Guides

Guide · interactive checklist

Build or buy: a small-company checklist.

Most of the time, the right answer is to buy. We are a software studio, and we still tell people that. Answer ten questions about the software you need and see which way your answers point.

Before the checklist

Start with the job, not the tool.

Before you look at a single product, write down what the software has to do during one real week of work. Who uses it, what information goes in, what has to come out, and where that information comes from today.

Then test every option against that week, not against a feature list. Products are good at storing information. The question is whether they fit the way the work actually arrives.

The checklist

Ten questions. Pick the answer that is closest to true.

  1. Is there a product built for your industry that does the core job?

    Why it matters: A product made for your industry already reflects how companies like yours work. That is years of design you do not have to pay for.

  2. Is the way you do this work standard, or part of what makes you good?

    Why it matters: If your process is the same as everyone else's, a product fits it. If your process is your advantage, bending it to fit a product gives that advantage away.

  3. Would a product cover most of what you need today?

    Why it matters: A product that covers most of the job is usually worth it. One your team fights every day costs more in lost time than it saves.

  4. Are the missing features on the product's published roadmap?

    Why it matters: If the features you need are already planned, waiting a few months can be far cheaper than building them yourself.

  5. Does the product work with the information people actually give you?

    Why it matters: Most products are good at storing information. The question is whether they help when the information you get is incomplete or arrives in a different shape.

  6. Does it need to connect to systems you already use?

    Why it matters: Ready-made connectors are cheap and reliable. When there are none, someone ends up retyping data between systems, or you pay for a custom connection anyway.

  7. Will the cost stay reasonable as your team grows?

    Why it matters: Subscriptions priced per person can be cheap for five people and expensive for fifty. Look at the cost over three years, not the first month.

  8. If the vendor raised prices, changed the product, or shut down, could you cope?

    Why it matters: The more central a tool is to how you make money, the more it matters who controls it and whether you can get your data out.

  9. Do you need it working within weeks?

    Why it matters: A product can be running this week. Custom software takes time to get right, so it fits problems you plan to live with for years.

  10. Has your team already tried off-the-shelf tools for this and given up?

    Why it matters: If you have not tried a product yet, try one first. If you have tried several and they all failed in the same way, that pattern is your answer.

0 of 10 answered

Answer every question to see your result.

When I said buy

The best advice I gave a client was not to build.

When I worked at a managed IT provider, one of our clients was planning a custom build. The quote they received was very high. Meanwhile, an off-the-shelf platform cost far less, covered most of what they needed right away, and had the rest of the features they wanted on its published roadmap.

I told them to buy it and wait for the updates. About six months later, they were very happy with that decision.

When we built

Every product stored the data. None of them fit the work.

When I worked desktop support at a call center, we looked at several products for tracking desks, network drops, switch ports, phones, and PCs. They all offered the same thing: a database to store the information.

The real problem was that people often did not say which desk they were at when they put in a ticket. Someone had to walk out to the floor, find the desk, and walk back. Our own floor map showed supervisor names on the desks, so we could simply ask who their supervisor was and find the desk from that. No product did that, because no product knew how our floor worked.

Read the full floor map story

The middle path

Often the answer is both: buy the core, build the gap.

What they kept

The Green Dog Consulting kept Microsoft 365 for documents, sign-in, and the tools the team already knew.

What we built

A client operations workspace for the part Microsoft 365 does not do: one home for each client's work.

What changed

Katie no longer spends time looking for information, and she collaborates better with her team.

Read the full story

Still not sure?

Send us your answers and a sentence about the job. We will tell you which way we would go, even if the answer is buy.

Start a project