The Claim-to-Proof Matrix

Every major pitch claim mapped to a specific proof artifact on the demo site, with imperative verification instructions and a shipping status. The structural anchor that all three review paths reference.

What it is

A compact 8-row matrix that says: here’s what SuiteCentral 2.0 claims, here’s the URL where the proof lives, here’s what to DO at that URL to verify, and here’s whether the claim is shipped. Per 09-claim-proof-matrix, all 8 rows are currently marked “Shipped.”

Why it matters (to the adoption case)

This is the reviewer’s fast-lane for verifying the pitch. A CFO, CTO, or COO who is skeptical of marketing language can walk this matrix in ~10 minutes and verify every load-bearing claim directly against live artifacts, rather than taking anyone’s word for it.

It also serves as the structural anchor of Paths A and B — both paths explicitly include a “Claim to Proof Matrix” step. Reviewers in Path A hit it as step 4 (2 minutes) and reviewers in Path B hit it as step 4 (4 minutes). Path C reviewers navigate the matrix implicitly as they walk through technical proof + compliance dashboard + Oracle comparison (which are three of the matrix rows).

The matrix also solves a specific problem: it eliminates ambiguity about what counts as proof. When a reviewer says “prove the AI mapping works,” the matrix points at a specific URL and specific action (“validate confidence + accept flow in narrated video”). There is no ambiguity about what the reviewer is supposed to look at.

The 8 rows (per 09-claim-proof-matrix)

ClaimProof ArtifactVerificationStatus
AI mapping is production-usable/watch/videos/player.html?video=ai-field-mappingValidate confidence + accept flow in narrated videoShipped
Context Sidecar delivers zero-click intelligence/watch/videos/player.html?video=context-sidecarConfirm context switching across record types in narrated flowShipped
Governance evidence is live/compliance-dashboard.htmlExpand SOC2 sections and export evidenceShipped
Oracle differentiation is explicit/squire-v2-media-demo/oracle-comparison.htmlReview 8-row matrixShipped
Watch evidence is complete/squire-v2-media-demo/watch/videos/index.htmlConfirm 19-card playlist structureShipped
Pilot economics are modeled/Squire-Executive-Package-v2/04-ROI-CALCULATOR-STANDALONE.htmlCheck scenarios and assumptionsShipped
Platform quality is enterprise-grade/Squire-Executive-Package-v2/05-TECHNICAL-PROOF-STANDALONE.htmlReview tests and controlsShipped
Pilot execution is gated/Squire-Executive-Package-v2/13-PILOT-30-60-90-STANDALONE.htmlValidate day-60/day-90 gatesShipped

Three observations about the matrix

  1. Every row is Shipped. No “Planned,” no “In Progress,” no “Roadmap.” The matrix is claiming production readiness across every load-bearing point of the pitch. This is consistent with the hook video’s present-tense framing (“is production ready today”) and the elevator pitch’s “This isn’t a pitch deck — it’s production code.”
  2. Verification is imperative. The verification column doesn’t say what the artifact contains — it says what the reviewer should do. “Validate.” “Confirm.” “Expand.” “Check.” “Review.” This is good async review design: the reviewer doesn’t have to guess what they’re looking for.
  3. Not every pitch claim is in the matrix. Notable absences: there’s no row for the four enterprise safety mechanisms (Reasoning Trace Engine, Governance Pacer, DLP PII Shield, Approved To Apply) from ai-governance-layer-video; no row for MDM Central as a differentiator; no row for the NL Action Gate; no row for dual-ERP equal-citizens support. These are implicitly covered by the Context Sidecar, Governance, Oracle, and Technical Proof rows, but they’re not explicit matrix rows. A reviewer could walk the matrix and still have questions about these specific claims.

The “19-card playlist” detail

The Watch evidence row says the verification is to “confirm 19-card playlist structure.” This gives us the total count of videos in the main Watch playlist: 19. Given the 16 modules + 2 Context Sidecar variants (full + highlight) + 1 Executive Reel = 19. The seven storyboard scenes are NOT in the 19-card count — they live in a separate playlist accessed from the storyboard page.

This is a useful structural fact for demo-site and three-review-paths: the Watch playlist is 19 cards, the storyboard is 7 scenes, and they are distinct surfaces.

The “05-TECHNICAL-PROOF” distinction — RESOLVED

Earlier I flagged this as a “05-vs-04 numbering discrepancy.” Investigation confirmed there are two separate technical-proof files in the repo, not a renumbering:

  • 04-TECHNICAL-PROOF.md (~4,400 chars): the full internal technical proof. 4-tier feature inventory with version numbers (v2.5, v3.2, v3.3, v3.4), architecture stack, component LOC counts, source file paths, NetSuite sandbox ID TSTDRV2698307, verified CRUD capabilities. See 04-technical-proof.
  • 05-TECHNICAL-PROOF.md (~1,300 chars): the condensed public-facing bullet summary. Six sections (tests, AI providers, connectors, modules, infrastructure, production readiness), no source-file paths, no version numbers. See 05-technical-proof.

This matrix correctly references 05-TECHNICAL-PROOF-STANDALONE.html as the proof URL for claim #7 (“Platform quality is enterprise-grade”) — because 05 is the right length and format for the 10-minute Technical Proof step in Path C. A reviewer who needs deeper detail can walk up to 04-technical-proof from there.

Important drift in 05-TECHNICAL-PROOF: the connector list in 05 says “NetSuite, Salesforce, Shopify, Oracle, Business Central” — that’s 5 connectors and includes Shopify. The canonical wording in 26-canonical-metrics-and-wording names 6 connectors: NetSuite, Business Central, Salesforce, HubSpot, ShipStation, Oracle (no Shopify). 05-TECHNICAL-PROOF is stale and does not conform to the canonical wording style guide. A Path C reviewer will see the stale Shopify claim. Flag for asset owner.

Relationship to the four enterprise safety mechanisms

ai-governance-layer-video names four enterprise safety mechanisms (Reasoning Trace Engine, Governance Pacer, DLP PII Shield, Approved To Apply). This claim-proof matrix does NOT have dedicated rows for each of these four. They are covered implicitly:

  • Reasoning Trace Engine → part of the “AI mapping is production-usable” row (the narrated video shows confidence + accept flow, which is the reasoning trace surface)
  • Governance Pacer → part of the “Governance evidence is live” row (the compliance dashboard would show rate-limiting posture)
  • DLP PII Shield → part of “Governance evidence is live”
  • Approved To Apply → part of “Pilot execution is gated” (the day-90 Governance gate metric checks “evidence package exported and reviewed”)

So the four safety mechanisms ARE verifiable via the matrix, just not in dedicated rows. A CTO who wants to verify them specifically would need to look at the Governance dashboard + the narrated AI mapping video together.

Relationship to the three review paths

All three review paths reference the Claim-Proof Matrix explicitly:

  • Path A (Executive) — Step 4: Claim-to-Proof Matrix (2 min). Lightweight skim before the Pilot Decision Memo.
  • Path B (Leadership) — Step 4: Claim-to-Proof Matrix (4 min). Deeper walk-through with time to actually open a few of the proof URLs.
  • Path C (Deep Proof) — Does not have an explicit matrix step, but the matrix’s rows for Governance, Oracle, Technical Proof, and Pilot execution all appear as Path C steps. Path C effectively walks through the matrix by visiting most of its proof URLs directly.

Sources