Presenter Script: Week 2 Climate Data Review for Sarah Mitchell (v1)
Portfolio edition: Kith Climate cohort case, fictional client. This is the script as used for the September 26, 2026 presentation of deck v2, built from memo v4. Later corrections are in memo v6 and deck v4.
Goes with: data-audit-deck-northwood-2026-09-v2-draft.html
Audience: Sarah Mitchell (COO). David Okafor and Lisa Brennan may join.
Length: about 20 minutes for pages 01 to 09, plus questions. Pages 10 to 12 are the cohort appendix; skip them with Sarah unless she asks how the work was done.
Source of every figure: memo v4 (data-quality-memo-northwood-2026-09-v4-draft.md). If asked for detail you don't have, say it's in the memo, section named below, and follow up.
Before you start:
- Open the deck full-screen.
- Move between pages with the arrow keys or the numbered rail on the left.
- Pointing at a chart element shows its detail; clicking works too.
- Have the memo open in another tab.
The one message to leave her with: LP disclosure for FY 2025 is achievable once known errors are fixed. Buyer diligence isn't, and Meridian is the urgent case. Four answers from her on Monday keep Week 3 on schedule.
01 · Headline (about 2 min)
On screen: "Your protocol exists on paper. It hasn't run for 2026." On the right, all ten portfolio companies are ranked by readiness score, with a dotted line at 40, where Partially ready starts.
Say:
- "Northwood has a good climate protocol, but the data room shows no sign the 2026 cycle has run. The ESG score tables are empty, and the Q4 2025 LP letter has no climate section, although the protocol requires one."
- "Is the data ready? It depends who's asking. For LP disclosure, yes, once known errors are fixed: the energy data exists for all ten, and the protocol accepts management estimates. For a buyer, no: all ten score between 26 and 38 out of 100, in the lowest band."
- "Meridian matters most. It's tied with Sterling for the best score, but it goes into confirmatory diligence in Q4 on a $363.9M to $400.9M process. The risk there is friction in diligence, not a discount on price."
Point to: the green rungs. "Six companies are one named fix from the next band. The four amber ones need more than one fix."
02 · Act now (about 3 min)
On screen: four stepped cards ranked by value at stake, with a detail panel on the right. Point at each card in turn.
Say: "We ranked findings by what's at stake for Northwood, not by how messy the data is. Four issues have high value, a deadline within about a year, and a fix within reach."
- Meridian. Two CFO-signed submissions disagree on gas and fleet, and the exit plan's diligence scope has no ESG workstream. Fix: reconcile against utility invoices before the LOI. Owners: Patricia Wells (CFO) and Mark Sullivan (deal lead). Effort: low.
- Harvest Ridge. Its two submissions disagree on every total. About $34.0M of revenue from three top-10 customers needs Scope 3 data at 2027 renewals. Fix: one clean resubmission. Owner: the Harvest Ridge CFO (Thomas Nguyen, to be confirmed).
- Sterling. The totals match, but every facility figure conflicts. It has $25.0M to $42.0M of defense RFPs, and Boeing Defense makes carbon data mandatory from 2027. Fix: reconcile the facility split against the three utilities' invoices.
- Northwood's own LP letter. The protocol requires a climate section in the Q4 2026 letter, due mid-February 2027. Owners: Sarah and Jennifer Park (IR). Kith drafts it in Week 6.
If asked about other issues: "The memo also has a 'resource and plan' group (Harvest Ridge's supplier programme, Pinnacle's refinery clients, Ridgeline's services line) and a list of quick wins. Those are in the memo's 'Where Northwood's Attention Should Go' section."
03 · LP-readiness scores (about 2 min)
On screen: ten horizontal bars out of 100. Green means one fix reaches Partially ready; amber means it doesn't. The dashed marker is at 40.
Say: "All ten are in the same band, but some are much closer to moving. Sterling and Meridian are at 38. Harvest Ridge is lowest at 26, but one clean resubmission takes it to 40."
Click a bar to show the financial exposure behind that company's data.
04 · How the score works (about 2 min)
On screen: eight weighted dimensions on the left, the three bands on the right.
Say:
- "Each company is scored 0 to 4 on eight dimensions. Energy data completeness carries the most weight, at 25%."
- "An honest caveat: today the ceiling is about 63. Four dimensions (emissions calculation, targets, governance, frameworks) carry 40% of the weight, and nobody scores above 1 on any of them. So the real differences between companies are in the energy data."
- "We set the weights on September 22, after the first checks had run, so with the data in view. They're frozen until the responses come back."
Likely question: "Why are Sterling and Meridian 2.5 points short?" "Their unrounded scores are 37.50. Both fell below 40 when we tightened a rule during the audit: naming someone who compiles data no longer counts as governance. Treat their band as a close call. The dimension profile is the more reliable read."
05 · Findings by company (about 4 min)
On screen: tabs for Meridian, Harvest Ridge, Sterling and Ridgeline, with short summaries of the other six below. Each finding is tagged Confirmed (from the files), Reported (the company's word) or Inferred (our interpretation).
Meridian tab: zero gas recorded at four sites where the questionnaire reports 12,500 therms, and 3,200 gallons of fleet fuel for a company that says it has no fleet.
