract with annual prepay - Auto-renew unless 90-day notice ### Term flexibility - < 12 months: requires Director approval - 24-36 months: Director approval - > 36 months: VP Sales approval - Multi-year discounts: per discount matrix ### Payment terms - Standard: Net 30, annual prepay - Net 45-60: Director approval - Net 90+: CFO approval - Custom milestone-based: CFO approval; revenue recognition impact reviewed ### Renewal - Standard: auto-renew, same terms, same price (or per published renewal pricing) - Renewal expansion > 20%: deal-desk review - Renewal contraction > 10%: deal-desk review + customer success consultation ### Termination - Standard: termination for convenience requires 90-day notice - Termination for cause: 30-day cure period - Customer-requested termination flexibility: Director approval - Mid-term termination rights: VP Sales approval ## Legal Policy ### MSA modifications - Pre-approved modifications: tracked list in approved-modifications appendix - Custom modifications: General Counsel approval required - Customer-supplied MSA: full GC review; default to push back to our MSA ### Liability cap - Standard: 1x annual fees - 2x annual fees: GC + CFO approval - > 2x annual fees: CEO sign-off - Carve-outs: IP infringement, gross negligence, willful misconduct — always uncapped ### Indemnification - Standard mutual indemnification per template - Customer-favorable indemnification: GC approval - Defense / settlement control: vendor by default; customer-controlled needs CEO ### Jurisdiction and governing law - Standard: vendor's jurisdiction - Customer jurisdiction: GC approval - Arbitration vs litigation: per template; deviations need GC ### IP - Standard: each party retains pre-existing; joint inventions per default - Customer-favorable IP terms: GC approval - Source code escrow: only for OEM / strategic; never standard customer ## Operational Policy ### SLA tiers - Standard published SLA (99.5%) - Enhanced SLA (99.9%): per published pricing - Custom SLA: Customer Success + Engineering approval; pricing premium per agreement - Custom SLA with penalties: CRO + CCO + Engineering approval ### Security commitments - Standard SOC 2 / ISO 27001 commitments per template - Custom security: CISO + GC approval - Customer audit rights: GC approval (limited to annual, with notice, third-party auditor) ### Dedicated infrastructure - Not standard; available only with CTO + GC approval - Premium pricing required ## Customer Commitments ### Reference / case study requests - Standard: requested but not required - Discounted deals (> 15%): case study or reference required as condition - Strategic logos: explicit case study + press release commitment ## Channel Policy ### Partner-mediated deals - Per Partner Agreement; discount per tier - Deal registration governs conflict - Direct rep authority same as direct deals on partner-led opportunities ## Special Terms ### Performance-based payment - Payment-on-acceptance / acceptance criteria: CFO + GC approval - Milestone payments: CFO approval ### Ramp deals - ≤ 3 months: Sales Manager - 3-12 months: Director - > 12 months: VP Sales ### Source code escrow (for customer) - Not standard; available only with CTO + GC approval ## Documentation Requirements Every non-standard deal documented per Deal Desk packet template: - Deviation explicitly listed - Justification documented - Approver identified - Customer commitments (if any) explicit - Expiration / conditions clear ## Annual Review This policy is reviewed annually by CRO + CFO + GC. Material changes communicated to sales with training. ## Effective Date <date> ## Last Updated <date> ## Approved By - CRO: <signature> - CFO: <signature> - GC: <signature> - Board (acknowledgement): <date>
See references/commercial-policy-charter.md for the full annotated charter with notes on each section's typical contentious issues.
See references/discount-and-terms-policy.md for deeper guidance on:
See references/contract-and-commercial-guardrails.md for deeper guidance on:
Before generating the policy, confirm these inputs. If any is unknown or vague, ASK — do not assume:
Stop rule: ask only the 2-3 that most change the output. If the user says "just draft it," proceed and list your assumptions at the top of the policy.
scripts/policy_compliance_checker.py --deals deals.csv --policy policy.yamlscripts/commercial_policy_generator.py --base policy.yaml --region <region> to get base + regional overlay| Script | Input | Output | |--------|-------|--------| | scripts/policy_compliance_checker.py | Deal CSV + policy YAML | Per-deal: compliant / non-compliant with policy violation listing; aggregate compliance metrics | | scripts/terms_deviation_analyzer.py | Deal CSV | Deviation patterns: which terms most often deviate? from which standard? by what magnitude? | | scripts/commercial_policy_generator.py | Base policy YAML + optional region overlay | Generated policy document (markdown), tailored to company stage, ICP, region |
All scripts: stdlib only, argparse CLI, JSON or markdown output.
