Bespoke software is built for one business and one job. The advantages are that it fits the work exactly, removes admin no product can reach, and belongs to you outright. The disadvantages are that it costs more up front, needs your time to get right, and depends almost entirely on who builds it.
We build bespoke software for a living, so read this knowing that. We have also told people not to buy it from us, which is the part of the argument most articles on this subject leave out. Below is what the advantages and disadvantages of bespoke software actually look like from the inside, including the cases where you should close this tab and buy something off the shelf.
What is bespoke software?
Bespoke software is a tool built for one business and one job, rather than a general product configured to fit. You commission it, it is written for the way your work actually runs, and you own the result. It is the alternative to bending an off-the-shelf platform around a process it was never designed for.
"Custom software" is the same thing in American English. UK buyers use both, and there is no difference in what you get.
The advantages of bespoke software
It fits the job, so nobody works around it
Off-the-shelf products are built for the average of thousands of businesses. Yours is not average, and the gap between the product and the job is where the admin lives: the re-typing, the copy and paste between two systems, the spreadsheet somebody keeps on the side because the software cannot do the one thing they need.
A regional auction house we worked with ran every lot enquiry through one shared mailbox. Three staff triaged it by hand, questions got answered twice or missed, and it took roughly the first two hours of everyone's morning during sale weeks. No product covers "route auction enquiries by lot and sale to the right valuer with the catalogue detail attached", because that is not a category. It is just their job.
The enquiry board we built took three days. Triage went from about two hours across three people to about ten minutes, which handed back roughly 1.5 people's worth of daily capacity. They took on more lots the following quarter without hiring.
It removes admin rather than reorganising it
Most software projects move work around. Bespoke work, done properly, deletes it.
Gorringe's, a long-established auction house, produced probate valuations by re-entering client and estate details across spreadsheets and paper, formatting documents by hand and totalling them manually. The tool we built captures the details once and produces the finished, correctly formatted valuation with the totals already calculated. That made valuations 70% faster to produce.
The valuer's expertise did not change. The typing disappeared.
You own it
This is the advantage buyers underrate most, and it is the one that compounds.
With bespoke software, the code is yours. You can host it elsewhere, hand it to a different developer, change it in five years or leave it running untouched. There is no licence to renew, no per-seat fee as you grow, and no vendor deciding to deprecate the feature your process depends on.
With a product, you are renting. The bill grows as you succeed, and the roadmap belongs to someone else.
The cost stops
A product is a permanent line on the P&L that rises with headcount. A build is a cost you pay once.
Ten hours a week of admin is most of a part-time salary, roughly £12,000 to £15,000 a year, every year. A tool that removes it is a few thousand pounds, once. The arithmetic is not close, provided the tool actually removes the ten hours, which is the whole question.
It can be small
The word "bespoke" suggests a two-year programme with a steering committee. It does not have to. Most of what holds a small business back is one bottleneck, and one bottleneck is usually a few days of work.
The disadvantages of bespoke software
These are real. Anyone who tells you otherwise is selling.
It costs more up front
A product costs £30 a month. A build costs a few thousand pounds before it does anything at all. Even when the five-year arithmetic clearly favours the build, the cash does not leave your account in five-year instalments. It leaves now.
If cash flow is tight, that matters more than the total cost of ownership, and no amount of spreadsheet maths makes the first invoice smaller.
It can overrun, and usually does
This is the disadvantage people have actually been burned by, and it is why "bespoke" makes finance directors flinch.
On time-and-materials terms, a quote is an estimate and an estimate is an opening position. Scope creeps because the thing being built did not exist when it was described, so nobody knew what they were asking for. The bill grows and there is no natural stopping point.
Worth knowing what to ask: who pays for the overrun? If the answer is "you do", you have not bought a fixed price. You have bought an estimate.
It needs your time
A product works the day you buy it. A build needs somebody at your end to explain how the job runs, answer questions during the build and tell the developer when they have misunderstood something.
That is real time from someone who is already busy, usually the person who understands the process best, which is usually the person with the least free time.
It depends entirely on who builds it
Products are mediocre in predictable ways. Bespoke software is as good or as bad as the person who wrote it, and you generally cannot tell which you are getting until it is running.
A bad build is worse than no build. It is a tool your team distrusts, works around, and eventually abandons, after you have paid for it.
It has to be maintained
Software does not sit still. Browsers change, integrations break, the business changes shape. Somebody has to host it, patch it and make small changes.
This is not usually expensive, ours is £100 a month and optional, but "we build it and walk away" is not a real offer, and you should be suspicious of anyone making it.
It is the wrong answer to an undecided process
If nobody has agreed how the job should run, software will not settle the argument. It will freeze the confusion in code and make it more expensive to change later.
