Independent CFO redline of any vendor SOW — before you sign.
Big-4 readiness proposal worth $425K on your desk? Anthropic-Blackstone JV term sheet? Palantir FDE engagement letter? Send it to us first. In 5 business days you get a line-by-line redline, an independent ROI recalibration, and paste-ready counter-proposal language — for a flat $15K. Refund if we don't save you 5× the fee.
When you should run the Calibrator first
Big-4 / boutique consulting SOW
$200K–$800K consulting proposal on your desk. Vague acceptance criteria, Phase-2 pricing deferred, team commitments unstated. We'll tell you which deliverables are real and which are padding.
Anthropic-Blackstone JV / OpenAI DeployCo
Model-provider services arm engagement worth $2M–$15M. Often impossible to back out of after month 3. We'll surface the lock-in clauses and ensure acceptance criteria are testable before you sign.
Palantir FDE engagement
Forward Deployed Engineer team for Foundry. Multi-year deal with methodology-IP retention. We'll ensure the ontology you own is portable, the IP terms are sane, and Phase 2 is bounded.
Six-stage independent pipeline
Same Hermes-style multi-sub-agent architecture as our Intelligence Graph, tuned for SOW analysis. Per-stage model routing: structured extraction on flash; judgment-heavy stages on v4-pro (1.6T parameters).
SOW Extraction
Parses the vendor proposal into structured fields — deliverables, milestones, pricing breakdown, team composition, IP ownership, liability cap, indemnification, data residency, portability commitments — quoting load-bearing legal language verbatim.
Ontology Cross-Reference
Runs the Intelligence Graph ontology pipeline on your environment docs to ground every scope-mismatch finding in your actual systems, workflows, and risk register — not generic consultant assumptions.
Deliverable Redline
Classifies every line-item deliverable: load-bearing, filler, mispriced, hidden Phase-2, or scope-mismatch. Each classification carries a rationale citing specific SOW text or IG ontology facts, plus a suggested revision and $-recovery estimate.
ROI Recalibration
Re-runs the consultant's savings claim against your actual workflow volumes and labor rates using deterministic ROI math, then produces a 5x5 sensitivity grid (labor ±20% × volume ±30%). Flags whether the claim survives worst-case sensitivity.
Counter-Proposal Language
Drafts paste-ready negotiation points by severity. Critical sign-blockers, important negotiation items, minor nice-to-haves. Each note cites specific SOW text or ontology facts, and proposes specific testable replacement language.
Pre-Contract Checklist Audit
Applies our 42-question pre-contract checklist against the SOW. Marks every question as answered (✓), dodged (red flag), or must-ask (vendor was silent). Surfaces the specific must-asks before signature.
What the redline reads like
Actual sample output from a calibration of a $425K Big-4 AP automation proposal. Each note is paste-ready — drop it directly into your reply to the vendor.
The current language limits liability to direct damages only and caps at fees paid in the preceding 6 months — for a $425K engagement, this caps the vendor at $212,500, less than the total fee. We require: liability cap at 2× total fees ($850K minimum), carve-out for indirect damages caused by gross negligence, and cyber liability + E&O insurance certificates naming Acme as additional insured.
The SOW defers Phase 2 scope and price to a "separately issued SOW". This is unacceptable risk — in similar engagements Phase 2 has run 2-3× the Phase 1 fee. We require: Phase 2 fee CAPPED at $250K written into this SOW, Phase 2 acceptance criteria pre-defined as part of Phase 1 deliverables, and option to engage an alternative implementation partner for Phase 2 without Phase 1 penalty.
Five deliverables duplicate documentation Acme has already produced (board-approved AI strategy doc, P2P SOP with SAP transaction codes, 15+ system landscape doc, 10-risk SOX register, AI inventory). Please re-scope these as gap-analyses against existing documentation, not net-new discovery. We've identified ~$140K of fee recovery available through this re-scope.
Four deliverables, hand-off-ready
Calibrated Redline (Word)
The paste-ready CFO artifact — line-by-line deliverable classification, drafted counter-proposal language by severity, ROI recalibration narrative, and Pre-Contract Checklist red-flag callouts. Hand to your CFO.
Calibration Workbook (Excel)
6-tab structured workbook — Executive Summary, Deliverable Redline, ROI Sensitivity (5×5 grid), Counter-Proposal Notes, Checklist Audit (all 42 questions w/ status), Parsed SOW. For analysts who need to integrate findings into procurement or vendor risk dashboards.
CFO Briefing Deck (PowerPoint)
8-slide visual summary — title + recommendation pill, executive summary, findings count, ROI recalibration, critical sign-blockers, pricing breakdown, checklist coverage, next steps. Steering-committee-ready in under a meeting.
Calibration Notes (Markdown)
Source markdown of the full calibration for archive, text-search, and embedding sections into other reports or vendor risk systems.
Why the Calibrator is the right second opinion
Independent — Never Aligned with Vendors
We have no consulting partnerships, no resale agreements, no friend-of-the-firm relationships. Every redline is from the buyer's perspective only. We don't hesitate to flag a Big-4 SOW as a hidden Phase-2 trap or call out a Palantir FDE pricing structure that locks the buyer in.
Grounded in Your Environment
The calibrator runs our Intelligence Graph ontology pipeline on your environment docs in Stage 2. When the SOW says "implement in S/4HANA" but your IG ontology shows ECC 6.0, we flag the 3-5× cost mismatch with specific SAP version citations — not a generic "environment may differ" disclaimer.
Deterministic ROI Math
Same compute_roi() function that powers the IG opportunity ROIs. Re-runs the vendor's savings claim against your actual volumes and labor rates, then produces a 5×5 sensitivity grid. The math is auditable line-by-line — your CFO can defend it in a board meeting.
Paste-Ready Counter-Proposal Language
Every drafted note reads like the CFO's lawyer wrote it. No AI-tells, no "let's work together" stilted phrasing. Counter-proposals are categorized by severity (critical sign-blocker / important / minor) so the buyer knows what to negotiate hard vs accept.
42-Question Pre-Contract Checklist
Our proprietary checklist distilled from real Anthropic-Blackstone JV, OpenAI DeployCo, Palantir FDE, and Big-4 readiness engagement RFPs. Every question is one a buyer who got burned wishes they had asked. The calibrator applies the checklist automatically and surfaces must-asks.
5-Business-Day Cycle
Submit the SOW Monday morning, get the redline + CFO briefing deck back by Friday. The compute itself runs in 5–10 minutes; the rest of the cycle is buffered for the live 60-minute debrief with our principal.
Refund if you don't see 5× the fee
If our calibration doesn't identify at least $75K of savings (5× the $15K fee) in your vendor SOW within 7 days of delivery, we refund. No questions, no clawbacks.
Calibrate a Proposal