Back to _LANDSCAPE-SYNTHESIS · _CROSS-COMPETITOR. Per-geo detail: _geo/us/note · _geo/canada/note · _geo/australia/note.
Built 2026-06-16 from three parallel exa/web scans (US, Canada, Australia) to de-bias the UK-only strategic review. Each geo scored 1–5 on the four beachhead axes by an independent agent against real gov/industry/vendor sources.
The one headline that changes the thesis
No market is empty. All four are VIABLE-but-contested. The UK-framing’s “two empty columns (area 15 claims-recovery + area 21 cross-firm cost)” was partly an artifact of incomplete scanning. Scanned properly, every English-speaking market already has a live cluster competing for the claims/recovery wedge and the QS buyer:
- US — Magra is already US-native and owns ~80% of the entitlement+quantum loop; dense incumbent stack around it.
- Australia — a whole cluster of AU-native AI claims startups (ClaimTrack, Varicon, Huntly AI, Consuite, DelaySolve, Styck) + Payapps/Autodesk on the payment rail.
- Canada — thinner, but narrow payment/lien tools (KiwiCode et al.) + Storia/CREO emerging + Revay owning the services end.
- UK — Gather Insights creeping from notices into the middle; Magra-clone risk.
What is still open everywhere: the full integrated loop (capture → Commercial Event → quantum → recovery → cross-firm cost) and the cross-firm realized-outcome benchmark (area 21). The edge is execution + data density, not an empty market.
Scorecard (1–5; whitespace 5 = wide open)
| Geo | Willingness-to-pay | Claims / recoverability | Competitive whitespace | Cold-start ease | Overall | Verdict |
|---|---|---|---|---|---|---|
| United States | 5 | 5 | 2 | 2 | 3.0 | VIABLE |
| Canada (Ontario-first) | 3 | 4 | 4 | 3 | 3.5 | VIABLE |
| Australia | 4 | 5 | 2 | 3 | 3.5 | VIABLE |
| UK (est. from v1 memo, same scale) | 3 | 4 | 3 | 3 | ~3.2 | services-first |
Per-geo one-liner
- US — Largest, most claims-rich ($60.1M avg North America dispute, 2024 Arcadis), highest WTP — but already half-occupied by Magra + a dense stack, brutal fragmentation, and a diffuse commercial buyer (no dedicated QS role; cost spread across PMs/estimators/accountants). Best entered services-first in one segment/metro around the open cross-firm-cost seam.
- Canada — A live UK-style prompt-payment + adjudication wave (ON 2019 → SK/AB/MB/Federal live 2025) plus the emptiest competitive field make it genuinely viable — but “win Canada” really means “win Ontario, then re-cut per province.” Smaller, thinner data density. ODACC adjudications 50→324 in 4 yrs; awards land far below claimed = our exact weak-evidence gap, proven.
- Australia — The world’s most pro-recovery regime (state SOPA statutory rapid adjudication, subcontractor-led; AS 4000 28-day EOT) — half our capture loop is legally mandated — plus unusually open public adjudication datasets (VIC/QLD/NSW) that are ready-made benchmark fuel. But contested by a native AI claims cluster + Payapps. Services-led.
What this means for the decision
- “Empty market” is dead — kill it from the pitch. The defensible story is the integrated loop + cross-firm realized-outcome data, not “no one does claims.” This holds in all four geos.
- Services-first is geo-independent. Every market has a high-value claims-services incumbent that proves the buyer pays for recovery (US: HKA, Arcadis, Long International; Canada: Revay, 50 yrs; Australia: Mitchell Brandtman, MBM; UK: the claims-QS world). The v1 SERVICES-FIRST vehicle call survives de-biasing.
- The axes trade off — there is no free beachhead:
- Want the biggest prize + highest WTP → US, but hardest cold-start + Magra already there.
- Want whitespace + regulatory tailwind → Canada/Ontario, but smallest + province-fragmented.
