Short answer
Bespoke software is software built for one business and its exact process. Off-the-shelf software is a product built for thousands of businesses and configured to fit. Buy off-the-shelf when your process is ordinary. Build bespoke when the software is part of how you win, or when the workaround has become someone's full-time job.
Most South African businesses should start with off-the-shelf and only build when it stops fitting. This guide shows where that line sits, with real rand figures from our own projects.
What is bespoke software development?
Bespoke (also called custom) software development means a team designs and builds an application around your workflow instead of asking you to adapt to theirs. You get a system that does exactly what your business needs, and you own the code.
The word covers a wide range. On our side it means a complex platform, an AI-native product, an internal tool that replaces a tangle of spreadsheets, or a long partnership where scope evolves each quarter. A fixed-scope marketing website is not bespoke software in this sense. That is a different product with a different price.
Typical examples:
- A multi-tenant platform where each client sees only their own data. We built Athleet, a club management platform, and had it live for three pilot clubs in under four weeks.
- A billing system that invoices, follows up and reconciles payments without a person touching the happy path.
- A booking or dispatch tool shaped around how your team actually schedules work.
- An internal dashboard that pulls from three systems your vendors will never integrate for you.
Bespoke vs off-the-shelf: the comparison
Here is the trade-off, one dimension at a time.
Upfront cost
- Off-the-shelf: Low. A subscription, usually per user per month, with no build bill.
- Bespoke: High. Our smallest bespoke engagement was an R85,000 discovery sprint. Our largest was a R2.4 million build over nine months. Custom web applications most often land between R180,000 and R450,000.
Running cost over three to five years
- Off-the-shelf: Grows with headcount. Ten seats today is thirty seats in two years, and the vendor sets the price rise.
- Bespoke: Hosting and maintenance, which stay roughly flat as you add users. This is where bespoke can win on total cost.
Fit to your process
- Off-the-shelf: Good for the 80 percent that every business shares (accounting, email, payroll). You bend the last 20 percent with workarounds.
- Bespoke: Fits the way you work, including the odd rules that make your business different.
Speed to start
- Off-the-shelf: Days. Sign up and go.
- Bespoke: Weeks to months. A discovery sprint takes two to four weeks, then the build.
Ownership and lock-in
- Off-the-shelf: The vendor owns the product and the roadmap. You rent access, and moving your data out can be painful.
- Bespoke: You own the code, the hosting account and the data. Ours ships to your GitHub and your Vercel, not ours.
Integrations
- Off-the-shelf: Limited to what the vendor supports, or to whatever a connector tool can bridge.
- Bespoke: Built to talk to your other systems, including local ones such as payment providers and WhatsApp.
Risk
- Off-the-shelf: Low technical risk, but the vendor can change pricing, features or terms.
- Bespoke: Higher delivery risk, reduced by a prototype of the hardest part before you commit to the full build.
When off-the-shelf wins
Be honest about this first, because it saves money. Off-the-shelf is the right call when:
- Your process is standard for your industry and the software already does it.
- You have fewer than a handful of users and no integration needs.
- You are still testing whether the business idea works. Do not build before you have proof.
- The cost of a workaround is small and nobody complains about it.
If that is you, buy the product. We would rather tell you that in a first call than sell you a build you do not need. We stopped three projects at discovery in the last year for exactly this reason.
When bespoke wins
Bespoke starts to pay when one or more of these is true:
- The software is the product or the edge. If your service depends on a process competitors cannot copy by signing up to the same SaaS, that process deserves to be owned.
- A spreadsheet has become a full-time job. When someone spends their week moving data between tools, the labour cost passes the cost of fixing it.
- Per-seat pricing is punishing growth. If every new hire adds a licence fee on a tool that is mostly a database and a form, building can undercut renting.
- No product fits. Some work, such as multi-tenant platforms with unusual permission rules or AI-native products, has no off-the-shelf equivalent.
- You need to own the asset. Investors and buyers value code and data you control more than a subscription to someone else's.
The middle path: buy the core, build the gap
You do not have to choose one side. A common and sensible pattern:
- Keep off-the-shelf tools for accounting, email and the parts that are the same for everyone.
- Build the one piece that is specific to you, and connect it to those tools by API.
- Add bespoke pieces one at a time as each one earns its keep.
This keeps the first bill small and avoids a big-bang rebuild. It is also how most of our web application work starts.
How to decide: five questions
1. What does the current workaround cost in hours per week, and what is an hour worth?
2. Would a competitor gain an advantage by using the same tool you do?
3. What will this software cost per user in year three if you rent it?
4. How hard would it be to leave this vendor if their pricing changed?
5. Can the hardest part be proven in a small prototype before you commit to the whole build?
If the answers point to rent, rent. If they point to build, start with a prototype and an honest estimate, not a full contract.
What a sensible bespoke engagement looks like
A good build partner should not ask you to sign for the whole project on day one. Ours runs in four steps:
- Deep dive: a candid conversation, sometimes ending in "do not build this".
- Discovery: two to four weeks producing an architecture, a prototype of the hardest part and an honest cost, fixed price from R85,000.
- Build: we build it, or hand your team a spec they can run with.
- Partner: advisory, fractional CTO work or a standing engagement as the product grows.
You can take the discovery output to any team. That is the point: you buy certainty, not a commitment. See the full detail on our bespoke builds page, or compare with our other pricing.
Frequently asked questions
What is the difference between bespoke and custom software?
In practice nothing. Both mean software built for one organisation's needs. "Bespoke" is the more common term in South Africa and the UK, "custom" in the US.
How much does bespoke software cost in South Africa?
It depends on the shape of the work, not a price card. Our range so far runs from an R85,000 discovery sprint to a R2.4 million, nine-month platform build. Custom web applications typically cost R180,000 to R450,000, quoted after a two-week discovery.
Is bespoke software worth it for a small business?
Sometimes. If your process is ordinary, off-the-shelf is cheaper and faster. It becomes worth it when a workaround costs real hours every week, when per-user fees compound, or when the software is your competitive edge.
How long does it take to build bespoke software?
A discovery sprint takes two to four weeks. A focused first version can ship in weeks: Athleet went live for three pilot clubs in under four weeks. Larger platforms take months.
Who owns the code?
You do. Our standard terms include IP assignment, and the code lives in your own GitHub and hosting accounts.
Can I start small and expand later?
Yes, and we recommend it. Build the one piece that matters, connect it to your existing tools, and add more once it proves its value.
Next step
If you are weighing a build against a subscription, a short call is the cheapest way to find out which side you are on. Read about bespoke builds or Book a 20-min call and bring the process you are trying to fix.