A founder built three independent SaaS applications. FM Control Hub. Care Hub. Compliance Hub. All in the span of months, all live in production, all backed by clinical AI running on real patient data and live demo environments seeded with months of simulated operational richness.
Three different AIs then audited them using three different methodologies. ChatGPT in January. Perplexity last month. Claude in May. Each pass assigned a different valuation range. Each pass found different gaps. Each gap was remediated.
This is the full story of what was built, what the audits revealed, and where the applications are worth today.
The Audit Story — From £30k to £280k in Four Months
ChatGPT First Pass: The Baseline (January 2026)
The founder ran ChatGPT 4 through the full codebase, architecture docs, demo environments, and database schemas. The verdict was honest and conservative.
Assigned valuation range: £30,000–£45,000.
Why so low? The audit was the right approach but limited by scope. ChatGPT evaluated the code structure and feature completeness, but the assessment didn't include: - Live verification of the AI running in production - Database richness and demo data depth - Governance documentation for handover - Industry compliance compliance packs (FHIR maturity, FM standards, etc.) - Security findings and remediation plan
ChatGPT's findings included: - Documentation gaps (standard for quick builds) - AI cost model unspecified - Secret management needs review - Demo environments lack seeding (no real-world-like data) - No handover or governance pack
The verdict wasn't wrong. It was accurate given the scope.
30-Hour Remediation Sprint (February 2026)
The founder took every ChatGPT finding as a literal task list:
FM Control Hub: - Seeded demo with work orders, preventive maintenance schedules, patrols, energy readings, asset inventory - Full cost model documented: Sonnet calls ~2p per summary, Haiku calls <0.3p per briefing - Governance pack created: org structure, handover docs, access controls
Care Hub: - Verified clinical AI running live on staged patient records (FHIR-compliant) - Demo environment enriched with synthetic patient journeys, wellness summaries verified - Clinical governance docs added (AI decision logging, audit trail, liability framework) - Security review completed; all findings addressed
Compliance Hub: - Industry packs tested against real regulatory frameworks - Demo data for all six verticals (FM, Care, Construction, Hospitality, Energy, Legal) - White-label engine documented with customer onboarding flow - Multi-site compliance rollup verified
Secret management overhauled: All credentials moved to .env, no hardcoded keys in commits.
Total effort: 30 hours across deployment, demo seeding, documentation, and security fixes.
ChatGPT Re-run: The Second Pass (March 2026)
Same auditor (ChatGPT 4). Same methodology. One month after remediation.
Assigned valuation range: £240,000–£260,000.
That's an 8× jump in four weeks.
What changed? - Live demo environments now fully seeded—real-world workflows, months of simulated data - Governance documentation 180+ pages, secret-free, ready for buyer handover - AI cost model specified (enables forecasting for acquirer) - Clinical validation (Care Hub AI verified running, not theoretical) - Security findings addressed (no open issues, remediation log clean)
ChatGPT's second pass didn't find new gaps—it found depth and completeness in the areas it flagged before.
Perplexity Independent Audit (April 2026)
A completely different AI. Different methodology (more research-oriented, web-context-aware, comparative benchmarking).
Assigned valuation range: £275,000–£350,000.
Perplexity's approach was to place these applications against market comparables. FM software (IWMS, Maximo, Archibus) in the SMB/mid-market space trades at 4–6× ARR when mature. Compliance software trades at 5–7× ARR. Care tech with clinical AI integrated trades at 3–5× ARR but commands a premium for live production verification.
Perplexity findings: - FM Hub strength: Full stack (ops, energy, procurement, lifecycle), 30+ modules. Comparative advantage over standalone systems. - Care Hub strength: Clinical AI live—not on roadmap. Genuine differentiator. Verified against FHIR. - Compliance Hub strength: Six industry packs + white-label engine = immediate multi-vertical positioning. - Gaps: One paying pilot would move valuation range to £450k–550k. Governance pack removes key-person risk (crediting work).
Perplexity's 4.2.4 model scored this as a £300k–350k mid-market acquisition candidate if one paying reference existed.
Claude Technical Deep-Dive (May 2026)
Code-level audit. Runtime verification. Database-level inspection. Not just what's documented—what's actually running.
Assigned valuation range: £200,000–£280,000.
Claude's methodology: read the code, trace the execution paths, verify the claims in the documentation. This is the most technically rigorous audit.
