‘appy now?

I’ve been thinking a lot about public sector apps.

Tom Loosemore’s We’re not ‘appy. Not ‘appy at all blog post celebrated its 10th birthday earlier this year. Like many others, I have cited this to add weight to arguments against building apps in the public sector.

More recently, Terence Eden’s blog post about mobile apps in government questioned the received wisdom that public sector apps are always a bad idea.

“Develop a GOV.UK App” is the top priority for the GOV.UK programme and that team recently recruited a new Head of Product.

And about 6 months ago I became Head of Product for the NHS App. So I really should have an opinion on the use of apps for public sector services.

Chatting apps

So I ran a session on public sector apps at Product for the People (London, April 2024) as a temperature check to see what attendees thought about public sector apps.

I started with an unfair question — “Public sector apps: yes or no?” and forced people to pick a side. No nuance. No sitting on the fence. There were 9 votes for ‘yes’ and 7 for ‘no’. We repeated this vote at the end of the session— more on that later.

Then, using Liberating Structures’ 1–2–4-All model, we debated the question:

Is there ever a case for public sector apps? If yes, what are the conditions? If no, why?

The case for public sector apps

We discussed a bunch of ways in which apps can offer a better experience than websites:

• Notifications. Have you ever accidentally allowed a website to send you browser notifications? It’s awful.
• Messaging people through an app is less expensive than sending SMS.
• Offline access coupled with on-device data storage allows users with variable internet access to complete their tasks.
• Biometric login is quicker and more convenient for most people (and more secure?) than entering a username and password.
• Digital wallets can make it easy to share tickets, vouchers and similar documents, for example a barcode to pick up a prescription.
• Integration with other native apps on your device, such as maps and calendars.
• Accessibility can be enhanced by hooking into devices’ in-built accessibility features, such as dark mode.

Yes, some of these things can be done to some extent on mobile websites.

Apps are also potentially convenient for services that people use often. People are unlikely to devote precious space on their device for an app they use once a year, but they might for something they use once a month.

The case against public sector apps

We also discussed many reasons why apps are inferior to websites:

• An app will be more expensive to build and maintain than a mobile website.
• You usually have to run a browser version of the service anyway, so the benefits gained from also having an app are usually marginal.
• You have to maintain multiple versions: iOS, Android and (usually) browser.
• It’s difficult to recruit and retain mobile app developers.
• There is a bigger barrier to first-time use because users have to download the thing from an app store.
• Apps take up space on people’s devices.
• Users with older devices that can’t upgrade to the latest operating systems (who may well be the users you most want to reach!) are more likely to be able to access a modern website on their browser than an app. Alternatively, maintaining your app’s compatibility with older operating systems is expensive.

Digression 1 — on that Tom Loosemore blog post

It should be noted that Tom’s blog post does not say you should never build an app. “We are not ‘banning’ apps outright.”

It was a challenge to the belief held by many at the time that an app was a better way to deliver a digital service than a website:

The benefits of developing and maintaining apps will very rarely justify their costs … departments should focus on improving the quality of the core web service.

But over the years the received wisdom in the public sector has become more “thou shalt not build an app” rather than “thou should probably not build an app”.

Digression 2 — on the nature of the NHS App

The NHS App’s design lead, Mike Gallagher, likes to remind us that we should think of the NHS App not as an app, but as a “all of the things we can do for you, and show to you, once we know who you are”.

The services in the NHS App — request a repeat prescription, book a vaccination appointment, check if you need urgent help — are not solely available through the NHS App. You can do these things through other appps, in your web browser, over the phone or even in-person.

In fact, many of the services in the NHS App are exactly the same as the ones you access through your web browser. We strip off the website header and footer and display the rest inside the ‘shell’ of the NHS App.

There is also a browser version of the NHS App that you can access via the NHS website. Last month, 4.5% of session were via this route. Is it an app? Is it a website? Is it a channel? Answers on a postcard, please.

The votes are in

At the end of the session I repeated the vote on the question “Public sector apps: yes or no?”

There were 16 votes for ‘yes’ and 3 votes for ‘no’ (reminder — those numbers at the start were 9 ‘yes’ and 7 ‘no’). Quite a few of those ‘yes’ votes were half-hearted, hands-kinda-raised votes but they still count.

Not only did we see a swing from ‘no’ to ‘yes’, we somehow had 3 more attendees at the end than at the start.

My view on apps

I have a pragmatic, Centrist Dad view on the question of public sector apps.

It’s very easy to make a bad app. It’s also easy to make something half decent and then see it rot because of a lack of foresight, maintenance and investment. Both scenarios represent a waste of money and a lost opportunity to do something simpler and actually useful.

Good public sector apps are possible, but they are expensive and should only be deployed when the benefits clearly outweigh the costs.