Can’t we just build the app ourselves?
With AI tools, a municipality can produce something that looks like a finished app in no time. So why pay for a platform? It’s a good question, so we go through the arguments for both paths.
Municipalities sometimes ask us: “Do we really need a vendor? Can’t we just build this ourselves now that AI does the work?”
It’s a fair question, and it deserves a proper answer. So let’s start by looking at the arguments for building it yourselves.
The arguments for building it yourselves
1. It has never been this fast. AI tools have lowered the bar considerably. In just a few days, a skilled person can put together an app with news, events and a map. A few years ago, that would have taken months.
2. Full control. You decide everything yourselves: design, features and timeline. You don’t have to wait for someone else’s roadmap.
3. No licence cost. A licence shows up in the budget and is easy to question. If you build it yourselves, that line disappears, and the app seems to be free.
4. No lock-in. You’re not dependent on a vendor’s prices, terms or future.
5. In-house expertise. You build your own digital capability instead of buying it.
These are good arguments, and we understand why they’re tempting.
What you don’t see
A prototype or an AI-coded app shows roughly ten percent of what a citizen app is. The rest lies beneath the surface. It’s a bit like admiring the nice wallpaper in a house you’re about to buy without checking whether the foundation is solid and properly drained.
“It’s quick to build”, but an app is never finished
The expensive part today isn’t building the app. It’s maintaining it. Every year brings new versions of iOS and Android, the App Store and Google Play change their requirements, and security updates keep coming. On top of that, accessibility requirements and WCAG have to be met, and the GDPR questions don’t go away.
In a self-built solution, all of this often ends up with one or two people who know the code. If they leave, the municipality is left with an app nobody knows how to maintain anymore. AI makes writing code faster, but it doesn’t reduce the responsibility for that code. Quite the opposite: when more code is written faster, more people end up having built things nobody really understands.
“No licence cost”, but the cost moves somewhere else
The licence is visible in the budget. The costs of an in-house solution are harder to see, because they’re spread across IT, communications, tools and various consultancy agreements. So make an honest comparison over 3–5 years that includes everything:
- developer or consultant time to build, maintain and further develop the app
- operations, hosting and monitoring
- security work and incident management
- AI tokens and third-party services
- support and training for admins and editors
Calculated over the whole lifetime, a platform shared by many municipalities is usually cheaper. That’s because each municipality pays for a small share of the development.
“Full control” also means full responsibility
Control sounds great until something goes wrong. When the app is used for crisis information, it has to cope with every resident opening it at the same time, and push notifications have to arrive immediately. That requires operations, monitoring and on-call readiness.
Control also means legal responsibility. The municipality is then itself responsible for personal data processing, security and agreements with all subcontractors.
“Tailor-made” quickly gets lonely
Your own app can be shaped exactly to the municipality’s wishes. At the same time, every integration becomes a small project of its own, for example with waste collection, fault reporting and news from the website. And the project has to be redone whenever the supplier of that system changes something.
On a shared platform, the integrations are already built and maintained. When one municipality drives a new core feature, everyone else gets it too. That gives a municipality access to solutions it would have struggled to find the budget to build on its own.
“No lock-in” is a requirement you should make
This is probably the strongest argument for building it yourselves, and the concern is justified. But the answer, we believe, isn’t to build everything yourselves. The answer is to set the right requirements for your vendor. The municipality should own its content, its data and its users. It should also be possible to take all of that with you if you choose a different solution. A serious vendor should be able to promise that in the contract.
The app isn’t what matters – the way of working is
This is where the biggest difference lies. The value of a citizen app isn’t in the code, but in how the work around it is organised. An app only comes alive when associations, culture and leisure, libraries and local players can publish their own content. That requires roles, permissions and moderation. Otherwise everything lands on the communications department, which quickly becomes a bottleneck.
On top of that, residents have to download the app and keep using it. That’s harder than building it. Experience of what actually works at launch and over time isn’t something you can code your way to, with or without AI.
We work to share knowledge between municipalities and have extensive experience from running app projects in almost 100 municipalities.
When do we think it can make sense to build it yourselves?
Sometimes it can make sense to build it yourselves, for example for an internal tool with a clear and limited purpose, for a campaign that will only be used for a short time, or in a large organisation with its own development department that can maintain the solution in the long term.
A citizen app, however, is not a side project. It’s infrastructure, just like the website and the e-services, and it has to work a few years from now and in a crisis too.
Five questions to ask before you decide to code it yourselves
- Who will maintain the app in three years, and what happens if that person leaves?
- What will the app cost in total over 3–5 years, when all costs are included?
- Can it handle a heavy load of users during a crisis?
- Who besides the communications department should be able to publish content, and how will that work?
- How do we get residents to download the app and keep using it?
Our recommendation
It’s no secret that we’re not impartial on this question.
You know your own organisation best and can judge for yourselves whether it can maintain an app in the long term. To be completely honest, there are municipalities that manage this without us. But we wouldn’t be doing what we do if we didn’t believe we do it best.
Finally, we’d like to highlight a thought from Nacka Municipality. When it comes to subcontractors for things like IT systems, they have a clear principle: “We should be the best at being a municipality.”
We think the same way. We want to be the best at citizen apps, so that you can put your energy into being a municipality.
