Custom software vs off-the-shelf, and the stack of apps you already have
Strategy · 11 min read
Most articles on this topic set it up as a clean choice: buy a product, or build your own. That is not how it looks inside most South African SMEs.
What we usually find is a business already running on five or six tools. Sage or Xero for the books. A POS in the shop. A job or task app someone signed up for. Google Sheets or Excel for stock, pricing and the owner's weekly numbers. And a handful of WhatsApp groups where the real coordination happens.
So the real question is often not "custom or off-the-shelf?" It is "do we keep adding apps, or is it time for one system that ties this together?" This guide is written for that decision.
There are three options, not two
- Off-the-shelf. One established product that covers most of what you do, configured to suit you.
- A multi-app stack. Several off-the-shelf tools, joined by exports, copy-paste, automation tools and staff effort.
- Custom software. A system built around your workflows, usually connected to a few established tools you keep, like your accounting package and payment gateway.
Each is the right answer for some businesses. The mistake is drifting into option two without ever deciding.
When off-the-shelf is the right call
We tell clients to buy rather than build more often than people expect. Off-the-shelf is usually right when:
- Your process is standard for your industry and the product was designed for businesses like yours.
- You are early, still figuring out how you want to operate, and speed matters more than fit.
- You have a small team and a small number of transactions you can check by hand.
- The function is not what makes you different. Payroll and accounting are good examples. There is no prize for rebuilding a general ledger.
- You need something working next week.
If a well-known product covers what you need without daily workarounds, use it. Building custom software for a standard problem wastes money.
The multi-app stack: cheap per tool, expensive in total
Each app in a stack looks affordable on its own. The cost shows up between the apps.
The integration tax
When tools do not share data properly, people become the integration. Someone retypes orders from WhatsApp into a sheet. Someone exports sales from the POS and pastes them into the owner's report. Someone checks stock in one app before quoting in another. When two tools disagree, someone phones around to find out which one is right.
None of this appears on an invoice, so it rarely gets counted. A useful exercise is to list every place where information is copied by hand from one tool to another, who does it and how often. That list is the running cost of your stack.
There are other costs too. Customer data spread across personal WhatsApp chats and several cloud accounts is harder to control under POPIA. Staff leaving means logins and knowledge leaving. And your per-user subscriptions grow every time you hire.
Configuration is not customisation
Many products are sold as "fully customisable". Usually that means configurable: you can rename fields, switch modules on and off and choose from preset workflow options. That is useful, but it is not the same as the software doing something it was not designed to do.
If your business depends on a rule the product does not support, such as pricing that changes by branch and customer type, or a job that needs two approvals only above a certain value, you will end up with a workaround. Ask vendors to show you your exact scenario in a demo, not their standard flow.
Where low-code and automation tools fit
Automation tools that move data between apps can be a good step. They are quick to set up and can remove a lot of copy-paste. They work best for simple, reliable triggers: a new form entry creates a row, a paid invoice sends a message.
They struggle when the flow has many exceptions, when you need a proper audit trail of who changed what, or when a failed sync goes unnoticed for a week. At that point you have built a system anyway, just one that is hard to see, test and support. If that sounds familiar, look at proper business automation systems, which are built with error handling and logging from the start.
When custom software makes sense
Custom software earns its cost when several of these are true:
- Staff work around your current tools every day.
- The owner's reports are rebuilt by hand from several sources and nobody fully trusts them.
- You run more than one branch, yard or team, and each does things slightly differently.
- Different people need to see different slices of the same data: customers, drivers, branch managers, head office.
- The way you deliver your service is part of why customers choose you, and generic tools flatten that.
- You need connections that your current tools do not offer, such as WhatsApp Business updates driven by job status, or stock that updates when a sale is captured at a branch.
- Staff need to keep working through power cuts or poor signal, and the system has to catch up when they reconnect.
