What It Actually Costs to Build a Custom Business System
MRP, CMS, LMS or an internal dashboard — the same six cost drivers decide the bill. Here is what actually moves the number, with an effort table.
What It Actually Costs to Build a Custom Business System
Ask a software vendor what their product costs and you get a price list. Ask what a custom business system costs — an MRP, a CMS, an LMS, or an internal dashboard built around how your business actually runs — and the honest answer is "it depends," followed by a contact form. That's not evasion, it's genuinely true. But it's also not useful. Here are the six things that actually move the number, in the order they usually bite.
1. User count and roles
Ten users on one role costs far less than fifty users split across five roles with different permissions. It's not the seat count that costs money — it's the branching logic: what an ops manager can edit that a warehouse operator can't, what a finance role can see that a sales role can't. Each additional role is a new set of test cases, not just a checkbox.
2. Integrations
A system that stands alone is cheap. A system that has to talk to your accounting software, your existing CRM, a supplier's EDI feed, or a legacy database with no documented API is not. Integrations are consistently the most underestimated line item in a quote, because the complexity lives in the other system's quirks, not yours.
3. Data migration
If you're replacing spreadsheets or an old system, someone has to get the existing data — years of it, inconsistently entered — into the new structure without losing anything. This is covered in more depth in our data migration guide, but as a cost driver: budget for it explicitly, because it rarely fits inside the build estimate by accident.
4. Reporting depth
A handful of fixed reports is a small feature. A flexible report builder where non-technical users can filter, group, and export arbitrary combinations of data is a product in itself. Know which one you actually need before you ask for a quote — most businesses overspecify this and pay for flexibility they never use.
5. Offline and connectivity requirements
If the system needs to keep working on a warehouse floor with patchy Wi-Fi, or sync data collected in the field once a connection returns, that's a meaningfully different architecture than a system that assumes constant connectivity. It's worth deciding early, because retrofitting offline support later is expensive.
6. Compliance and audit trail
Regulated data — health records, financial transactions, anything GDPR-sensitive — adds logging, access control, and retention requirements that a simple internal tool doesn't need. See our note on GDPR for custom software for what this actually involves.
Rough effort by project size
| Project shape | Typical effort | What's included |
|---|---|---|
| Single-purpose internal tool | 4–8 weeks | One or two roles, no external integrations, fixed reports |
| Departmental system (e.g. stock or scheduling) | 8–16 weeks | Several roles, one or two integrations, basic reporting, data migration |
| Company-wide platform (e.g. MRP or LMS) | 4–9 months | Multiple roles, several integrations, flexible reporting, audit trail, phased rollout |
None of this replaces an actual quote — your project will land somewhere specific on that table depending on which of the six drivers apply. But if an agency hands you a number without asking about roles, integrations, and existing data, ask again. The estimate isn't real yet.
Sheraian scopes every custom system project against these six drivers before quoting, so the number you get reflects your actual requirements rather than a template. If you're at the point of comparing options, our build vs buy framework is a useful next stop.