When is custom software worth the investment?
You can't really ask us. We build custom software, so the answer is always yes. Right?
No. In fact, we often see that custom work shouldn't be the first step. Custom software is a sizeable investment, and you only earn it back if the problem is big enough and no smaller solution exists. In this article you'll read how we weigh that up. Even if the outcome is that you don't need us yet.
01 / First this
Four things you can try before you have something built.
Many bottlenecks seem to call for new software, when a smaller step would be enough.
Get more out of what you already have
ConfigureMany packages can do more than you use. An afternoon with someone who knows the package well sometimes saves a whole project.
Let systems talk to each other
IntegrateDo employees retype data from one system into another? Then an integration is often enough, and it is much smaller than a new system.
Look at the process first
SimplifyA cumbersome process doesn't get better when you automate it. Sometimes a step turns out to be unnecessary as soon as someone asks why it exists.
Look for an off-the-shelf package
BuyFor accounting, CRM or scheduling there is almost always something to buy. If your process isn't special, adapt the work to the package, not the other way round.
When custom work does make sense.
There are situations in which off-the-shelf software falls short structurally. Not once, but every day. We usually see one or more of these signs:
- Employees keep their own spreadsheets alongside the system, because the system doesn't do what they need.
- The same data is entered several times, and that is where things go wrong.
- Your way of working is exactly what sets you apart from competitors, and no package supports it.
- With every growth step you need just as many extra people for the admin.
- You want to offer customers something digital that doesn't exist yet.
Recognise one or two of these? Then it's time to do the maths.
02 / Worked example
What does the problem cost now?
A simple way to start. The figures are an example; fill in your own situation.
- How many employees do this work?
- 5
- How many hours per week does the manual part take per person?
- 4 hours
- How many working weeks per year?
- 46
- Hours per year
- 920 hours
- Labour cost per hour
- € 45
- Cost of the problem per year
- €41,400.00
That amount is your upper limit. Software that solves this problem should not, over a few years, cost more than the problem itself. Otherwise you are better off carrying on as you work now.
Hours are not the whole story, though. Mistakes cost money too: a wrong invoice, a missed appointment, a customer who drops out because they wait three days for an answer. Those are harder to count, but often bigger. And sometimes it isn't about saving, but about something you can't do now. A new service, a new market. Then you calculate with what it brings in rather than what it saves.
The costs you don't see on the quote
To be fair: custom work also costs money once it is finished. Hosting, security updates, small changes when your way of working changes. Count on that from the start. Hosting, updates, monitoring and back-ups are covered by our hosting, management and support. Changes are done on an hourly basis or through a retainer. Include those costs when you work out what the software will cost you over a few years.
The biggest risk, by the way, isn't the cost. It is software that nobody uses. That is why we start small, with a first version that does one thing well. You can read how that works in [From idea to MVP](/en/blog/from-idea-to-mvp-how-we-reach-a-first-working-version).
03 / Weighing it up
Does custom work pay off for you?
Custom work pays off if
- the problem comes back every day and affects many people;
- you can put the cost of the problem in hours or euros;
- your way of working sets you apart and needs to stay that way;
- you have already tried integrating or configuring things better;
- you expect to grow and the workload grows with you.
Choose something else first if
- an off-the-shelf package gets it eighty per cent right;
- the problem comes up a few times a year;
- the process itself is not yet clear;
- nobody internally has time to think along and test.
Frequently asked questions
What does custom software cost on average?
An average says little. An integration between two systems is something very different from a complete platform with users, roles and dashboards. A first, clearly scoped version, such as an integration, dashboard or simple web application, often starts at a few hundred to a few thousand euros. An extensive platform quickly runs into the tens of thousands. We split every project into phases and agree a fixed price per phase in advance. That way you always know what you get and what it costs before you build further.
Can I start small?
Yes, and that is usually what we recommend. A first version that solves one problem quickly shows whether the investment pays off before you build further.
What if I want an off-the-shelf package later after all?
We agree ownership of code and data in advance. Your data is yours and goes with you if you switch.
Can you help with the business case?
Yes. In a short advisory engagement we set the options side by side, including the routes where you don't need us.
Not sure whether custom work is the right route?
Tell us where things are getting stuck. In a first conversation we look together at whether custom work pays off, or whether an integration or a better-configured package is enough.