Custom software vs off-the-shelf: an honest comparison from people who build custom

A disclosure before anything else: we’re a custom software studio, so we have an obvious bias. Here’s why you can read this anyway — about a third of the enquiries we get end with us saying “don’t hire us, use this off-the-shelf product.” Building the wrong custom software is bad for the client and, eventually, bad for the builder. So this is the framework we actually use on those calls.

The short version

• Buy off-the-shelf when your problem is one thousands of businesses share, and you’re willing to adapt your process to the tool.
• Build custom when your process is your edge, when off-the-shelf keeps almost-fitting, or when the real problem is systems that don’t talk to each other.
• Most small businesses should do both: off-the-shelf for the commodity functions (accounting, email, payroll), small custom tools for the workflows that make them different. The interesting decision is almost never “replace Xero” — it’s the gap between the systems.

What each one really costs

The sticker prices mislead in both directions.

Off-the-shelf looks cheap: $20–$150 per user per month. But multiply it out — a $49/user/month tool for 12 staff is roughly $7,000 a year, every year, rising with each price increase and each “that feature is on the Premium plan.” Five years in, you’ve spent $35,000+ and own nothing. For many tools that’s still a great deal. For some, it’s a custom build paid for three times over.

Custom looks expensive: a one-off $8,000–$60,000 depending on scope (see our breakdown of real custom software prices). But there are no per-seat fees, and ongoing cost is hosting plus maintenance — typically 10–20% of build cost per year. The catch runs the other way: you pay upfront, you wait weeks instead of minutes, and you own the responsibility for keeping it alive.

The honest comparison is five-year total cost for your head count, not month one versus month one.

Where off-the-shelf wins (buy it, don’t build it)

1. Solved problems. Accounting, payroll, email, document storage, video calls, generic CRM. Xero has had thousands of engineer-years invested in Australian tax compliance. Nobody should build that, including us.
2. Speed. Sign up today, working this afternoon. No custom project beats that.
3. Someone else carries the maintenance. Security patches, server failures, browser updates — the vendor’s problem, amortised across thousands of customers.
4. Best practice built in. Good off-the-shelf tools encode how well-run businesses handle that workflow. If your process is messy because it grew by accident, adopting a tool’s opinionated process is sometimes the upgrade.

Where custom wins (build it)

1. Your process is your competitive advantage. If the way you quote, schedule, or serve customers is why customers pick you, forcing it into generic software sands off the edge that wins you work. Commodity processes deserve commodity software; differentiating processes don’t.
2. The 80% trap. The most common reason people call us: an off-the-shelf tool does 80% of the job, and the missing 20% is the part they needed most. The symptoms are workaround spreadsheets orbiting the SaaS tool and staff maintaining the same data in two places. If you’ve bought two or three tools for the same problem and abandoned each one, the problem probably isn’t generic.
3. Integration gaps. Each tool is fine; the seams are the problem. Re-keying data from the job system into Xero, from the inbox into the CRM. Small “glue” software that moves data between systems is the highest-ROI custom work that exists — most of the tools we’ve built are exactly this, and several cost less than a year of one unused SaaS subscription.
4. Per-seat pricing punishing your structure. Twenty casual staff who each touch the system five minutes a day make per-user SaaS pricing absurd. Custom software doesn’t care how many logins you have.
5. You’re the wrong shape for the market. Niche industries, unusual compliance, processes spanning what software vendors consider three different categories. If every demo ends with “we don’t quite do that,” stop demoing.

The risks nobody puts in the brochure

Off-the-shelf risks: price rises you can’t refuse (migration pain is the lock-in), features killed or paywalled after you depend on them, vendors acquired or shut down, your data hostage in a proprietary export format, and slow drift into workaround spreadsheets that recreate the chaos the tool was meant to end.

Custom risks: scope blowouts from vague requirements, builders who disappear after launch (ask “who maintains this in year three?” before signing), and over-building — commissioning a $60,000 platform when a $6,000 tool plus Airtable would have done. The mitigation for all three is the same: start small, fixed scope, code ownership in writing, and a maintenance plan agreed upfront.

A decision framework you can run in ten minutes

Work through these in order; stop at the first clear answer.

1. Is this a commodity problem? (Accounting, payroll, email, storage.) → Buy. Always.
2. Does a well-reviewed tool cover ~95% of your need, and can you live with adapting to it? → Buy. The remaining 5% is cheaper as a habit change than as software.
3. Does the gap between tools — double entry, re-keying, copy-paste — hurt more than any single tool? → Build small glue automation. Keep the SaaS tools; connect them. Usually $2,000–$10,000. (Our automation ideas guide lists a dozen examples with costs.)
4. Is the painful workflow the thing that makes your business distinctive, or shaped like nothing the market sells? → Build custom, starting with the smallest version that removes the worst pain.
5. Still unsure? → Buy the cheap tool and use it hard for three months. You’ll either be happy (great, you saved tens of thousands) or you’ll have produced the world’s most precise custom-software specification: a documented list of exactly where it falls short.

Step 5 is real advice we give paying clients. A failed $300 SaaS experiment is the cheapest requirements document you’ll ever write.

The hybrid most small businesses actually end up with

The pattern we see working, over and over, in Australian small businesses:

• Xero or MYOB for accounting — never custom.
• Google Workspace or Microsoft 365 for email and documents — never custom.
• One or two SaaS tools for commodity workflows that fit well enough.
• One small custom tool for the workflow that makes the business money and never fit anything off the shelf — often a replacement for the spreadsheet that had grown past its limits.
• Thin automation stitching the above together so nobody re-keys anything.

Total custom investment in that picture is usually $5,000–$25,000 — not the six-figure “digital transformation” the phrase custom software tends to conjure.

The bottom line

Buy software for the problems you share with every other business. Build software for the problems that make you you — and build the smallest version first. If you’re not sure which side of the line your problem sits on, describe it to us. If off-the-shelf will do the job, we’ll name the product and you’ll have saved yourself a project; if it won’t, you’ll get an honest fixed quote instead of a discovery retainer.