The decision usually arrives framed as build or buy, and that framing eliminates the two options that most often turn out to be correct. Integrating what you already own and consolidating what has accumulated are both cheaper than either alternative and both get skipped because neither involves acquiring anything new.

The Four Options

What you are actually choosing between

The Question That Decides It

Before evaluating any option, establish how distinctive the underlying process actually is.

Most processes that feel unique are conventional processes with accumulated local variation. Intake, approvals, scheduling, reporting, case management: the shape is common across thousands of organizations even when the details differ. Where that is true, buying is almost always correct, and the variations that feel essential are usually habits rather than requirements.

Where a process genuinely is distinctive, and where it is central to how the organization delivers value, building becomes defensible. That combination is rarer than it feels from the inside.

Distinctive and central justifies building. Distinctive but peripheral is usually a habit worth abandoning.

Working Through the Decision

Question
What are you already paying for that could do this?

Most organizations run productivity and cloud platforms whose capabilities are used at a fraction of what is licensed. Before evaluating a new purchase, establish what the existing stack covers. This question alone resolves a meaningful share of platform decisions.

Question
Is the problem the tool, or the process running through it?

Replacing a platform to fix an unclear process produces a new platform and the same problem, plus a migration. If nobody can describe the current process cleanly, that is the work to do first.

Question
How many systems does this need to talk to?

Integration cost scales with connection count and is consistently underestimated. A product that is a strong fit in isolation and connects to nothing you run is often a worse outcome than a weaker product that connects cleanly.

Question
What does the second year cost?

First-year pricing is negotiated. Renewal pricing, per-seat growth, storage overages, and the modules that turn out to be separately licensed are where the real number lives. Ask for year-two and year-three figures in writing before deciding.

Question
How do you get your data out?

Ask the question during evaluation, when you have leverage, rather than during an exit, when you have none. A vendor who cannot describe a clean export path has told you something about the relationship.

Question
Who owns this after go-live?

Every platform needs an owner responsible for configuration, access, and the decisions that accumulate after launch. A platform without a named owner drifts, and the drift shows up two years later as a system nobody quite understands.

The Cost That Gets Missed

Platform decisions are usually evaluated on license cost and implementation effort. The larger number is what the choice costs in flexibility.

Every platform encodes assumptions about how work should be done. Adopting one means adopting those assumptions, and changing them later ranges from expensive to impossible. That is fine when the assumptions match how you operate. It is a slow problem when they do not, because the organization gradually reshapes itself around the tool rather than the other way around.

Worth checking before signing

Ask the vendor how customers typically handle the two or three parts of your process that do not match their model. If the answer is that customers change their process, decide now whether you are willing to.

A Note on Building

Building is not primarily a development decision. It is a commitment to maintain something indefinitely: security updates, dependency upgrades, changes as requirements shift, and the knowledge of how it works staying inside the organization.

Organizations that build successfully treat that ongoing commitment as part of the cost from the beginning. Organizations that regret building usually scoped the initial delivery and not the decade after it.

Where to start

Before scoping any platform decision, inventory what the organization already licenses and what portion of it is actually in use. That inventory changes the shape of the decision more often than it does not.