Findings: - FM Control Hub: Full-stack implementation verified. Soft FM suite complete. Energy tracking live and working. Operations layer functional. 30+ modules enumerated and tested. No half-built features. Cost model accurate. Verdict: Production-ready, acquisition-grade code. - Care Hub: Clinical AI verified running. FHIR implementation audited (DSTU2, R4 both present). PWA functionality checked. KLOE (key life observation events) schema verified. AI decision logging present. Verdict: Clinically defensible, live and working, not prototype. - Compliance Hub: Six industry packs scanned. White-label routing logic working. AI briefing engine (using Haiku for cost efficiency) tested. Multi-site rollup verified. Verdict: Enterprise-ready white-label stack.
Gaps identified: - One FM demo needed enrichment (work order dates inconsistent in simulation) - Care Hub AI cost model verified but annual projection document missing - Compliance Hub governance docs complete but buyer SOP guide missing
All remediable in 4–6 hours.
Claude's valuation places the suite at £200k–280k based on code maturity, feature completeness, and clinical/regulatory defensibility. The range reflects: - Lower bound: Single-app acquisition + IP licensing - Mid-point: Multi-vertical operator acquiring all three for portfolio expansion - Upper bound: Strategic buyer in FM or care tech, adding into existing product
What Was Actually Built
FM Control Hub: 30+ Modules, Live
Feature set: - Soft FM suite (cleaning, security, catering, courier, uniform, linen) - Hard FM modules (HVAC, electrical, plumbing, roofing, grounds) - Energy tracking and sustainability reporting - Preventive maintenance scheduling - Work order management, assignment, and completion tracking - Mobile-first patrol app (photo attachments, real-time marking) - Supplier and contractor management - Asset lifecycle tracking - Budget and cost forecasting - KPI dashboard with drill-down analytics
Technology: - Next.js frontend (responsive, PWA-ready) - Node.js backend with REST API - PostgreSQL for operational data - Redis for caching and real-time updates - Claude API for AI summary generation (cost: ~2p per FM summary) - Stripe for billing (white-label ready)
Demo data: 6 months of simulated facilities operations (100+ asset types, 500+ work orders, 1000+ patrol records, energy readings for 50 buildings)
Status: Fully functional, acquisition-ready. Code maturity: production-grade.
Care Hub: Clinical AI, Live and Verified
Feature set: - FHIR R4 and DSTU2 compatibility (patient records, observations, care plans) - Clinical AI for wellness summaries (patientjourney analysis, care recommendation generation) - Patient PWA (access own records, request features, secure messaging) - Clinician portal (safe data view, intervention logging, audit trail) - Audit trail and decision logging (every AI summary logged with input/output/confidence) - Multi-clinic support with role-based access - Compliance reporting (regulatory audit-ready) - KLOE schema (key life observation events for continuity of care)
Technology: - Next.js frontend with Tailwind (accessibility-first) - Node.js/Express backend (medical-grade error handling) - PostgreSQL for HIPAA-grade patient data - Redis for session/caching - Claude API for clinical summaries (cost: <0.3p per Haiku briefing, ~2p per Sonnet summary) - Audit logging to immutable event stream
Demo data: 50 staged patient records with synthetic but clinically realistic journeys (chronic conditions, medication histories, appointment logs, wellness events)
Clinical verification: Care Hub AI tested against staged patient data. Summaries reviewed for accuracy, tone, and safety. No hallucinations. Recommendations grounded in FHIR data.
Status: Clinically defensible, live, audit-ready. Crown jewel of the suite.
Compliance Hub: 6 Industry Packs, White-Label Engine
Feature set: - Six industry-specific compliance packs: 1. Facilities Management (COSHH, fire safety, asbestos management, legionella) 2. Care & Healthcare (CQC compliance, safeguarding, infection control, data protection) 3. Construction (CDM 2015, H&S file, site induction, risk registers) 4. Hospitality (allergen management, food safety, staff training, licensing) 5. Energy (MEES, EPC, carbon reporting, decarbonisation plan) 6. Legal (contract compliance, audit trail, document management, privilege log) - White-label customer portal (rebranded, custom domain, custom logo) - AI-powered compliance briefing (Haiku cost: <0.3p per brief) - Multi-site rollup and consolidated reporting - Automated audit trail - Dashboard with compliance KPIs and gap tracking
Technology: - Next.js frontend (responsive, white-label-ready CSS) - Node.js API with industry pack routing logic - PostgreSQL for compliance records - Redis for audit event streaming - Claude/Haiku API for compliance briefing generation - Stripe for white-label billing
Demo data: Full compliance scenarios for all six verticals (audit findings, remediation trackers, certification schedules, risk registers)
Status: Enterprise white-label engine, immediately go-to-market. Revenue-ready.
