SEQ — Living Roadmap¶
What this page is — strategic memory for SEQ
The persistent roadmap: the single north-star document AI agents work toward and humans review. It holds why we believe what we believe, how we plan to win, what we expect, and what actually happened — so you can look back and see whether the bet paid off.
Research → Strategy → Expectations → Return → Report, all in one place, all dated.
Marker key (carried from the strategy docs): 🟢 Confirmed · 🟡 Assumption · 🔴 Gap · 🎯 Target The markers are the "living" part — a 🔴 gap that research fills becomes 🟢; a 🎯 target that a monthly return hits gets checked off. Watching them move is the roadmap working.
Narrative depth (how a build works, full risk register) stays on the entity page → Superior Equipment. This page is the measured, decision-grade companion.
0 · MARKET RESEARCH & ANALYSIS — the why¶
What we know about the market, what we're assuming, and what we still need to learn. Dated so we can see what we believed at the time.
Demand drivers (as of 2026-06-07) — 🟢 the group sells into construction, dust-control compliance, fire/wildland response, and municipal water management. Water-truck demand is seasonal.
Seasonality — 🟡 peak water-truck season runs May–September; sourcing should run counter-seasonally (acquire salvage off-peak, deliver into peak). Validate against actual build completion + sell-through timing once RETURN data lands.
Supply side (SEQ-specific) — 🟢 inputs are salvage chassis bought at auction; two parts supply chains SEQ prices against each other — new parts from SWTP, used from CTP. 🔴 Gap: salvage-chassis availability + price trend over a year (the supply that caps throughput).
Addressable market — 🔴 Gap: dollar size of the Phoenix/Southwest salvage + water-truck segment. Needed to justify the 8-builds/month throughput target and size the opportunity for a stakeholder (NextGear/Tamarack). Assigned to research pass.
Competitive set — 🔴 Gap: named competitors and SEQ's quantified edge. Working thesis (🟡): SEQ's edge is four build stages under one roof (weld / paint & body / tank / dump-bed), no outsourcing on the critical path → control of quality, cost, schedule.
SWOT (working — refine as gaps close)
- Strengths 🟢 four-shops-in-house; ~$10K/truck cost discipline; dual parts sourcing; longest operating history.
- Weaknesses 🔴 single-customer concentration (bills STE); a bottleneck station caps output; parts lead times (China new-parts, teardown pace for used).
- Opportunities 🟡 outside build work; counter-seasonal sourcing arbitrage.
- Threats 🔴 salvage-price inflation; a demand dip at STE/STR flows straight back to the shop.
1 · SALES & GO-TO-MARKET STRATEGY — the how¶
SEQ is the production engine, not a storefront — its "sales" is throughput sold internally to the dealership. Customer-facing GTM lives on the STE/SWTP roadmaps.
- Customer 🟢 — principal buyer is the dealership STE, at arm's-length per-build pricing. Whether SEQ takes outside build work is an 🟡/🎯 open strategic question (would spread single-customer risk).
- Positioning 🟢 — in-house, four-stage rebuild = quality/cost/schedule under one roof.
- "Channel" — demand is pulled by the group's own sell/rent channels (STE sales, STR rental), not external marketing. 🎯 The lever is throughput + mix, not lead-gen.
- Pricing 🔴 — build price to STE is a capture-needed gap (blocks margin math).
- Play for the year 🎯 — hold 8 builds/month, drive per-shop utilization to find the bottleneck, and decide the outside-work question with a real addressable-market number.
2 · EXPECTATIONS — targets (the contract)¶
Set by Brandon + MC-ATG; change only on a logged decision. Every SEQ work item should move a row.
| KPI | 🎯 Target | Horizon | Owner | Source of truth |
|---|---|---|---|---|
| Finished builds / month | 8 | Monthly | Shop | Build completions (QB invoice to STE / ops log) |
| Rebuild cost / truck | ≤ $10,000 | Per build | Shop | QB job cost |
| Build price to STE | 🔴 capture | Per build | Brandon | QB P&L — SEQ |
| Gross margin % | 🔴 set once price+cost captured | Monthly | AM | QB P&L — SEQ |
| Per-shop utilization (weld/paint/tank/bed) | No station < 70% | Monthly | Shop | Utilization tracker (to stand up) |
| AR days to STE | ≤ 30 | Monthly | AM | QB AR aging — SEQ |
3 · RETURN — actuals (the books)¶
Fed monthly after close. No hand-typed numbers — each cell cites source + as-of. — = pipeline not yet wired (Phase 2).
| KPI | 🎯 Target | Actual | Variance | As of | Source / status |
|---|---|---|---|---|---|
| Builds / month | 8 | — | — | — | ops log / QB — wire (quick win) |
| Rebuild cost / truck | ≤ $10K | — | — | — | QB job cost — wire (quick win) |
| Build revenue | — | — | — | — | QB P&L — pending gl_fact (#55) |
| Gross margin % | — | — | — | — | QB P&L — pending |
| Per-shop utilization | ≥70% ea. | — | — | — | tracker not stood up |
| AR days to STE | ≤ 30 | — | — | — | QB AR aging — wire (quick win) |
Wiring: QB → warehouse/GL → Metabase → this table, via the monthly job. Build count, job cost, and AR wire from QB now; revenue/margin follow the GL-detail pipeline (#55).
4 · REPORT — monthly review (the loop)¶
After each close: diff Expectations vs Return, name the misses, set next month's actions, append a dated entry to the decision log. Humans read it here; agents ingest it.
Latest month: (pending first close)¶
- On track: —
- Behind target: —
- Research that changed (gap closed / assumption confirmed or broken): —
- Actions (owner → by when): —
- Decisions logged: —
How the monthly report is produced
- AM confirms SEQ books closed. 2. Job pulls RETURN metrics (Metabase → table). 3. An agent pass writes Latest month: variances, misses, 3–5 actions, and any marker changes in §0/§1.
- Dated review → decision log; "Last review" updates. 5. Quarterly: MC-ATG refreshes targets and works the 🔴 research gaps (addressable market, competitive set, salvage supply).
How AI uses this page¶
This is direction for every SEQ-facing agent. Work items trace to an Expectations row; the Return says whether the work moved the number; the Report is the monthly recalibration; and the markers in §0/§1 are the standing to-do list — every 🔴 is research to run, every 🟡 an assumption to prove or kill. That is the persistent memory that keeps autonomous work pointed at outcomes and honest about what's still unknown.