Record ids. D-6-n is a Week 6 decision, J-n a journal entry, F-0n a review finding, DR6-n a data request, A-n an assumption; all are in the engagement record.
D6, part 2 of 3. Final version. Source: the Week 1 proposal to Sarah Mitchell (my-work/week-1/northwood-pitch-presentation.html, pinned in the manifest). The instruction of Daniel Vetsko (engagement owner) (D-6-6): it is fine to go against some Week 1 promises, but each alternative route must be explained for transparency. Version 2 records the close-out (D-6-60 to D-6-67).
| Week 1 promise | What was delivered | Route taken, and why | Status |
|---|---|---|---|
| "TCFD/CSRD-aligned LP narrative, ready to send this quarter" (fund level) | An LP climate report in TCFD order (Governance, Strategy, Risk management, Metrics and targets) with framework content indices for IFRS S2, ESRS E1, SFDR PAI and ILPA EDCI, now including Scope 3 (estimated), PCAF financed emissions and the exposure by fund | TCFD structure because the client's protocol requires it and says Northwood does not claim full TCFD compliance. CSRD is covered in the library and the content index, not claimed in the report: the data room shows no EU LP and the CSRD scope questions are legal (A-14). | Delivered, not yet releasable (Northwood actions). Final; released from draft on the engagement owner's approval (D-6-60); Investor Relations review and sending are Northwood's |
| "Company-specific disclosure summary per portco" | Ten board-tier climate pages, one per company (Tier 3 of the protocol), single-company data only | The protocol forbids portco-level data in LP disclosures. Its own per-company vehicle is the annual board climate review (D-6-12). | Delivered; peers only for two companies |
| "Reusable template and methodology docs" | The reference library (playbook, 11 framework files and 2 method standards, with verification labels, applicability rules, methodology notes including the Scope 3 and PCAF method), the engine and quality checks, a re-run note | As promised | Delivered. The playbook's hours table is the scout's estimate and should not be quoted |
| "Your team runs the next update without us" (persisted playbooks) | A re-run note whose nine steps rebuild everything and a clean re-run check that proves it | The steps are commands, so a technical colleague runs them; Sarah reviews the result. | Delivered with that caveat |
| Week 1 promise | What happened | Why | Status |
|---|---|---|---|
| Four priority companies for full build, the other six "queued for the same build at a fraction of the cost" | All ten companies were covered in Weeks 3 to 6 through one pipeline | Daniel ruled that all ten are in scope (D-6-32); one pipeline made the extra six cheap | Done earlier than promised |
| Scope 3 screen for Harvest Ridge and Meridian (Week 3) | A spend-based Scope 3 estimate for all ten companies, grade C, with a range; mainly categories 1, 4 and 6 combined | Owner rulings D-6-61, D-6-65, D-6-66; the data room has no supplier spend by category, so the estimate uses cost lines less labour | Delivered for all ten, wider than promised; category detail needs DR6-16 |
| Financed emissions (implied by the LP narrative) | PCAF financed emissions for all ten companies, by fund | D-6-64 | Delivered |
| "Persisted in CLAUDE.md, reference architecture, playbooks" | query-pack-v2/CLAUDE.md (the query pack's instructions), the kit, the library and the re-run note persist | As promised | Delivered |
| Schedule: proposal window October 19 to 23 | The build and the gates ran later: gates 6 to 10 November 2026 | The schedule moved; the four gates were run later than the proposal's window. | Delivered later than the proposal's window |
| Cadence: a quarterly refresh was proposed | An annual refresh retainer (option A) was recommended | Option A re-runs the ten companies on the protocol's annual cycle, with the fixed work already built (pricing reflection, section 4). | Different route, stated |
| Named consultant bios within 48 hours of signature | Not a Week 6 deliverable | n/a | Outside this record |
Portfolio edition: Kith Climate cohort case, fictional client.