V360HQ · Discovery & Recommendation

JPR Combustions — New Website & Integration Strategy

Why the existing site is a liability, and how to build its replacement as the front door to one integrated system — not a brochure.

Prepared by V360HQ For JPR Combustions Ltd Date 13 June 2026 Status Research & recommendation — pre-build

The one-line thesis. Don't commission a new brochure website. Build the integration spine once — a small secure backend that talks to BigChange and Xero — and let both the public website and the planned customer app be two faces of it. The website becomes a lead engine that writes jobs straight into BigChange, so it starts reducing inbound phone calls from day one, before the app ships.

1Executive summary

JPR's current website is roughly eight years out of date, is publicly displaying a database error where its testimonials should be, and works against a firm that is about to launch a rebrand and a customer app. That is the problem. The opportunity is larger than a redesign: the same engineering work that makes the customer app possible (a secure backend wired into BigChange + Xero) can power a modern website at the same time. Build it once, deploy it twice.

2Current site audit — the evidence

We reviewed the live site at jprcombustions.co.uk and mapped every page. It is a static, multi-page HTML template built in 2018 by an SEO agency (i-promote.eu). The findings below are not cosmetic — several are actively damaging credibility right now.

FindingDetailSeverity
Broken testimonials pageThe /testimonials.html page is throwing a live, public database error: "Fatal error: Uncaught mysqli_sql_exception: Access denied for user 'icontrol'@'localhost'". Customers see a raw SQL crash where reviews should be.Critical
Stale contentThe homepage still announces an office move "as from the 1st January 2019." Copyright reads 2018. Nobody has maintained the site in ~7 years.High
Dead / wrong social linksA Google Plus icon (the platform shut down in 2019) and a YouTube link that points to Worcester Bosch's channel — not JPR's.Medium
Mobile-hostileViewport set to user-scalable=no, maximum-scale=1 — pinch-zoom disabled, an accessibility and usability fail on the device most local customers use.Medium
No real lead captureEvery call-to-action is "Contact Us" or a phone number. Nothing books, quotes, or qualifies. The site generates phone calls — the exact opposite of the stated goal.High
SEO-template DNAKeyword-stuffed titles ("Gas Fire Servicing Wooten Bridge, Power Flushing Isle of Wight"). Generic OG/social image hosted on the agency's domain. Looks dated to both users and modern search.Medium
One genuine asset to keepA "Boiler Insight Tool" exists (a property/boiler questionnaire). The idea is right — it should be rebuilt as a modern instant-quote + finance lead magnet (Section 3).Salvage
Immediate, regardless of the bigger project

The testimonials page should be taken down or fixed this week — a public SQL error is the single worst thing on the site and costs nothing to remove. Flag to James as a quick win even before the rebuild is approved.

3"All possible integrations" — the ecosystem mapped

James asked for a future-proof site that uses every integration for best output. Here is the full surface, grouped by what it does for the business. Most of these JPR is already paying for — the value is in wiring them into one coherent customer experience.

Operations & the call-reduction engine (the core)

IntegrationWhat it deliversStatus
BigChange — Jobs APIWebsite enquiry / fault report writes a job straight into BigChange. No phone call, no re-keying. This is the mechanism that reduces inbound calls.On their plan
BigChange — Customer PortalNative self-service booking, live job tracking with engineer ETA, documents-on-demand. Switch-on-able today at near-zero cost — link or embed it as the logged-in layer.Available now
BigChange — Documents APIPull completed job PDFs / certificates by property. Confirmed retrievable via the Documents resource (Section 5). Powers tenant/landlord certificate access.Confirmed
BigChange — WebhooksEvent push (job updated, invoice created, etc.) — newly available in 2026 and expanding. Lets the site react in near-real-time instead of polling.New 2026

Money & recurring revenue

IntegrationWhat it deliversStatus
Xero APIInvoice + payment status pulled into the customer's account view. BigChange ↔ Xero already sync natively — we read one source of truth, never double-enter.On their plan
Online paymentsStripe / GoCardless to pay an invoice or deposit online. Removes "pop a cheque in the post."To add
Service-plan subscriptionsAnnual boiler-service plan billed monthly (recurring revenue + guaranteed re-booking + a captive list for gas-safety renewals). A genuine new revenue line, not just a website feature.Opportunity
Worcester / boiler financeFinance application flow for new-boiler installs (JPR already offers this — make it an online journey, not a phone call).To wire

Trust, conversion & reach

IntegrationWhat it deliversStatus
Google Business Profile reviewsLive star rating + recent reviews — replaces the broken DB testimonials page with real, self-updating social proof.To add
Gas Safe live verification"Verify our registration" badge linking to the official Gas Safe record — instant trust for a safety-critical trade.To add
Instant boiler quote toolRebuild the existing Boiler Insight Tool as a modern fixed-price quote + finance calculator (the BOXT / Heatable model). The single biggest conversion lever for a boiler installer.Upgrade
AI chat / triage assistantDeflects routine "is this urgent / can you come out / how much" questions and books or escalates — directly attacks call volume.Phase 2+
Reminders (email/SMS)Automated annual service + gas-safety renewal reminders. Drives re-bookings and underpins the landlord-compliance angle.To add
GA4 + Local SEO schemaProper analytics + LocalBusiness/HVAC structured data + Core Web Vitals. Modern search visibility the current template can't reach.Baseline

4Recommended architecture — build once, deploy twice

The key decision: the new website and the planned customer app should not be two separate projects. They share one backend.