business-growth/deal-desk — operational enforcement of policybusiness-growth/pricing-strategy — sets prices that policy governs deviations frombusiness-growth/contract-and-proposal-writer — drafts contracts respecting policybusiness-growth/channel-economics — channel deals subject to policy (with overlay for partners)business-growth/partnerships-architect — partnership terms subject to commercial-policy oversightc-level-advisor/cs-cro-advisor — CRO is co-owner of policyc-level-advisor/cs-cfo-advisor — CFO is co-owner of policyra-qm-team/soc2-compliance-expert — policy compliance is audit-relevant evidence| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 33,304 | 30,517 | -8% | 1 | 1 | 0% | 5,048 | 8,289 | +64% | 0 | 0 | — |
case-02 | fail→fail | 28,778 | 29,184 | +1% | 1 | 1 | 0% | 4,320 | 7,917 | +83% | 0 | 0 | — |
case-03 | fail→fail | 23,974 | 32,563 | +36% | 1 | 1 | 0% | 3,481 | 8,484 | +144% | 0 | 0 | — |
case-04 | fail→fail | 15,105 | 15,785 | +5% | 1 | 1 | 0% | 2,121 | 5,778 | +172% | 0 | 0 | — |
case-19 | pass→pass | 16,075 | 11,867 | -26% | 1 | 1 | 0% | 2,335 | 5,105 | +119% | 0 | 0 | — |
case-05 | fail→fail | 14,970 | 16,090 | +7% | 1 | 1 | 0% | 2,255 | 5,855 | +160% | 0 | 0 | — |
case-06 | fail→fail | 13,593 | 14,658 | +8% | 1 | 1 | 0% | 2,270 | 5,753 | +153% | 0 | 0 | — |
case-07 | fail→pass | 18,613 | 14,371 | -23% | 1 | 1 | 0% | 2,654 | 5,534 | +109% | 0 | 0 | — |
case-08 | fail→pass | 11,112 | 8,279 | -25% | 1 | 1 | 0% | 1,629 | 4,765 | +193% | 0 | 0 | — |
case-09 | fail→pass | 10,493 | 7,022 | -33% | 1 | 1 | 0% | 1,547 | 4,625 | +199% | 0 | 0 | — |
case-10 | pass→pass | 13,329 | 10,383 | -22% | 1 | 1 | 0% | 2,083 | 5,071 | +143% | 0 | 0 | — |
case-11 | fail→pass | 12,013 | 10,277 | -14% | 1 | 1 | 0% | 1,857 | 5,069 | +173% | 0 | 0 | — |
case-12 | fail→pass | 12,712 | 9,520 | -25% | 1 | 1 | 0% | 1,943 | 4,900 | +152% | 0 | 0 | — |
case-13 | fail→pass | 17,818 | 19,203 | +8% | 1 | 1 | 0% | 2,376 | 6,464 | +172% | 0 | 0 | — |
case-14 | fail→fail | 17,206 | 22,755 | +32% | 1 | 1 | 0% | 2,837 | 7,654 | +170% | 0 | 0 | — |
case-15 | fail→pass | 10,531 | 4,719 | -55% | 1 | 1 | 0% | 1,753 | 4,252 | +143% | 0 | 0 | — |
case-16 | pass→pass | 10,646 | 8,763 | -18% | 1 | 1 | 0% | 1,549 | 4,813 | +211% | 0 | 0 | — |
case-17 | fail→pass | 10,379 | 3,415 | -67% | 1 | 1 | 0% | 1,603 | 4,010 | +150% | 0 | 0 | — |
case-18 | fail→pass | 14,242 | 9,747 | -32% | 1 | 1 | 0% | 1,968 | 4,856 | +147% | 0 | 0 | — |
case-20 | pass→pass | 8,946 | 8,186 | -8% | 1 | 1 | 0% | 1,231 | 4,707 | +282% | 0 | 0 | — |
case-21 | fail→pass | 17,493 | 13,342 | -24% | 1 | 1 | 0% | 2,428 | 5,360 | +121% | 0 | 0 | — |
case-22 | pass→pass | 20,653 | 19,416 | -6% | 1 | 1 | 0% | 2,830 | 6,231 | +120% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +45 percentage points is the difference between those two pass rates over the 22 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.