Custom software or off-the-shelf? A decision checklist for small businesses

The custom software vs off the shelf question usually starts with frustration. A subscription tool almost fits, the team has built workarounds around the gaps, and someone asks whether it would be cheaper to build exactly what you need.

Sometimes it is. But most of the time it isn't. We build custom software for a living, and we still tell most people to buy first. This post gives you a checklist to score your own case, plus the option most people miss: buy the core and build only the connection.

Start with the honest default: buy

Off-the-shelf software has advantages that are hard to beat. Someone else pays for security updates, hosting and new features. Thousands of other customers have already found the bugs. You can start this week.

Custom software makes sense when the thing you need is close to how you make money, or when every tool on the market forces you into a process that costs real time every day. If neither is true, buy.

The custom software vs off the shelf checklist

Score each question 0, 1 or 2. Be honest; nobody is grading you.

1. How close is this to what makes you different?

• 0: It is standard work every business does: email, payroll processing, basic accounting.
• 1: It is standard, but your version has a few rules nobody else has.
• 2: It is how you win work, price jobs or serve customers in a way competitors don't.

For example, accounting is a 0 for almost everyone. A quoting process that encodes twenty years of pricing judgment can be a 2.

2. How much do the workarounds cost each week?

• 0: A few minutes here and there.
• 1: A few hours a week across the team.
• 2: A person's job is mostly re-typing, reconciling or checking.

Workarounds hide. So count the copy and paste between systems, the weekly spreadsheet that merges two exports, and the time spent fixing mistakes from both.

3. What will it cost over three years, not one?

Compare total cost over three years, not the first invoice.

For a subscription, add the per-seat price times the seats you will have in three years, plus the price increases you can reasonably expect, plus the paid add-ons you will end up needing.

For a custom build, add the first release, the changes you expect in years two and three, and hosting plus maintenance every month.

• 0: The subscription is clearly cheaper over three years.
• 1: They are close.
• 2: The subscription costs more, usually because of seat counts or add-ons.

4. Do you need to own the data and the way it is stored?

• 0: Exporting a CSV now and then is enough.
• 1: You need reliable access through an API for reports or other tools.
• 2: The data model itself is the business, and you need full control of it.

5. How painful would it be to switch later?

• 0: You could move to a competitor in a weekend.
• 1: Switching would take a few weeks of cleanup.
• 2: You would be locked in, with your history stuck in a format only that vendor reads.

Reading your score

• 0 to 3: Buy. Then spend the money on setting the tool up well and training the team.
• 4 to 6: Look hard at the hybrid option below.
• 7 to 10: A custom build is worth scoping.

The score is a conversation starter, not a verdict. For instance, one 2 on question 1 can outweigh three 0s elsewhere.

The hybrid: buy the core, build the glue

The best answer is often neither pure option. Instead, you buy the parts that are standard and build only what connects them or fills the one real gap. A few examples from our own work show the pattern:

• Real estate listings. For the Live Local Vermont site, listing search comes from an IDX provider connected to PrimeMLS. We did not build a listing database. We built the site around it, plus a check that blocks any deploy where a page is missing a brokerage or MLS display rule.
• An online store. TACOVICI's store runs on Shopify, which handles checkout, taxes and hosting. The custom part is what Shopify can't know: a theme built around one product and a tool for placing a mirror on a bathroom wall at real size.
• Payroll and hiring. The workforce platform we build for a staffing company exports each payroll run to QuickBooks and runs background checks through an outside service, rather than reinventing either. The custom part is the staffing workflow those tools don't cover.

In each case the bought tool does the heavy, regulated, well-understood work. The custom code is small, focused and easy to replace.

Questions to ask before you commit either way

If you are leaning toward buying:

1. Can we export all our data, in a usable format, at any time?
2. Does it have a documented API for the tools we already use?
3. What did the price do over the last three years?

If you are leaning toward building:

1. Will we own the code, the hosting account and the domain?
2. What is the smallest first release that removes the most pain?
3. Who maintains it, and what does that cost each month?

For a rough sense of what sets the price of a build, see how much custom business software costs.

When we say "don't build this"

Part of every scope we write is a recommendation on what to build, what to buy and what to keep. Often the honest answer is a tool you already pay for, connected properly to the rest. If that is your answer, we will say so. If you want a second opinion on your own checklist, send us your scores and the tools you use now. The internal tools page shows what a small custom build looks like.

Cover photo by Kaleidico on Unsplash.