Bespoke Software
Replacing spreadsheets with a business management system
The short answer
Replacing spreadsheets with a proper business management system typically costs £8,000 to £45,000 depending on how many processes move across, and takes six to sixteen weeks. It is worth doing once more than one person needs the same data at the same time, once an error in a sheet could cost real money, or once a single individual is the only person who understands how the file works. Below that threshold, a well-structured spreadsheet is a perfectly good tool.
Indicative price ranges
One process
£8,000 – £18,000
The single sheet that runs a critical process becomes a proper system with roles, validation, history and reporting. 6–10 weeks.
Multi-process system
£18,000 – £45,000
Several linked processes — jobs, customers, stock, scheduling, invoicing — in one database with dashboards and automation. 10–16 weeks.
Full operating platform
£45,000 – £90,000+
Adds portals, payments, mobile access for field teams and integrations with accounting or ERP.
These are indicative UK ranges based on projects we have actually delivered. Scope drives the number, not the sales conversation — we quote a fixed price after a scoping session, and we will tell you if an off-the-shelf tool is the cheaper answer.
Typical timescales
- 1–2 weeks
Spreadsheet archaeology
Working out what the formulas actually do — including the rules nobody documented.
- 2–3 weeks
Data model and design
Turning columns into proper entities and relationships.
- 4–10 weeks
Build
Delivered in increments, with the spreadsheet still running alongside.
- 2–4 weeks
Parallel running and cutover
Both systems live until the numbers match and the team trusts the new one.
Who this suits
- Businesses where two or more people need to edit the same data simultaneously.
- Processes where a single mistyped cell has financial or safety consequences.
- Teams that need history — who changed what, and when.
- Field or remote staff who need mobile access rather than a file on a laptop.
- Any process where one person leaving would create a genuine operational crisis.
Who it does not suit
- Genuine ad-hoc analysis and modelling. Spreadsheets remain the best tool for that and always will.
- Processes that change shape every month and have not settled.
- Very small teams where one person owns the process end to end and volume is low.
- Businesses unwilling to run both systems in parallel during cutover.
What actually drives the cost
Undocumented logic
Hidden columns, nested formulas and manual overrides all have to be understood before they can be rebuilt. This is the classic underestimate.
Data quality
Inconsistent formats, duplicates and free-text fields where a list should be. Cleaning is often a third of the migration effort.
Number of processes
Each additional sheet brought into scope adds screens, rules and testing.
Reporting expectations
People are attached to the exact layout of the sheet they have used for a decade. Replicating that view is often cheap and buys enormous goodwill.
The five signs the spreadsheet has become a risk
- The file has an owner and a fear. One person understands it, and everyone quietly worries about them going on holiday.
- Versions have appeared. "Final", "Final v2", "Final USE THIS ONE".
- You cannot see history. Nobody can say who changed a figure or why.
- Rules live in heads. The sheet is correct only if you know the six conventions nobody wrote down.
- It has outgrown itself. It is slow, it breaks on open, and someone maintains it as a part-time job.
At that point the spreadsheet is not saving money — it is a piece of unsupported, undocumented, single-maintainer software running a critical part of your business.
What you gain
Beyond the obvious multi-user access: validation that prevents bad data at entry, permissions so the wrong person cannot change the wrong number, a full audit trail, automation for the steps that were manual, reporting that does not require an export, and mobile access for people who are not at a desk.
How to migrate without pain
Do not rebuild everything at once. Take the process that hurts most, move it, prove it, then move the next one.
Run in parallel. Keep the spreadsheet alive until the numbers match for a full cycle. This is non-negotiable and it is how you earn the team's trust.
Recreate the familiar view. If the team has read the same layout for ten years, give them a screen that resembles it. It costs very little and dramatically reduces resistance.
Keep an export. People need to know they can still get their data into Excel for ad-hoc analysis. Say yes to that — it removes the last real objection.
Where this has worked
The Association of Independent Celebrants went from spreadsheet-and-inbox administration to a full association operating system covering membership, billing and an admin back office. Hartswood Holiday Park replaced manual booking administration with a direct booking engine and mobile-first admin, removing the dependency on both spreadsheets and commission-charging third parties.
Both had the same starting point: a business that had outgrown its files and knew it. If you recognise that, start with a conversation about which single process to move first.
Proof
Where we have done this
Common questions
How much does it cost to replace a spreadsheet with a system?
One critical process typically costs £8,000–£18,000. Moving several linked processes into one system usually falls between £18,000 and £45,000.
Can we not just use Airtable or Notion?
Sometimes, yes, and we will tell you when that is the right answer. They work well for structured lists with light workflow. They struggle with complex permissions, real relational integrity, high record volumes and customer-facing portals.
What happens to our historical data?
It is migrated, cleaned and kept. We usually import full history so reporting continues uninterrupted, and archive the original files unchanged as a reference.
How do we stop the team drifting back to spreadsheets?
Make the system faster than the sheet for the tasks people do daily, keep required fields minimal, and provide a proper export for genuine ad-hoc analysis. Drift happens when the system is slower than the thing it replaced.