Sooner or later every growing business reaches the same fork. The packaged software you bought years ago no longer fits how you work, and someone suggests building your own. Custom software vs off the shelf is rarely an all-or-nothing choice, but real money rides on it, and the sales pitches on both sides can make it harder to see clearly.
This guide clears away five common myths, then gives you a short scorecard you can fill in with your team in an hour.
What each option really means
Off-the-shelf software is a product built for many companies: an accounting package, a CRM, a field-service app. You pay a license or subscription, configure it within the options it offers and adapt your process to fit.
Custom software is built for your business alone, around your process. You pay to have it designed and built, you own it, and you pay to maintain it.
Between them sits a third route that many mid-size firms end up taking: keep a packaged system for the standard work and build custom pieces around it.
Custom software vs off the shelf: five myths
Myth 1: Off-the-shelf is always cheaper
It's cheaper to start. Over five years, per-seat fees, add-on modules, integration work and staff time spent on workarounds add up. Licenses also go unused.
Zylo's 2024 SaaS Management Index, drawn from the software spend it manages for its customers, found organizations were using only 49% of the licenses they had provisioned. Compare the total cost over the life of the system, with year one as just the first line.
Myth 2: Custom software takes years and always overruns
Big-bang projects do carry real risk. A McKinsey and University of Oxford study of more than 5,400 large IT projects found that, on average, they ran 45% over budget and 7% over time while delivering 56% less value than predicted. Those were projects with initial budgets above $15 million. The lesson for a mid-size firm: keep the first release small, see working software every week and fix the price of each stage before it starts.
Myth 3: Custom means you're locked in to the developer
Only if the contract allows it. Insist that the code sits in a repository under your name, runs in your own cloud account and comes with documentation. Then any competent team can take it over. Packaged software can lock you in too, through your data format and the vendor's price rises.
Myth 4: Once custom software is built, it's done
All software needs upkeep: security updates, new browser versions, changes in the systems it connects to. Skipping that builds up technical debt. In a McKinsey survey of CIOs, respondents said 10% to 20% of the technology budget meant for new products was diverted to resolving tech-debt issues. Budget for maintenance from day one, whichever route you take.
Myth 5: Off-the-shelf works out of the box
Configuration, data migration, integration and training are real projects in their own right. A packaged ERP can take months to fit to a business, and the parts that don't fit tend to end up in spreadsheets on the side.
A decision scorecard for build vs buy
Score each question from 1 to 5 with the people who run the process. Low totals point to off-the-shelf, high totals to custom.
Question | Score 1 (lean off-the-shelf) | Score 5 (lean custom) |
|---|---|---|
How standard is the process? | Same as most firms in your industry, such as payroll | Unusual, and part of why customers choose you |
How well do packages cover it? | A package covers almost everything | Every package forces big workarounds |
How many systems must it connect to? | One or two, with ready-made connectors | Several, including older in-house systems |
How many users? | A handful | Hundreds, where per-seat fees add up |
How often does the process change? | Rarely | Often, and the software has to change with it |
How sensitive is the data? | Standard business records | Regulated or contract-restricted data you must keep in your own systems |
6 to 12: buy and configure. 13 to 20: buy the core and build around it. 21 to 30: a custom build is worth pricing properly.
Treat the total as a starting point for discussion. One answer can outweigh the rest. If a process is the reason customers pick you over a competitor, that alone may justify building it. If you'd like an outside view on your scores before spending anything, that's what our software consultancy work is for.
The middle route: buy the core, build the edges
Most mid-size firms don't need to replace everything. Common patterns include:
A client portal on top of your packaged ERP, so customers check orders and download documents without emailing your team.
A mobile app for field staff that writes straight into the office system.
Integrations that stop people retyping data between the CRM, accounts and warehouse system.
A custom module for the one process no package handles well, such as quoting for made-to-order products.
This keeps standard work on standard software, with the vendor's updates and support, and puts your money where your business is different. Our custom software development page shows the kinds of portals, apps and modules this covers.
Compare five years of cost, line by line
Quotes for the two routes rarely compare like with like. Put every cost line side by side over five years before you decide.
Cost line | Off-the-shelf | Custom |
|---|---|---|
Up front | Setup, configuration and any implementation partner | Design and build, usually in stages |
Every year | Subscriptions or licenses per user, plus add-on modules | Hosting, maintenance and small improvements |
Integration | Connectors, where they exist, and custom work where they don't | Built in from the start |
Data migration | Moving and cleaning your records | Moving and cleaning your records |
Hidden costs | Workarounds, side spreadsheets and price rises at renewal | Upkeep you didn't budget for |
Exit | Export your data and retrain staff on a new tool | Hand the code and documentation to another team |
Ask your finance team to fill in real figures from your current software spend. The comparison often settles the question faster than any debate about features.
Questions to ask before you build
Who owns the code, the database and the cloud accounts?
What is in the first release, and when can our team use it?
How will we see progress? A working demo every week tells you more than a status report.
How will our data move across from the old system, and how will we check it?
What will maintenance cost each year after launch?
Will it pass our IT team's security and procurement review?
Frequently asked questions
How does the cost of custom software compare with a subscription?
It depends on scope, the number of systems involved and the number of users. Compare the build plus five years of maintenance against five years of subscriptions, add-ons, integration and workaround time. Ask any developer for a fixed price on a small first stage before committing further.
Can we start with off-the-shelf and switch later?
Yes, and it's often sensible. Before you sign, check that you can export all your data in a usable format, so a later move doesn't begin with a data-recovery project.
Is custom software less secure than a well-known product?
Not by nature. Security depends on how software is built, tested and maintained. A custom system can be built to your IT team's standards and run inside your own cloud account, under your own controls.
How soon can we use a custom system?
Keep the first release small enough to use within a few months. If a proposal can't show you working software within its first few weeks, ask why.
If you're stuck between the two, a short conversation is a quick way to test your thinking. Tell us what you run today and where it pinches, and we'll tell you plainly whether building makes sense or a package will do the job. Book a 30-minute call with a founder whenever it suits you.
