Skip to content

ServiceWeb platforms & SaaS

Web platform and SaaS development, end to end

We design and build web platforms and SaaS products end to end: the interface, the APIs, the background processing and the integrations that connect them to the rest of your business. It’s for founders, scale-ups and enterprise teams who need a product that works for paying customers, and a Build Project starts from £12,000.

Software Stories is a UK software consultancy. Our case studies include analytics for construction programmes, a multi-channel platform for advertising agencies, ticketing and commerce for event hosts and audit-ready carbon reporting for UK SMEs, each designed and built end to end.

What we build

Most of the platforms we build are B2B or two-sided SaaS products, where several kinds of user work on the same data and the product has to connect to systems the client already runs.

  • SaaS products with sign-in, roles, billing and admin tools
  • Data-heavy dashboards and analytics, such as earned value, performance and forecasting views
  • Two-sided platforms with profiles, bookings, ticketing and payments
  • Integration layers and middleware that connect your product to accounting, telephony and automation tools
  • Internal tools that replace spreadsheet-based reporting
  • AI features inside the product, built through our AI agents and LLM service

Architecture that holds up as you grow

We keep the architecture as simple as the product allows and add structure only where it pays for itself. A single, well-organised application suits most early products. Separate services make sense when parts of the system change on someone else’s schedule, as Kaivo’s ad-platform integrations do, so one outage doesn’t take the rest of the product down with it.

Slow work stays out of the web request. File parsing, data syncs and calls to outside APIs run as background jobs with Celery and Redis, so a large upload returns straight away and the result is ready when the job finishes. That pattern is what lets Hallward process large Primavera schedules without freezing the browser or timing out at the gateway.

Integrations get the same care as the core product. We build them with retries, backoff, duplicate-event filtering and logging, because a third-party API will eventually fail, rate-limit you or send the same event twice.

Case studies: Hallward, Kaivo, Evntt and Carbon Giant

Hallward is a construction intelligence platform for project-controls teams on roads, rail and pipeline programmes. It parses Primavera P6 schedules into Earned Value, Critical Path, Time Chainage and milestone dashboards, on React, Django REST, PostgreSQL, Celery and Redis, with AWS Cognito for sign-in.

Kaivo gives advertising agencies one portal for client ad accounts across eight channels, with a Next.js front end, FastAPI microservices and an in-product AI assistant, deployed on Cloudflare and AWS.

Evntt is a two-sided platform for event hosts and guests: branded profiles with live-preview customisation, ticketing, analytics, merchandise and instant promotion through QR codes and share links. It runs on React, TypeScript, Node.js and PostgreSQL, with Stripe billing and AWS hosting.

Carbon Giant’s Carbon-Passport turns invoices and Xero, QuickBooks and Sage data into SECR-ready Scope 1–3 emissions reports, with OCR through AWS Textract and a single accounting integration through Apideck.

Payments, sign-in and integrations

Much of a SaaS product’s value depends on what it connects to. These are integrations we’ve built into products we’ve shipped.

  • Payments and billing: Stripe and Affirm
  • Sign-in: AWS Cognito, NextAuth.js and Firebase Auth
  • Accounting data: Xero, QuickBooks and Sage through Apideck
  • Messaging and automation: RingCentral SMS events forwarded to Zapier, and email through Nodemailer
  • Health and research data: Terra for wearables, and REDCap Cloud
  • Venue systems: POS and reservation platforms, in CityOS

The stack we use

We pick tools that fit the product and the team who’ll own it after handover. These are the ones we use most for web platforms.

  • Front end: React, Next.js, TypeScript and Tailwind CSS
  • Back end: Python (Django REST, FastAPI, Flask) and Node.js
  • Data: PostgreSQL, MongoDB, Redis and SQLite
  • Background jobs: Celery and Redis, plus webhook-driven async processing
  • Desktop, when the job needs it: Electron, as in Cosmic IDE
  • Hosting: AWS and Cloudflare

What you get

  1. A written scope and architecture covering users, data, integrations and hosting
  2. A production web application, deployed in your own cloud account
  3. APIs and integrations with retries, logging and monitoring
  4. Sign-in, roles and billing set up for your business model
  5. Regular demos and a staging environment you can test on throughout the build
  6. Source code, documentation and a handover session for your team

How it works

  1. 1.Discovery and scope

    Before the build

    An NDA if you need one, then a working session on your users, your data and the systems the product must connect to. You get a written scope, architecture and price before any work starts.

  2. 2.Design the core

    Early in the build

    We agree the data model, the main user journeys and the riskiest integration first, because those are the decisions that are hardest to change later.

  3. 3.Build

    Agreed milestones

    Features delivered in working increments, with regular demos on a staging environment so you can test as we go.

  4. 4.Launch

    Before go-live

    Load and failure testing on the slow paths and integrations, monitoring and alerts in place, then release to real users.

  5. 5.Hand over

    At launch

    Documentation and a handover session, so your team can run and extend the product. If you’d like us to keep building with you, an ongoing retainer is priced on request.

Proof

“They acted as true collaborators who cared about the success of the product as much as I did.”

ViralzyClient

Viralzy

Client · 5-star review on Upwork

From the very beginning, they approached the project with a deep sense of responsibility. They took the time to fully understand the vision behind Viralzy, asking thoughtful questions and proactively offering ideas to make the platform more robust and scalable. Rather than simply executing tasks, they acted as true collaborators who cared about the success of the product as much as I did.

Their work ethic has also stood out. They handle challenges calmly, troubleshoot effectively, and never cut corners. Even when timelines were tight, they prioritized quality and made sure that everything they delivered was stable, clean, and well thought out.

Their commitment, reliability, and constant availability made them an essential part of the project’s success, and I would recommend working with them to anyone who wants a team that genuinely cares about bringing a vision to life.

Price

From£12,000

For a scoped first release, from architecture to launch, with regular demos and a full handover. Every engagement starts with a free one-hour call, and you get a written scope with a fixed or capped price before any work starts.

FAQ

Our Build Projects start from £12,000 for a scoped first release. The price depends on the number of user types, the integrations involved and how much data processing the product does. You get a written scope with a fixed or capped price before any work starts.

Usually not at the start. A single, well-structured application is faster to build and cheaper to run. We split out services where there’s a clear reason, such as integrations that change on a third party’s schedule or heavy processing that needs to scale on its own.

Usually React or Next.js with TypeScript on the front end, Python (Django or FastAPI) or Node.js on the back end, and PostgreSQL for data. If your team already works in one of these, we’ll fit to it, so the product is easy for them to own.

Usually, yes, if they offer an API, webhooks or a data export. We’ve built integrations with accounting systems, telephony, payment providers, wearables platforms and research databases. We check the specific systems in discovery before committing to a scope.

Yes. Several of the platforms we’ve built include AI inside the product, such as Kaivo’s “Kai” assistant and Hallward’s AI-assisted activity categorisation. Our AI agents and LLM service explains how we test and control those features.

Tell us what you’re building.

Book a free one-hour call, or send a short brief and we’ll reply within two working days.