A market problem rarely gets solved by one component. The framework's Buy, Build or Partner box asks a specific question, which is how to complete the solution where the current offering has gaps, and it treats building as one option among three rather than the default.
The three options and what each costs
Build. Full control of the roadmap, the data and the experience. Slowest, and the cost does not stop at delivery. Every component built is a component maintained, documented, supported and migrated for as long as the product lives.
Buy. Acquire the capability, whether that is a company, a product, a licence or a data set. Fastest route to a capability that already exists elsewhere. Costs capital and integration effort, and the integration is usually underestimated by a wide margin.
Partner. Someone else's capability, reached through an agreement. Low upfront cost and fast to arrange. In exchange the partner's priorities, pricing and viability become yours, and unwinding it later costs more each year.
The test that decides
The question is not what is cheapest today. It is whether the gap sits on the distinctive competency.
If the component is the thing customers cannot get elsewhere, build it. Handing that to a partner gives away the reason the product wins.
If the component is necessary but undifferentiated, buying or partnering is usually correct. Every week spent building a commodity is a week not spent on the part nobody else can build.
If the component is differentiating but the company has no route to building it in a useful timeframe, buying is what the option exists for.
Questions worth asking before committing
How long until the capability is in customers' hands under each option.
What the total cost of ownership looks like over three years rather than the first year.
What happens if the partner is acquired, changes their pricing, or closes.
Whether the option can be reversed, and what it would cost to reverse it after a hundred customers depend on it.
Whether the buyer would notice or care that this part came from somewhere else.
Revisiting the decision
The right answer changes as a market matures. Capabilities that were differentiating become commodity, at which point a built component is a maintenance burden and switching to a bought one releases the team. Partnerships made when a capability was scarce become expensive once three vendors offer it. The box is reviewed on the same cycle as the competitive landscape, not once at the start.