When not to build

Most enquiries we receive describe a system the client wants built. Half the time, the right answer is not to build it.

The process is still changing. If the business is still working out how the workflow should run, software will not clarify it. Software makes a process repeatable, not correct. A spreadsheet or a manual checklist surfaces the gaps faster, and costs less to throw away when the process changes again.

The volume does not justify it. A monthly reconciliation that takes three hours is annoying. Automating it might take two weeks to build and a day every quarter to maintain. The payback period is years. Hire a part-time bookkeeper instead.

An off-the-shelf tool already does it. If Xero, HubSpot or Notion already solve the problem, the cost of building and maintaining a replacement will exceed the licence fee every year. Custom software makes sense when the process is specific enough that a general tool forces expensive workarounds. If the tool fits, use it.

It requires domain knowledge we do not have. Payroll, logistics routing, actuarial models — some domains have enough regulatory or algorithmic depth that building from scratch means rediscovering a decade of lessons already baked into existing software. If the problem is well-trodden, the vendors are probably ahead.

The real constraint is not software. If the bottleneck is that three people need to review every approval, software that routes the approvals faster will not remove the bottleneck. It will just surface it more clearly. Fix the process, then decide if software helps.

We charge for advice, including the advice not to build. Scope is where budgets are lost.

---

Sample post — this reflects Sageware’s current positioning on advisory and delivery, but is marked as illustrative content.