"Should we buy a product or build our own?" is usually asked at the wrong level. The better question: where does your business need control, and where is a template good enough? This framework walks through that decision the way we walk through it with clients — starting from operations, not from a feature grid.
Start with the workflow, not the feature list
List the work people actually perform from input to outcome. Mark three things: where data gets re-entered by hand, where approvals disappear into chat threads, and where the existing tool forces a workaround. Those seams usually matter more to the business than any number of missing screens.
A workflow that is genuinely distinctive — your pricing rules, your fulfilment sequence, the way your branches coordinate — is expensive to force into someone else's template. A workflow that every business in your industry shares is expensive to rebuild for no advantage.
“The question is not custom or off-the-shelf in isolation. It is where the business needs control.”
Three viable paths
Configure the existing product when its workflow is close enough and the gaps are tolerable. This is almost always the fastest path and should be the default assumption until proven otherwise.
Add an integration layer when the tools are individually useful but disconnected — orders in one system, stock in another, accounting re-entering both. Automation between tools often delivers most of the value of a rebuild at a fraction of the risk.
Build custom software when the operating model itself is the differentiator, when compliance or data ownership demands it, or when the recurring cost of workarounds has quietly grown larger than the build cost. If you reach this path, insist on owning the code and the data.
Signals you are on the wrong path
People maintaining private spreadsheets beside the official tool. Decisions waiting on a report nobody fully trusts. A vendor roadmap that keeps moving your priorities backwards. Any of these means the current arrangement is costing more than it looks on the invoice.
The honest first step
Map the workflow before choosing the software. If you cannot draw the flow from input to outcome on one page, no product decision made before that drawing exists will be a confident one.