The spine: a Backend-for-Frontend (BFF). A small, secure server holds the BigChange and Xero credentials, refreshes tokens, caches data, listens for webhooks, and exposes one clean, safe API. Secrets never live in a browser or a phone app. This backend is the real engineering — and it is identical whether the front-end is a website, a web app, or a native app.

Two front-ends off one spine:

Recommended stack

  • Marketing site: static/edge-rendered on Cloudflare Pages — fast, cheap, excellent Core Web Vitals, strong SEO.
  • Backend (BFF): Cloudflare Workers / lightweight Node service — holds creds, caches, handles webhooks + polling fallback.
  • Content: headless CMS so JPR edit copy/prices without a developer.
  • Portal: PWA on the same domain, same backend.

Why this is "future-proof"

  • Every new integration is added once, at the spine, and instantly serves both faces.
  • Front-end can be re-skinned for the rebrand without touching the plumbing.
  • Webhooks-first where supported, polling as fallback — resilient to BigChange's roadmap.
  • Scales from 5 engineers / 400 jobs a month to many multiples without re-architecture.

5Three things to confirm + two findings in JPR's favour

We pressure-tested the technical feasibility against live sources. Two items work firmly in JPR's favour; one needs careful framing.

Finding 1 — in JPR's favour

BigChange now has webhooks. BigChange has shipped webhook events through 2026 (Users, Stock, Invoices, Job line items, Quotes, Purchase Orders — added Mar–May 2026) and is expanding them. We architect event-driven first, polling as fallback — better UX (a completed job surfaces near-instantly) and lighter API load.

Finding 2 — in JPR's favour

Certificate retrieval is confirmed. BigChange exposes a Documents resource that returns completed job PDFs / jobsheets, and Jobs carry their documents too. The "can we even get the CP12 out?" risk — the one that could have killed the compliance idea — is answered. (Still worth confirming the exact endpoint + licence scope with BigChange, but the mechanism exists.)

Finding 3 — needs careful framing

The compliance angle must be framed on current law. Section 21 was abolished on 1 May 2026 under the Renters' Rights Act, so any "keeps your Section 21 valid" angle no longer applies. The durable driver is stronger: the 28-day requirement to give the tenant the gas safety record under the Gas Safety (Installation & Use) Regulations 1998 — unchanged, and criminal law — while the Renters' Rights Act is raising the overall compliance bar. The compliance product is more relevant now, framed on the law that still applies.

The three confirmations to put to James / BigChange

  1. Certificate endpoint & scope. Confirm completed CP12 / LGSR PDFs can be pulled by job/property on JPR's licence scopes (the Documents resource indicates yes — verify the exact call).
  2. Webhook coverage. Confirm which events JPR's plan can subscribe to (especially "job completed / document generated"), so we set the event-vs-poll boundary.
  3. API access on the plan. Confirm the "Essentials" REST tier is active and the "Administrate API Keys" permission can be granted to generate credentials. (400 jobs/month is trivial against the 100k calls/month ceiling.)
Reality check on the "differentiator"

Landlord-compliance tools already exist (ComplyGuy, August, CertNudge). The honest framing: those are all landlord-side, manual data-entry tools — someone has to upload the cert. JPR's edge is that the certificate is delivered automatically because JPR did the work and it's already in BigChange — zero landlord admin, timestamped proof. That's a distribution moat (you have to be the contractor to do it), and it's real — but it's worth being precise about, because it shapes how the offer is priced and positioned.

6Phased build plan

Phase 1 New marketing website + the spine

Ship a fast, credible, modern site that already reduces calls — before the app exists.

Phase 2 Customer portal (PWA) on the shared spine

Phase 3 Landlord/tenant compliance layer + native (if justified)

7Recommendation

Green-light the build. Feasibility is confirmed: the integrations are real, mostly already paid for, and certificate retrieval — the scariest unknown — is answered. The backend-then-website-then-portal sequencing is the right shape.

Lead with the website, not the app. Phase 1 delivers visible value fast (a site that stops embarrassing JPR and starts capturing leads + cutting calls) while quietly building the spine the app needs. It de-risks everything downstream and delivers an early, visible win to build on.

The remaining risk is positioning, not engineering. Get the legal framing right (28-day Gas Safety Regs duty, not Section 21), be precise about the distribution moat, and confirm the three BigChange items. From there, the rebrand is applied to the design, the package is priced and the commercial terms agreed, and the build proceeds.

Immediate next steps

  1. Take down / fix the broken testimonials page this week — a quick win worth doing immediately.
  2. Put the three confirmations to James → BigChange (certificate endpoint, webhook coverage, API access on plan).
  3. Once Phase 1 is approved: lock scope, apply the rebrand direction to the design, and begin the BFF skeleton and site build.