Start with the cost you cannot see
A licence fee is easy to compare. The expensive part of a near-fit product is the work it generates: the export into a spreadsheet, the second system that holds what the first cannot, the person who reconciles them on a Friday. That cost never appears on an invoice, which is exactly why it survives for years.
Before comparing products, write down the manual steps your current setup requires each week and roughly how long they take. That number is the real benchmark any new option has to beat.
Four questions that usually settle it
- Is the process a genuine differentiator, or is it the same as everyone else in your sector? Differentiators are worth building; commodity processes are worth buying.
- Would you have to change how you work to suit the product, and is that change an improvement or a compromise?
- How much of the licence are you paying for capability you will never use?
- If the product disappeared tomorrow, how much of your operation would stop?
Buy when the process is standard
Accounting, payroll, email, storage, most CRM functions: these are solved problems with mature products and long support lifetimes. Building your own version of them is almost always a way of spending money to end up slightly behind.
The same applies to the first version of a process you have not run yet. If you are unsure how the work should flow, a configurable product is a cheap way to find out before committing to a build.
Build when the fit is the whole point
Building becomes sensible when the way you do something is the reason customers choose you, when the workaround has become a job, or when the connection between two systems matters more than either system does. It is also sensible when the product you need does exist but only as a module inside a platform three times the size of your requirement.
A useful test: if the software vendor made you change one thing about how you work, would your customers notice?
It is rarely all one or the other
Most of the work I do sits between the two: keep the products that are doing their job, and build the connecting layer, the portal or the internal tool that stops people acting as the integration. That usually costs less than replacing a platform and removes more manual effort than either option alone.
What to do next
Write down the process, mark every step where a human moves information rather than making a decision, and total the time. If that total is uncomfortable, the decision is no longer about licence fees.