- Want the best recoverability regime + bootstrappable benchmark data → Australia, but contested by native AI cluster.
- Australia uniquely de-risks the moat. Its public SOPA adjudication outcome datasets let us seed the area-21 realized-outcome benchmark without the cross-firm-trust cold-start that blocks it elsewhere. That is a genuinely new strategic option.
- Closest whole-model analogs to watch (across geos): Magra (US — entitlement+quantum loop), Huntly AI (AU — software bundled with commercial consultants = literally our services-first shape), Storia/CREO (Canada — source-linked record + schedule + claims narrative). These three are the nearest things to what we’d build.
New competitors surfaced (beyond the ~60-tool corpus)
- US: Magra, SmartPM, InEight, Clearstory, Extracker, StratusVue, HCSS, Gordian/RSMeans, Kahua, Siteline, Flashtract, Levelset (Procore), + services HKA, Arcadis, Long International, Delta Consulting, Precision Scheduling.
- Canada: KiwiCode (targets our exact $15–80M ICI band), BuilderLien, GCPay, PaySub, Billeasy, RationalGo, 4castplus, Altus Group (cost data), Storia/CREO, Revay (services).
- Australia: BuildPass, Payapps (Autodesk), Procore AU, ClaimTrack, Varicon, Huntly AI, Consuite, DelaySolve, Styck, Aphex, Cordell/Cotality (cost data), Rawlinsons, Mitchell Brandtman, MBM.
v2 grading — independent scan vs ChatGPT v2 (DONE 2026-06-16)
v2 memo: _handoff/strategic-review-memo-v2.md (589 lines, geo-balanced, all 16 corpus files used, real gov citations). It weighted the ranking 30% WTP / 30% recovery-regime / 20% competition / 20% cold-start → Australia 4.35, UK 4.05, US 3.55, Canada 3.35.
5 strong convergences (high confidence — two independent passes agree):
- Australia = best first beachhead. Both landed here separately. v2: “cleanest evidence-to-cash loop via Security of Payment.” Us: “world’s best recovery regime, half the capture loop legally mandated, open SoPA data seeds the area-21 moat.”
- US = wrong first market despite biggest TAM + highest ACV ($50–150k). v2 ranked it 3rd (“scale market, not learning market; Magra direct; platforms denser”). Us: US overall 3.0 (lowest), whitespace 2 / cold-start 2.
- Services-first, geo-independent. Both; every market has a high-value claims-services incumbent (Revay/HKA/Mitchell Brandtman) proving the buyer pays for recovery.
- “Empty market” is dead → precise claim = “no scaled, trusted, jurisdiction-specific recovery platform with cross-firm outcome data.” v2 and this scan reframed identically.
- Single-beachhead density first; area 21 moat gated by trust not modelling. Both.
1 divergence (weighting, not contradiction) — Canada rank: our scan put Canada tied-top (3.5, whitespace 4 — emptiest field + 2025 prompt-pay/adjudication tailwind + ODACC awards << claimed = our weak-evidence gap proven). v2 put Canada last (3.35 — province-fragmented, less mature, “smaller than US without being as clean as AU”). Reconciliation: v2 is probably right for a data-moat business — Canada’s emptiness is partly because it is too thin to build density. Both agree “win Ontario, not Canada,” and both keep Canada as expansion #3 (Ontario/Alberta). Not worth overriding AU-first.
Net: the independent scan validates v2 on every decision that matters. Verdict to carry forward: SERVICES-FIRST, Australia-first, expansion UK → Canada(Ontario) → US, moat = cross-firm realized-outcome data (AU public SoPA data uniquely bootstraps it).
Real gap both surfaced (next research, if any): our dossier voice-of-user is biased to field/admin/safety/QA users (Raken/Fieldwire/BuildPass reviewers) — it proves adoption + workflow gaps but not QS/commercial willingness-to-pay for recovery. That is the #1 thing to validate with real AU buyers before committing capital.