Inconsistencies, Inaccuracies and Bad Design: The UK’s COVID-19 App Fiasco
When COVID-19 hit the UK earlier this year, it quickly became clear that something needed to be done to help stem the tide of infections. There needed to be a way to track COVID cases across the population, to figure out what areas had the most infections, and to encourage those in said locations to isolate to minimise further effects.
But gettig that data was not an easy task. After all, the initial attempts to keep the virus under wraps meant it’d spread to much of the world, and the initial lacklustre government response in the UK made it worse, with millions potentially infected after the initial ‘herd immunity’ plan fell through.
So the obvious solution here was to create an app, and track COVID infections that way. By doing this they’d gain an understanding of where people were getting infected, and provide a way to notify those in the area to isolate themselves for a few weeks to prevent further issues.
Unfortunately, this hasn’t exactly go according to plan. No, while the idea of a tracking app for COVID-19 was a good one, the actual implementation leaves a lot to be desired.
Even after the initial disaster that was the government forging their own path and ignoring what worked in other countries. There Apple and Google had worked together to create one model for COVID tracing apps, but the UK government and NHS decided they knew better, and completely ignored it. This was a huge disaster in of itself, and one that resulted in them just using the Apple/Google model anyway, after millions of pounds in wasted work went down the drain.
But that’s not what I’m talking about today. No, I’m talking about the current track and trace app, and how that too is a complete and utter dog’s dinner.
You ready? Let’s begin.
Starting with an obvious issue that someone surely would have foreseen by now.
That being that the UK doesn’t have one app here.
This means going to a different area involves installing an entirely different app if you want to track cases there, even if said area is just a few hours away by car. That’s not ideal to say the least, and is a huge hassle in a region as small as the UK.
What’s more, it also means you can’t rely on muscle memory to navigate said app quickly either. Why? Because despite them all (mostly) doing the same thing, they’re also entirely different in terms of structure and basic UI design.
For instance, here’s the app used in England and Wales:
And here’s the one used in Northern Ireland:
Notice anything? Yep, the latter doesn’t show you how ‘serious’ infections are in the area, and has all the options laid out in a completely different way. So not only do people in different areas now have different data available to them, but it’s also accessible in a different way too.
Additionally, it’s not really made clear how much data is shared between said apps either, if any. So if you don’t have the relevant app for an area, will you get notified if someone who does reports they have COVID? Will someone who visited Scotland from London or vice versa know what the situation is there?
Who knows, it’s not explained at all. We’d hope they’d be connected, but given this is the UK… we’re not holding our breath.
Regardless, you’d at least think the app would work well once you got it right? That if you’re in the right area, the tracing app would be a good way to tell if you may have been infected?
That the information the app would give you would be accurate at least?
The information these apps give you can actually be flat out wrong to the point of being dangerous. For instance, when I checked recently, I got told the level of cases was Medium to High, whereas someone else I just just got told it was Medium.
Problem is, we live in the same town. On the same street in the same town to boot.
So how the hell does this work? Why was I getting one set of results while they’re getting another set?
It’s not personalised. There’s nothing in the app that appears to be able to tell who exactly is using it or how much risk COVID puts them at. Hell, if it was, the results would likely be the other way around, since I’m 30 and the other person mentioned is about twice my age. So no. I would never be the one at greater risk.
Besides, if you click through the results to the official site, the site itself says what my app says; that the area’s risk level is Medium-High. Hence that’s the real value, and the one both apps should say.
Fortunately, this particular issue has been fixed recently. Now both apps say Medium to High as expected.
But who knows how many other people are having similar issues. How many are getting told there are more or less cases in the area than there actually are? That it’s fine to leave the area when it’s actually in lockdown? That another area can be entered fine despite that?
It’s impossible to know, and means anyone using the app will have to check the official site too, in case the app’s flawed information leads them to ignore lockdown rules or wander straight into an area where they’d be at severe risk.
Good job UK governments. You had one job when making this app, and you completely blew it here.
Still, that’s not exactly a rarity where this app is concerned. No, apparently quite a few bits of functionality don’t work as expected here.
For starters, you can’t post negative test results to the app in anyway. So if you were hoping to make things clear that you didn’t catch COVID from being in an ‘infected’ area, then you’ll have to think again. The app flat out doesn’t support that.
It also didn’t used tosupport test results from certain sources either. Did you test take place in a Public Health England lab, an NHS hospital or as part of a national surveillance scheme by the Office of National Statistics?
Well until recently, tough luck. You couldn’t report those results in the app at all:
You also can’t get paid for isolating if the app reports you have COVID:
Covid-19 app users can't get isolation payment
An alert issued via the app will not mean the recipient can claim a payment for self-isolating.
Since those payments are only available to those that receive the results by phone, at least as of right now.
And that’s probably just the tip of the iceberg here. Point being, it’s broken in a lot of ways, which isn’t exactly a good look at some recent IT fiascos interfered here as well.
Excel: Why using Microsoft's tool caused Covid-19 results to be lost
The decision to use a spreadsheet format that dates back to the 1980s has proved to be unwise.
