Watch how a software stack actually grows. Not from some master plan, but one reasonable yes at a time. A tool here to fix a headache, a subscription there because a competitor mentioned it, a free trial that quietly started billing in month two. A year later the business is paying for a pile of apps, and nobody can say for certain what half of them do. The average small business now runs around 87 of them.
So before you add the next one, it is worth slowing down. We build software for a living, and here is the honest thing up front: most of the time, you should buy, not build. Accounting, email, payments, scheduling, these are solved problems. Writing your own version of QuickBooks is a great way to lose a year and still end up with something worse. The catch sits on the other side of that 87 number. A quarter to a third of those licenses go unused, which adds up to tens of billions in wasted spend every year. Buying is the right default. Buying on reflex is how the pile grows.
Underneath the whole decision is one idea: buy the commodity, build the difference. Buy the things every business does the same boring way. Build, or have built, only the part that is actually yours. Five questions sort one from the other.
1. Is this your secret sauce, or is it table stakes?
If a capability is how you win, the thing your customers feel and your competitors cannot easily copy, that is worth owning and shaping to fit you exactly. If it is something every shop in your trade does the same way, buy it and move on. Nobody ever beat a competitor with a slightly better expense report.
2. Will the tool fit how you work, or will you bend the business to fit it?
Off the shelf software assumes a standard process. If yours is standard, wonderful, go buy it. But if your business lives in its exceptions, the odd approval, the pricing only you understand, the step that only makes sense in your trade, you will spend forever bending the tool, and it will keep snapping back to how its makers pictured the world. We got into that fight in why off the shelf software fails service businesses. A tool you wrestle every day costs more than its price tag.
3. What does it cost over three years, not in month one?
The price you see is per user, per month, and it climbs. Add the seats you grow into, the higher tier you need for the one feature that matters, the add-ons, the hours spent feeding it. Now set that real three year number next to the cost of building the thing once. Sometimes buy still wins clean. Often the gap is a lot smaller than the trial price made it look.
4. Will it connect to what you already use, or start another island?
Every tool that does not talk to your others becomes a place where data goes to get re-typed. The real cost of the 87 app pile was never the subscriptions. It is the person quietly copying numbers from one app into another, all day, as the human glue. Before you add one more, ask whether it plugs into where your team already works, which for most businesses is Google Workspace, or whether you are about to build one more island somebody has to row out to.
5. When you want to leave, do you keep your data and your logic?
Some tools are easy to walk into and very hard to walk out of. Your data lives in their format, your process lives in their settings, and the cost of leaving climbs with every month you stay. Owning your core means owning the data and the rules behind it, so the business is never held hostage to one vendor’s next price increase.
A word on the one seat trick
Here is one I see constantly. To save money, a business buys a single seat and shares the login around the team. On paper it works, and it is cheap. In practice it comes apart fast. Nobody can tell who changed what. Two people sign in at once and step on each other’s work. Every “who did this?” turns into a guessing game, and the history, if there is any, points at one shared account that everybody used. There is also the part people do not say out loud: nobody actually likes sharing a sign in. It feels cramped, and a little sneaky. If the only way a tool pencils out is by pretending five people are one person, that is worth listening to.
The honest answer is usually both
For most growing businesses the right move is a mix. Buy the commodities, all of them, with no guilt at all. Then build only the thin slice that is truly yours, which is usually not a giant platform but a small, focused tool that ties together the apps you already pay for and takes over the steps your team does by hand. That kind of thing can live right inside Google Workspace, on top of the tools your team already trusts.
The takeaway
The goal was never more software, and it was never custom software for its own sake. It is the least software that runs your business well. Default to buy. Build when it is your core, when nothing on the market actually fits, or when the manual workaround has quietly grown into real hours, which is the line we drew in the 20 hour rule. And if you are about to buy yet another subscription to prop up a tracker you already outgrew, the fix might not be a tool at all, which is the trap we covered in from spreadsheet to app.