Most comparisons between building and buying fail the same way: one option is priced in full and the others are priced from memory. A subscription looks like a clean monthly number, a build looks like one large invoice, and the status quo looks free.
This guide is a method for putting all three on the same sheet. It will not tell you which to pick. It will make sure you are comparing real costs.
#The three options
Compare these side by side, not just build versus buy:
- Keep what you have. The spreadsheets, the manual steps, the tools you already pay for.
- Buy or subscribe. An off-the-shelf product, configured to fit as well as it can.
- Build. Software made for your process.
Keeping what you have is an option with a cost. It is just a cost that nobody reviews, because it shows up as hours instead of invoices.
#The cost worksheet
Pick a time frame (three years works well for most small businesses) and fill in each line for each option. Write zero where a line truly does not apply, and write a guess where you are unsure. A rough number beats a blank.
Costs people forget:
- Seat growth. Price per user today, and price per user when the team is larger. Include the seats you would add if hiring goes as planned.
- Add-ons and integrations. The features you need that are not in the base plan, plus the connectors or middleware that move data between tools.
- Time spent on workarounds. Hours per week that people spend copying, reconciling or fixing things the tool does not handle, multiplied by what those hours cost.
- Migration. Moving existing data in, cleaning it, and running two systems side by side for a while.
- Training. Time for everyone who will use it, plus time for new hires later.
- Maintenance and support. For a subscription, the support tier you actually need. For a build, hosting, monitoring, fixes and improvements.
- Switching costs later. What it takes to leave: exporting your data, rebuilding reports, retraining. Ask each vendor how you would export everything, and in what format.
Add the one-time costs and the recurring costs separately. Then total them over your time frame.
Imagine a five-person team paying a per-seat price that doubles once they hire five more people. The first year looks cheap and the third year does not. Writing the headcount plan next to the price is what exposes that, and it is the line people skip most often.
#A short scoring table
Cost is only half the comparison. Score fit as well, from 1 (poor) to 5 (strong), with a short note.
| Question | Keep | Buy | Build |
|---|---|---|---|
| Does it match how we really work? | |||
| Can it connect to the systems we use? | |||
| Can it handle the volume we expect? | |||
| Can we change it when the process changes? |
Weight the rows that matter most. If one question is a deal-breaker, treat a score of 1 there as a disqualifier regardless of the total.
#A decision rule of thumb
Some practical corollaries:
- If you would change your process to fit the tool, and that change is harmless, buy.
- If you would change your process and it would hurt, price a build.
- If the honest answer is “keep,” do it on purpose and schedule a review, because volume and headcount drift.
#Check your numbers
Before you decide, have someone who will not benefit from the answer look at your worksheet. The most common errors are an underestimated workaround line and an overestimated build line.
Ask for a fixed-price quote for the build option, so the number you compare is a commitment and not a hope. For an opinion piece on the same topic, read custom software vs another SaaS subscription. If you want a second pair of eyes on your sheet, you can book a 30-minute call with an engineer.