Side-by-side comparison
| Factor | Off-the-shelf | Multi-app stack | Custom software |
|---|---|---|---|
| Time to start | Fast | Fast, then grows over time | Slower, needs scoping and build |
| Upfront cost | Low | Low per tool | Higher |
| Ongoing cost | Subscription per user | Several subscriptions plus staff time on workarounds | Hosting, support and changes |
| Fit to your process | You adapt to it | Each tool fits one part | Built around your process |
| Single source of truth | Yes, within the product | Rarely | Yes, if designed properly |
| Reporting | Standard reports | Manual consolidation | Built from your own data |
| Control over changes | Vendor roadmap | Several vendors' roadmaps | Your priorities |
| Data control and POPIA | Vendor terms | Spread across many services | Designed in, with clear access rules |
The hybrid most SA SMEs end up with
In practice, the best answer is often a mix. Keep what is good and standard: Sage, Pastel or Xero for accounting, your payment gateway for payments, maybe your POS. Build a custom operations layer for the part that is specific to you: jobs, stock across sites, bookings, dispatch or approvals. Then connect them so data flows once, in one direction, with clear rules about which system owns which record.
This avoids paying to rebuild things that already work, while replacing the WhatsApp groups and spreadsheets that are doing the heavy lifting. It also means you are not betting everything on one big switch-over.
If your "ops layer" is really stock, purchasing and branches in one place, that is ERP territory. See our sibling post "Custom ERP vs Sage / Odoo / Dynamics for SA SMEs" and our ERP development page.
A quick self-check
Answer yes or no for your business today:
- Does anyone copy information by hand between two or more tools every day?
- Is there a spreadsheet that the business would struggle without, maintained by one person?
- Do important decisions or approvals happen in WhatsApp with no record in a system?
- Does a monthly or weekly report take someone hours to assemble?
- Would opening another branch or adding another team mean more workarounds?
- Have you asked a software vendor for a feature and been told it is "on the roadmap"?
If you answered yes to most of these, it is worth scoping a custom system or a proper automation layer. If you answered no to most, a better-configured off-the-shelf product is probably the cheaper path. Our sibling post "When does an SA SME actually need custom software?" goes deeper on this.
Frequently asked questions
Custom software vs off-the-shelf: which should I choose?
Choose off-the-shelf when your process is standard and a known product covers it without daily workarounds. Choose custom when your workflow, integrations or multi-site setup do not fit available products and the workarounds are costing staff time and accuracy. Many SA SMEs use both.
Is it worth building custom software instead of using multiple apps?
It can be, when the time spent moving data between apps, fixing mismatches and rebuilding reports is significant and growing. Count the manual hand-offs between your tools first. If there are many, one system connected to the few tools you keep is often better value over time.
Should my business use low-code platforms or custom development?
Low-code and automation tools are good for simple, stable flows and quick wins. When the process has many exceptions, needs an audit trail, or failures must be caught and handled, custom development is usually more reliable and easier to support.
Can I use custom software and off-the-shelf tools together?
Yes, and it is common. A typical setup keeps accounting in Sage or Xero and payments on an established gateway, with a custom system handling operations and syncing to both.
What is the difference between custom and enterprise software?
Enterprise software is a large packaged product sold to many big organisations and configured for each one. Custom software is built for one business. SMEs usually want the control of a tailored system without the size and implementation effort of enterprise packages.
What happens to my data if I change systems later?
With any product, check export options before you sign. With ThinkinCode's managed platforms, you keep ownership of your operational data and can export it. Specific IP and licensing terms for dedicated builds are agreed in the proposal.
Get a straight answer for your business
We are not in the business of talking people into custom software they do not need. If a product fits, we will say so. If your stack is costing more than it saves, we will show you where and suggest a first phase.
Email [email protected] to book a Discovery call with a senior engineer, or read more about how we do custom software development.
Ready when you are
Book a Discovery call
Tell us what you are trying to fix. We will tell you honestly whether custom software is the right next step.