Should you buy off-the-shelf software or build something custom?
Buy when a generic tool covers what your business actually needs, and the small percentage it does not cover genuinely does not matter. Build custom when that uncovered part is the part your business runs on, because a subscription that does not fit will cost you in workarounds every month for as long as you keep paying for it.
Most businesses treat this as a budget decision. It is really a fit decision, and the two get confused constantly because the price tags are so far apart at the start. A subscription is cheap on day one. Custom software is not. That difference is real and it matters, but it is only half the sum.
When off-the-shelf is the right call
We say this to clients often enough that it is worth being upfront about it: plenty of problems do not need custom software at all. Buy an existing tool when:
- Your process is close to how most businesses in your industry already work
- The tool's defaults cover the workflow, and the gaps are minor enough to live with
- You need something running this week, not this quarter
- The category is genuinely commoditised: accounting, email, basic scheduling
Accounting software, email, generic scheduling tools. Buying is usually right there, and we tell clients so. Where it stops being right is the moment your business is winning on the part that tool cannot do.
When custom software earns back what it costs
The businesses that come to us for custom internal tools and business systems have usually already tried the off-the-shelf route. Not because they skipped a step, but because the fit only breaks once a business has actually grown into the part that does not match the shape of the tool.
The tell is consistent: staff keeping a second spreadsheet for the thing the software cannot do, manually re-typing data between two systems that were never going to talk to each other, or a process getting quietly simplified to fit what the tool allows rather than what the business needs. Each of those looks like a minor annoyance in isolation. Together, they are a business paying, in time, for software that was supposed to save it time.
The cost comparison nobody puts in the sales deck
A subscription's sticker price is the cheapest number in the conversation, and it is also the least complete one. It compounds monthly, scales per seat, and rarely covers what the workaround costs in staff time. Custom software is the opposite shape: a larger number once, then a system you own outright that does not charge you more as you add people or grow the process it was built for.
Neither shape is automatically cheaper. A five-person business with a simple process usually loses on total cost by building custom for something a R500-a-month tool already does well. A business whose whole operation runs on the part no off-the-shelf tool handles usually loses, for years, by not building. The honest question is not "what does this cost to start", it is "what does this cost to run for three years, including everything it doesn't fix."
What this looked like for BSL Auction Services
BSL runs vehicle and machinery auctions, where a single sale day generates more records than most businesses produce in a month. Lots, bidders, bids and settlements were being recorded by hand, and the record only became complete once someone had gone back afterwards and pieced it together. A bigger sale day did not just mean more revenue. It meant proportionally more filing, arriving after the auction instead of during it.
Generic auction or e-commerce platforms exist, and they can list items and take bids. What they do not do is run bidding, listings, and settlements as one operational record that a small admin team can read in real time on a sale day, rather than reconstruct afterwards. Bending BSL's operation to fit a generic tool would have meant giving up the parts of the business that make it BSL, in exchange for software that was never built for an auction house's sale day.
So we built one platform instead of buying one: a public auction site where bidding opens up beyond whoever can be in the room, and an operations dashboard where listings, bids and records live together, captured the moment they happen rather than written up once the day is over. A bid placed by a customer on the public site is already in the record the team works from. That is the part no off-the-shelf tool was going to give them, because it is specific to how an auction actually runs.
ShiftTech built our auction website and admin platform exactly to our needs. They understood our business and delivered a system that simplified operations and reduced manual work. We're very happy with the result.
— Conrad, Operations Manager, BSL Services
The result, in BSL's own terms: the filing that used to follow every sale is now largely produced by the system as the sale happens, the bidding pool is no longer limited to people who can attend in person, and lots, bids and settlements stopped living in separate paper records. Read the full BSL Auction Services case study for the detail on how the platform is built.
The questions that actually decide it
We do not start a discovery call by pitching custom software. We start by trying to talk a client out of needing it, because we quote fixed price and a wrong recommendation costs us, not them. The questions that settle it:
- Is the process you are fixing the same as everyone else's, or specific to your business?
- What is the workaround costing you today, in hours, not just in frustration?
- Would a bigger version of your business still fit inside this tool, or would you outgrow it the way BSL outgrew paper?
- Does the category genuinely have a mature, well-fitting product already, or are you comparing against a tool built for a slightly different problem?
These are the same questions we ask before quoting any custom build, because "should we buy or build" is really the same question as "what does this business actually need", asked one step earlier.
The honest version of our advice
If an off-the-shelf tool genuinely fits, we will tell you that in the first call and save you the build cost. We turn that recommendation into revenue for us exactly zero times, which is precisely why it is worth trusting. When the fit genuinely is not there, custom software stops being the expensive option and becomes the one that stops charging you every month for a workaround.
Not sure which one you need?
Tell us what you are trying to fix. If a tool you can buy today already solves it, we will say so. If it does not, you will get a fixed price and a timeline within 48 hours.
Book a Discovery Call