There is no responsible single price for bespoke software because two products with the same number of screens can carry completely different integration, data and operational risk. A useful budget begins with the decisions the software must support and the systems it must connect.
Typical budget ranges
A focused discovery phase commonly starts from a few thousand pounds. A production-ready MVP for one clear workflow often begins around £18,000 to £35,000. Multi-role platforms with migration and several integrations are usually larger programmes delivered in stages.
- Discovery and prototype: typically £2,500–£8,000
- Focused MVP: typically £18,000–£50,000
- Integrated business platform: commonly £50,000 and above
- Ongoing product development: planned monthly capacity
What actually changes the cost
The largest drivers are rarely visual design alone. Permissions, data migration, external APIs, offline behaviour, regulatory controls and unusual reporting requirements add engineering and testing effort.
Unclear ownership also increases cost. When no stakeholder can resolve process questions quickly, a development team either waits or builds assumptions that later need to be undone.
Why discovery is not wasted budget
Discovery creates a shared model of users, workflows, constraints and success measures. It can reveal that a smaller integration or process change will solve the problem, or that a planned feature is not needed in the first release.
A short paid discovery is usually less expensive than committing to a fixed feature list based on incomplete assumptions.
How to compare proposals
Compare what is included around the code: product management, design, testing, deployment, monitoring, documentation and post-launch support. Ask who owns the repositories and cloud accounts, how changes are approved and how progress will be demonstrated.
- Is the scope linked to business outcomes?
- Are integrations and migration explicitly covered?
- Who owns infrastructure and source code?
- What happens when assumptions change?
- How are security, accessibility and testing handled?
A lower-risk way to start
Begin with the smallest release that removes one meaningful bottleneck or proves one commercial assumption. Launch it to a controlled user group, collect evidence and use that evidence to fund the next stage.