Fix the process first. Then build.
When you should buy off-the-shelf instead
We say no to work on this basis regularly, so this section is not a formality.
Buy the product when the job is genuinely standard. Accounting, payroll, email, calendars, file storage, video calls. These are solved problems, the products are excellent, and nothing you could commission would be better. If someone offers to build you a bespoke accounting system, walk away.
Buy the product when the process is still moving. New businesses change how they work every few months. Build too early and you pay to encode a process you are about to abandon.
Buy the product when the pain is mild. If the admin costs an hour a week, it is not worth a few thousand pounds. Come back when it costs a day.
Buy the product when compliance is the point. Anything carrying certification, regulatory reporting or audit obligations, such as HMRC-recognised accounting, clinical records or payment processing, is better handled by a vendor whose job is keeping that certification current.
Build when the job is specific to you and the pain is measured in hours a week. That is the whole test. Everything else is detail.
How the disadvantages get solved
Read the list of disadvantages again and notice something: almost every one is a disadvantage of how custom software is usually sold, not of bespoke software itself.
Overruns come from time-and-materials billing. Lock-in comes from developers who keep the code. Long timelines come from scoping a platform instead of a bottleneck. Bad builds come from working off a requirements document instead of watching the job.
Each has a structural answer, and they are the questions worth asking whoever you speak to:
| The disadvantage | What removes it |
|---|---|
| It will overrun | A fixed price agreed before starting, where the overrun is the developer's problem |
| I will be locked in | Owning the code outright, with no licence and no per-seat fee |
| It will take months | Scoping one bottleneck rather than a platform |
| It might be bad | A developer who watches the job before quoting, not one who works off a wishlist |
| I will be stuck paying forever | Maintenance that is optional and cancellable, not access you are renting |
For what it is worth, here is how we answer them. We quote a fixed number of days before starting and if it overruns, that is our problem. Gorringe's was quoted one day. It took three. They paid for one. You own what we build, code and all. Maintenance is £100 a month and you can stop anytime without losing the tool.
We are not the only people who work this way, and you should ask anyone you are considering the same five questions. But if a developer will not put a number on it before starting, that tells you what the relationship is going to be like.
What bespoke software actually costs
Most articles on custom software development will not answer this, so: our day rate is £500, and most builds take three to five days. That puts a working tool between £1,500 and £2,500, with maintenance at £100 a month if you want it.
That is small-tool pricing for small tools, which is what most businesses actually need. A larger programme, several connected tools, integrations across multiple systems, a build that runs in stages, costs more, and anyone quoting a two-week build for a two-month job is either guessing or hoping.
The comparison that matters is not against a £30-a-month product. It is against the ten hours a week the admin currently costs you, which is £12,000 to £15,000 a year, forever.
Frequently asked questions
Is bespoke software worth it for a small business?
Usually yes, if the pain is specific to how you work and costs several hours a week. It is usually not worth it for standard functions like accounting or payroll, where good products already exist. The test is whether an off-the-shelf tool can do the job without your team working around it.
How long does bespoke software take to build?
A single tool addressing one bottleneck typically takes three to five days. Larger builds run in stages over weeks or months. If someone quotes a year for something you described in one sentence, the scope has grown past the problem.
What happens if the developer disappears?
This is why code ownership matters. If you own the code, another developer can pick it up. If the developer holds it, or it only runs on their platform, you are stranded. Ask who owns the code before you sign anything, and get the answer in writing.
Is bespoke software better than off-the-shelf?
Neither is better in general. Off-the-shelf wins for standard, solved problems where the product is mature and the regulation is somebody else's job. Bespoke wins where the work is specific to your business and no product reaches it. Most businesses should run both.
Can bespoke software work with the tools we already use?
Usually, yes. Most builds connect to what is already there, the accounting package, the email system, the calendar, rather than replacing them. Replacing working systems is expensive and rarely necessary.
What is the biggest risk with bespoke software?
Paying for a tool your team then works around. It happens when the build was scoped from a wishlist rather than from watching the job, and it is the reason to care more about how a developer scopes than what they charge.
The short version
The advantages of bespoke software are fit, ownership and a cost that stops. The disadvantages are the up-front price, the overrun risk, the time it asks of you and the fact that quality depends entirely on who builds it.
Most of the disadvantages are solvable by how the work is bought: fixed price, owned code, one bottleneck at a time. The up-front cost is not solvable, and if the admin is only costing you an hour a week, it is not worth paying.
If two or three hours of somebody's day are going into work a tool could do, that is worth a conversation. If not, buy the product. We will tell you the same thing on the call.