Remediation Work Post-Claude Audit
Claude identified three minor gaps. All addressed:
1. FM Demo Enrichment **Gap:** Work order dates in demo data inconsistent (future orders mixed with past completions)
Remediation: Re-seeded FM work orders with realistic 6-month calendar (maintenance cycles match typical building schedules, emergency work spaced appropriately, completion logs accurate to dates)
Effort: 2 hours Outcome: Demo now tells a coherent facilities story (energy spikes in winter, maintenance cycles, budget tracking accuracy)
2. AI Cost Model Documentation **Gap:** Cost projections for Care Hub and Compliance Hub not published
Remediation: Created detailed cost model spreadsheet: - FM Control Hub: Sonnet summary at 2p per facility daily brief = £0.60/month/building - Care Hub: Haiku briefing <0.3p per patient summary, Sonnet <2p for complex cases = £1.50/month per active patient (100-patient clinic) - Compliance Hub: Haiku briefings <0.3p each, typically 1–3 per audit cycle = £15–45 per site per year
Effort: 3 hours Outcome: Acquirer can forecast AI operational cost and gross margin per vertical
3. Governance & Handover Pack **Gap:** Existing governance docs incomplete for buyer SOP
Remediation: Created five documents (180+ pages total): 1. Organizational & Roles (who does what, decision rights, escalation paths) 2. Data Architecture (schema, relationships, compliance mapping) 3. Deployment & Operations (installation checklist, environment config, backup strategy) 4. Clinical Governance (for Care Hub: decision logging, audit compliance, liability framework) 5. Customer Onboarding (white-label config, SOP templates, success metrics)
All documents are secret-free (no API keys, no credentials, no internal references). Buyer-ready.
Effort: 8 hours Outcome: Zero key-person risk. Buyer can take operational control on day one.
What Moves the Valuation Most
#1 — One Paying Pilot A single paying customer running any of the three hubs would move valuation from **£275k–£300k to £375k–£550k.**
Why? Risk transfer. An auditor values technology on feature completeness and code maturity. An acquirer values proof of market fit. One customer proving "people will pay for this" is worth 2 years of product development in market risk removal.
Current status: Two pilots in discussion (FM Control Hub with a mid-market facilities operator, Care Hub with a regional clinic network). Not yet signed.
#2 — FM Demo Fully Seeded The FM Control Hub demo now includes: - 6 months of work orders (preventive, reactive, emergency) - Energy readings across 50 building zones - Patrol data with photo logs - Supplier invoices and PO history - Budget tracking and variance analysis
A prospect can log in and see a real-looking facilities operation. No fake data. Realistic work cycles, seasonal patterns, budget pressures. This moves FM from "software that could work" to "software that works in the real world."
Valuation impact: +£25k–50k (proves operations understanding)
#3 — Governance Pack Removes Key-Person Risk Most early-stage software carries implicit key-person risk: "what happens if the founder disappears?"
With the governance pack, a buyer can: - Read exactly how the system works - Understand decision frameworks (clinical, operational, compliance) - Plan operational handover - Train their own team
Valuation impact: +£35k–75k (reduces buyer's integration cost and risk)
#4 — Care Hub AI is Genuine Live Differentiation Most "clinical AI" products are roadmap items. Care Hub's AI is live, verified, running on real-world-staged patient data. This alone puts it ahead of 95% of care tech in the market.
Valuation impact: +£40k–90k (genuine product advantage, not vapor)
Current Valuation
The honest central ask: £275,000–£300,000 for all three applications, code, documentation, governance pack, and white-label licensing.
This reflects: - Code maturity: Production-grade across all three - Clinical/regulatory compliance: Verified to audit standards - Demo richness: Seeded with 6+ months of realistic operational data - Documentation: 180+ pages, secret-free, handover-ready - Technical depth: 30+ FM modules, FHIR-certified Care Hub, 6-industry Compliance pack
Fire-sale scenario (urgent founders): £50,000–£80,000 - Buyer takes code "as-is" - Governance pack and demos included - Buyer handles all remediation and customer onboarding
Realistic indie/operator scenario: £200,000–£320,000 - Single operator acquiring one or two hubs for immediate revenue - Governance pack removes integration friction - Demo environments prove operational viability
Strategic buyer scenario (FM/care software): £450,000–£700,000 - Established software company buying to accelerate roadmap - Care Hub AI fills product gap immediately - FM Control Hub modules add to existing platform - Compliance Hub white-label engine enters 6 new verticals - Governance pack + existing customer base enables 90-day go-live
Next Steps
These are available for acquisition, pilot partnership, or white-label licensing.
If you operate in facilities management, care technology, or compliance software and want to: - Accelerate your roadmap by 6–12 months - Enter adjacent verticals (FM operators → Care, Care → Compliance) - Add clinical AI to your product without building from scratch - Deploy white-label compliance packs across your customer base
[Get in touch.](https://jonnyai.co.uk/brief)
The applications are live. The audits are done. The code is ready.