When to build, when to buy, when to stop
Most build-vs-buy debates miss a third option: don’t do it at all. Here’s the test we actually use with clients.
Every year we are asked to help an owner decide whether to build custom software or buy an off-the-shelf tool. Before we answer, we always check a third option: not doing it.
A surprising number of build-vs-buy conversations end with “actually, let\u2019s not” \u2014 and the business is better for it.
The three tests
1. Is the problem real, today?
Not aspirational. Not “when we grow”. Is the problem costing you measurable time, money, or customer trust right now?
If not, the answer is: stop. Most software purchases are hopes disguised as plans.
2. Is the problem specific enough to solve?
A solvable problem looks like: “Our fulfilment team spends 90 minutes a day reconciling orders between Shopify and MYOB.”
An unsolvable one looks like: “We need to be more data-driven.”
If you can\u2019t describe the problem in a sentence with a specific number in it, pause. Go back to interviews and observation until you can.
3. Is your current process a documented, understood baseline?
If a motivated new hire could not understand your current process on day one, you are not ready to automate it. You are ready to document it.
We have watched too many businesses spend six figures on software that encoded a broken process in a more expensive form.
When to buy
Once you\u2019ve passed those three gates, buying almost always beats building, as long as:
- A credible vendor has a product that does 70% of what you want.
- That 70% covers the parts that actually matter.
- The vendor is still likely to exist in three years.
- You can migrate off them later without catastrophic rework.
Buying is rarely glamorous. It usually wins.
When to build
Build when:
- The problem is your core operational advantage and buying would give it to a competitor.
- You\u2019ve looked genuinely hard at existing tools and none of them clear the 70% bar.
- You can afford to maintain what you build, not just ship it.
- You understand the true cost of ownership — about 20% of build cost, every year, forever.
When to stop
Stop when:
- Nobody owns the outcome.
- The business case is “we should probably…”.
- The proposed system is upstream of a decision you haven\u2019t made yet.
- The ROI case relies on headcount reduction you have no appetite to enact.
Stopping is often the correct, cheapest answer. It almost never gets the applause it deserves.
If you\u2019d like a sober outside view on one of these conversations, that\u2019s exactly what Dynamis Advisory does for a week or two at a time.