Either way, we’ve now got three distinct COVID apps that barely work, don’t resemble each other in the slightest, give out incorrect information under certain circumstances and apparently don’t integrate well with the government’s systems at all.
How could this possibly get any worse from here?
Well, how about said apps being completely ignored by most of the population?
Yep, most of these apps are not only poorly designed, but also barely used by anyone to begin with.
For instance, the NHS England app has been downloaded 17,679,872 times so far, at least according to the latest statistics here.
That’s not bad in theory, since 17.7 million people is at least a decent number.
But it completely pales in comparison to the number of people in the country as a whole. Indeed, assuming the same app covers both England and Wales, that leaves us with a total population of 59,439,961.
So the app’s been downloaded by abount 29–30% of the total population. Not too great when you look at it like that, is it?
No. Not it isn’t
But hey, it’s not as if we can be too sure of those numbers to begin with.
After all, our totals above assume that every person has exactly one phone/device and has only downloaded this app once in their entire life. That’s not necessarily the case, given people tend to have work and personal phones, or both phones and tablets, or just get new phones every year or two if they’re on a contract/happen to be an Apple fanboy.
Add this to families sharing devices, and well the stats could be anywhere from 20% of the population having the COVID-19 app installed to 35–40% having it. Regardless, it’s still less than half the population, and leaves out many people who are told not to download it by their bosses (like police officers).
Yet the statistical woes don’t end there.
No, it’s also worth noting that many restaurants and pubs (which should display a QR code to check in with this app)… don’t.
This means that depending on where exactly you go, the premise might not even provide a way to track cases there to begin with. Like that Pizza Express I ate at with a friend a few weeks back. Neither of us had to check in there, hence if a case broke out we’d be none the wiser.
And the list of issues just goes on and on. Did you bring a phone or device with the app installed that day? Does the owner of the pub or restaurant you went to believe COVID is a big deal? Have any infected people actually inputted their test results?
If the answer to either of these questions is no, then the data is just as wrong as it’d be if the app didn’t exist at all.
Especially given that last point on the list. Yes, you’re not reading that wrong. This app basically works on the honour system, and assumes that people are going to casually enter test results into the app after they came back positive for the disease.
They won’t. Or they’ll promise themselves they will, then forget to do so later.
Either way, the results are the same; people will test positive for COVID-19, and the others they’re with will be none the wiser.
So what can be done here? What could the UK government do to make their track and trace app actually worth using?
Well to some degree, their options are limited there. They can’t force people to use an app, since they have no idea who even uses a phone or device that can support it to begin with (and it’d be a huge attack on everyone’s civil liberties). They also can’t fully prevent human error, or fix many of the millions of edge cases that a system like this will always have.
But they can do a few things to improve the situation regardless.
First, they can drop this whole ‘every government/country uses a different app’ malarky and mandatee a single app for the entirety of the UK. After all, the UK is one nation, not four. We don’t need to separate the results for England/Scotland/Wales/Northern Ireland at all here.
Instead, come up with one app, declare it’s going to be used country wide and roll with it. That way, people won’t need to juggle multiple apps when travelling, nor will they need to learn a new user interface whenever they enter a new area.
Secondly, fix the systems that underly these apps. They shouldn’t be giving out incorrect or cached data, nor should they be refusing test results or payments based on random technical issues.
Fix that, and you’ll at least be able to guarantee the app works the way it should do for most people.
Next, rethink many of the core principles behind how data is provided to the app here. Instead of making people enter test results codes, actually connect it to the NHS databases/systems instead.
That way, whenever someone is tested positive for COVID, anyone who’s been in the same area will be automatically notified they’re at risk without having to rely on the affected person having to provide the data themselves. It’d be a much better setup, and one that’s actually somewhat reliable here.
It may also be a good idea to reward people for using the app too. Pay those who download it, or offer them rewards for doing so, like cheaper food at restaurants or items in shops. That way, a larger percentage of the population would be encouraged to make use of the system, not just the 30% that does so right now.
Finally, enforce rules for businesses better too, to get them to use the QR codes for check ins more often. This will make sure fewer companies flout the rules, and get things working much better than they do at the moment.
By doing this, the system will at least be half decent here, and it’ll give a much better overview of how COVID-19 is spreading within the country, as well as how to stop it.
Update: Since this story was published, another few obvious issues have now come up.
Firstly, the app is apparently sending incorrect, ‘phantom’ notifications saying you’ve been exposed to COVID-19. This has been going on for 3 weeks, and can apparently affect people who’ve never even left the house.
Like myself. Apparently despite not checking in anywhere, the app randomly decided I may have exposed to the virus, and warned me about it.
This then led to two more problems:
- When I clicked the notification, I couldn’t see it in the app. You can’t see this info then, and viewing the app from the notification deletes it from your history too.
- Apparently instead of fixing the bug, the app developers decided the best solution was to send a second message basically saying “Don’t worry, everything’s fine”
It’s ridiculous to say the least.