// every new enquiry, scored and bookedexport async functiononLead(lead) {const score = await ai.qualify(lead)if (score > 0.8) await cal.book(lead)await crm.save({ ...lead, score })}
Deployed to production
Growing businesses often run on spreadsheets, copied data and tools that never quite talk to each other. We map how your work flows, build software around it, and use AI to test faster with a person reviewing each release.
What we build.
Six areas of work, often combined into one project, all planned and delivered by the same team.
SaaS products
Subscription software, built to grow.
We take your SaaS idea from first screens to a working product with accounts, billing and an admin area. You get a codebase built to grow with your customers.
Replace the spreadsheet holding everything together.
We build custom tools for quoting, bookings, stock, jobs or approvals, shaped around how your team already works. That means less copying between systems and fewer mistakes.
We connect your systems through their APIs so orders, contacts and stock move between them on their own. Where it fits, we use Zapier, Make or n8n instead of custom code.
We set up Stripe for one-off payments, subscriptions, invoices and customer portals. Webhooks keep your records in step with every payment, refund and renewal.
After launch we monitor your software, apply updates and fix issues as they come up. New features are planned with you and released in small, tested steps.
The parts that make the biggest difference, and how we handle each one.
Planned before a line of code.
Every project starts with a short discovery. We map your users, data and workflows, then agree a clear scope so you know what is being built and why.
Build plan
Working
Map users and workflowsDone
Agree scope and prioritiesDone
Design key screens in FigmaDone
Build core featuresLive
Test and review releaseNext
Faster builds, checked by people.
We use AI to draft code, write tests and spot likely bugs earlier in the build. A developer reviews every change before it reaches your users.
AI review summary
Example
Pull request ready for review. Adds invoice export to the admin area, with tests for empty and partial data. One query may be slow on large accounts. Suggest adding an index before merging.
Checked by a person before it goes out
Built on tools that last.
We choose proven, well-supported technology so your software is easier to maintain, extend and hand over later. Every project is documented as we go, so nothing lives only in one person's head.
How we would approach it. Illustrations, not client results.
Example project · SaaS products
A training provider turning courses into a subscription
A training company sells courses one at a time. It wants a members platform with monthly plans, team accounts for employers and progress tracking, without building something too large to launch.
Agree the smallest product worth charging for
Design sign-up, onboarding and the learner dashboard
A growing e-commerce brand checks Shopify, Meta Ads, Google Ads and Xero separately every Monday. Each report tells a different story, and nobody is sure which products make money after ad spend.
A trades business quoting from a shared spreadsheet
A heating and plumbing firm prices jobs in one large spreadsheet, books engineers by phone and chases sign-off on WhatsApp. Quotes go out late, and details get lost between office and site.
Every project is different, so it is quoted after a short call, with a fixed price agreed before we start. Ongoing support and maintenance runs on a fixed monthly fee.
How long does a build take?
A focused tool or dashboard usually takes a few weeks, and larger SaaS products take longer. We agree a date before we start and share working versions along the way.
Can you work with our existing software?
Yes. We can review an existing codebase, fix issues and add features, or connect it to new tools through its API. We start with a short audit so we understand what is already there.
How do you use AI in development?
We use AI to draft code, write tests and review changes for likely bugs. A developer checks everything before it goes live, and we agree with you up front which data AI tools can see.
Which technology do you use?
We usually build web software with Next.js, React and TypeScript, with PostgreSQL for data and Stripe for payments. We pick well-supported tools so your software is easier to maintain later.
What happens after launch?
We can stay on for a fixed monthly fee to monitor the software, apply updates and build new features. If you prefer to take it in-house, we hand over the code with full documentation.