Off-the-Shelf vs. Custom Software: How a Small Business Should Actually Decide
Buy it. Almost always, buy it. Custom software is the most expensive way to solve nearly any small business problem, and the businesses that ask me about building something usually don't need anything built. They need to stop paying for five subscriptions that almost do the job.
That said, building is right sometimes. Here is how to tell which situation you're in, with the actual numbers.
1. Do the real math, not the sticker math
A $50-per-seat tool for 12 people is $7,200 a year. That sounds like a lot until you price the build. In 2026, a small custom business app runs $5,000 to $25,000, a mid-size system $25,000 to $100,000, and one pricing roundup of Clutch-listed projects puts the average build around $132,000. The build price is the down payment, not the total: hosting, maintenance, and the day your developer moves on all come later.
So run five years. Five years of the subscription: $36,000. Five years of a $60,000 custom system plus maintenance and the inevitable phase two: almost always far more than $36,000. The subscription wins unless it genuinely can't do the job. Where subscriptions lose is the pile-up: five tools at $40 a seat for 20 people is $48,000 a year, and now you're paying custom-build money for a stack nobody loves. Audit the stack before you build anything.
2. Ask whether this is plumbing or product
Plumbing is everything that keeps the business running but isn't the business: scheduling, invoicing, payroll, CRM, inventory. Product is what customers pay you for. Buy plumbing. Build only when the software is the thing that makes you different.
A Missoula HVAC shop with 12 techs does not need custom dispatch software. It needs the standard dispatch software, set up right. A company whose whole pitch is a booking experience competitors can't copy: that is product, and that is the build.
3. Price the workaround tax
Count the hours your almost-good-enough stack costs you: the Friday CSV export, the manual re-entry between two systems, the spreadsheet only one person understands. Three hours a week at a loaded cost of $40 an hour is $6,240 a year. If the workaround costs more than the fix, the decision already made itself. You just haven't done the arithmetic.
4. The third option: buy the core, build the edges
Most businesses pick a camp: buy everything and suffer the gaps, or build everything and suffer the bill. The third option is where the smart money goes. Buy the platform that does 90 percent of the job, then build only the missing piece as an integration. Nearly every serious SaaS product has an API, and a small integration connecting your CRM to your invoicing, or a low-code tool covering the one weird process you have, usually kills the case for a full custom build at a fraction of the cost.
Say you run a shop with one tool for the storefront and another for bookings. A modest integration that moves every booking into the customer record automatically costs a fraction of the unified custom platform quote, and it ships in weeks, not quarters. That is the third option in action.
5. If you build, own everything
Before a dollar moves: you own the code, the repository, the hosting accounts, and the domain, in writing. Fixed scope with a fixed price, or walk away. Milestones with working demos, never a big reveal at the end. And ask what year two costs before you agree to year one. A builder who can't answer that is selling you a project, not a system.
Buy first. Build only what makes you different. Integrate the rest. Most businesses do this backwards and pay twice: once for the subscriptions they half-use, once for the custom system they never needed.