Taking a working prototype into production
When something built with Claude, ChatGPT, Cursor, Lovable, Replit or Bolt starts doing a real job, the engineering requirements change. Security, data, testing, deployment and clear technical ownership all need deliberate attention.
Prototype → Engineering → Production
A prototype can prove an idea and still be a long way from software a business can safely depend on. Moving it into production means finding the gaps, deciding what can stay and strengthening what matters.
What it proves
- A working version of the core idea
- A faster route to useful feedback
- Evidence about which features matter
- Early technical decisions to review
- A clearer view of what to build next
What we review and change
- Decide what to retain, refactor or rebuild
- Strengthen authentication and permissions
- Reshape data where future changes require it
- Add tests around critical user journeys
- Set up repeatable deployments and environments
What the business can rely on
- Monitoring and alerts that reach the right person
- Backups with a tested restore process
- Repeatable releases with a route back
- Accessible, tested critical user journeys
- Documentation and a prioritised plan
What we check before your prototype goes into production
We review each prototype across the areas below, then explain the findings in plain English and prioritise them by severity. The written assessment is yours to keep, whether or not you ask Branch to carry out the work.
Security
How secrets, sensitive data and user input are protected.
Authentication
How people sign in, recover access and stay signed in safely.
Permissions
Which data and actions each user can reach.
Database architecture
Whether the data model can support likely features without fragile workarounds.
Scalability
Where greater traffic, data volumes or background work could cause problems.
Testing
Which important journeys are checked automatically before a change goes live.
Monitoring
Whether errors and outages are recorded and reach the right person.
Backups
What is backed up, how often and whether it can be restored.
Integrations
How the software handles slow, unavailable or changed third-party services.
Code quality
How safely another developer can understand and change the code.
Accessibility
Whether important journeys work with a keyboard, screen reader or other assistive technology.
Deployment
How changes are released and how a failed release can be rolled back.
Maintainability
Where complexity could make future changes slower, riskier or more expensive.
Documentation
What another developer needs to run, release and change the software.
Compliance
How personal data is collected, stored, shared and deleted.
Ownership
Whether the business controls its source code, domains, hosting and third-party accounts.
Start with a production readiness review
Before recommending any rebuild, we work through the code, data model, deployment setup and the main user journeys. We look for security and reliability gaps, then separate urgent risks from improvements that can wait.
The result is a written assessment that explains what can stay, what needs attention and what would be safer to rebuild, with each finding prioritised by severity. It also sets out the likely cost and sequence of the work, so you can decide what to do next.
- Code and architecture review
- Security, data and compliance assessment
- Prioritised findings with indicative costs
- A phased plan to production, yours to keep
A production readiness review is £1,500 excluding VAT, and the findings are yours whether or not you go on to work with us.
Request a production-readiness review
Show us what you have built and we will assess its security, data, testing, deployment and maintainability, including what can stay and what needs attention.