Buy or Build: A Straightforward Way to Decide
Off-the-shelf software is cheaper until the workarounds start. A practical test for which side of the line a given process sits on.
Article
The buy-or-build question usually gets answered by temperament rather than analysis. People who have been burned by a long build buy everything. People who have been trapped by a rigid tool build everything. Both positions produce bad decisions about half the time. A more useful question: is this process a source of advantage, or is it just something the business has to do? Buy the ordinary Payroll, accounting, email, storage, calendars. These are solved problems where thousands of companies want nearly identical behaviour. A vendor spreading development across all of them will always outbuild an in-house effort, and the process is not where the business wins anyway. Buying here is not a compromise. It is the correct answer, and adapting the way you work to match a good tool is usually cheaper than the alternative. Build what makes you different The processes worth building are the ones that would look strange to a competitor: an approval flow that reflects a particular regulatory position, pricing logic specific to how the business actually sells, a client-facing view nobody else offers. These are exactly the processes where off-the-shelf software fits worst, because the vendor built for the average and the advantage is in the deviation. Bending the business to fit the tool here means giving up the thing that made it worth doing. The workaround test For anything in between, count the workarounds. Every exported spreadsheet, every field being used for something other than its label, every process step that exists only to move data between two tools: each is a small ongoing payment. Workarounds do not announce themselves. They accumulate, and the team stops noticing them because working around the tool has become the job. Once a process has three or four, the tool is no longer saving what its price suggests. The middle option Buy-or-build is a false binary more often than not. Most businesses should buy the platforms and build the connective tissue: the integration that moves a signed proposal into the accounting system, the internal view that combines three tools into the screen someone actually needs. This is usually where the highest return sits. It is a fraction of the cost of building a platform, it removes real daily friction, and it does not commit anyone to maintaining software they did not want to own.