- Read the "If it isn't fixed" box: buyer questions under time pressure, a slower ESG workstream, and possibly a request for a specific representation.
- Then say: "No price effect is expected for an office-based business. The cost is management time and credibility at the worst moment." This is our inference, tagged as such.
Harvest Ridge tab: submissions differ by 14.6% on electricity, 18.7% on gas and 62.2% on fleet diesel. The board rates the customer requirement "Low-Medium" against about $34.0M of revenue.
Sterling tab: all six facility pairs conflict, and the R-22 chiller replacement has been deferred twice.
Ridgeline tab: its board has sized a sustainability services line at $1.5M to $8.0M a year, but it depends on the boundary decision on page 09.
The six summaries: mention two only.
- Pinnacle: the FY 2026 insurance budget grows about $0.1M against a broker forecast of about $0.2M. That's an EBITDA point for David Okafor.
- NovaCare: two misfiled rows overstate its gas by 90.5%.
06 · Three items can follow (about 1 min)
Say: "The memo has seven items for Northwood to confirm. Four are needed Monday and are on the last page. These three can wait: fuel-card classification (NW-3), who provided the physical-risk scores (NW-4), and the kickoff inputs we asked for (NW-6). None of them changes a Week 3 figure."
07 · Data requests and timeline (about 2 min)
On screen: a vertical timeline. Requests are issued Monday, September 28. Tier 1 is due Monday, October 5, and Tiers 2 and 3 on Monday, October 12.
Say:
- "There are eleven requests, one per company plus one to Northwood, all sent in Northwood's name. Each company sees only its own data. There are 70 asks in total."
- "The tiers set the order of work, not importance. Tier 1 is whatever would block the Week 3 baseline. Sterling has no Tier 1 asks, yet its invoices matter most commercially."
- "Harvest Ridge, Patriot and Ridgeline are most at risk of answering late."
- Point to the grey box: "If answers are late, Week 3 still covers all ten. Where the template is known to be wrong, we use the questionnaire figure and flag it. Where the two submissions conflict, we show a range. Nothing is quietly filled in."
08 · How this was built (about 1 min)
On screen: a ring of seven pipeline steps.
Say: "Every figure comes out of scripts that re-run from the raw files, and a final check re-runs everything before anything goes out. The scripts need only standard Python, so Northwood's own team can run the same process next quarter. That's the 'system that compounds' you asked for. The playbook moves to a Northwood location at the Week 6 handover."
Keep this short with Sarah. It's reassurance, not the point.
09 · Monday's decisions (about 3 min; end here)
Say: "Four answers from you on Monday keep Week 3 on schedule."
- NW-1: confirm the CFO for Harvest Ridge and Patriot. The data room lists Thomas Nguyen for both, and it blocks two of the eleven requests.
- NW-2: Ridgeline's reporting boundary.
- Our recommendation: its own offices and fleet, with the client buildings it manages (about 285M kWh) recognised but reported separately.
- The reasoning: the owners pay for that energy and make the capital decisions. This is subject to a review of the management contracts.
- Why it matters: the answer changes Ridgeline's footprint by orders of magnitude.
- NW-10: where Meridian's sale process stands, confirmed with Mark Sullivan. Our newest status document is from Q1 2026.
- The sequencing call: whether the six additional company requests (Keystone, NovaCare, Patriot, Pinnacle, Ridgeline, Trident) go out Monday, follow later, or wait for the January 2027 cycle.
Close: "We scoped this for four companies and covered all ten at no extra cost. Whether six more CFOs hear from us this week is your call."
10 to 12 · Appendix (cohort review only)
Use these pages with Sarah only if she asks how the work was done.
- 10: eight build stages, and how the memo grew from 4,231 to 7,749 words as two review rounds sharpened it.
- 11: who did what at each stage (instructions, the authoring pass, scripts, read-only review agents, the Kith lead).
- 12: how the deliverables connect, from evidence to what Northwood keeps to Weeks 3 to 6.
Questions to be ready for
| Question | Answer |
|---|---|
| "What happens to the Meridian deal if this isn't fixed?" | Buyer questions under time pressure, a slower ESG workstream, possibly a specific representation. No price effect expected. Management time and credibility are the cost. |
| "Why not put a dollar figure on Meridian?" | The risk is friction in diligence, not a calculable discount. EV conversions exist in our working files for discussion only; they aren't in the memo because full-loss figures would overstate the exposure. |
| "Have Patricia Wells and Mark Sullivan been told?" | Don't guess. Confirm with Daniel before the meeting. |
| "Can we trust these numbers?" | Every figure traces to a logged source. The questionnaires were transcribed twice and compared: 151 shared figures, 0 mismatches. A final script re-checks the memo and deck against the pipeline. |
| "Is Patriot's fleet really 42.5% of its footprint?" | It's a Kith indicative estimate (EPA 2025 factors, US-average grid). It gets refined in Week 3. The fleet volume itself looks implausible for 8 vehicles, which is why it's a Tier 1 ask. |
| "When were the thresholds set?" | The weights on September 22 and the materiality thresholds on September 25, both after we'd seen data. The memo says so. From next cycle they're set before any data is reviewed. |