How Do Successful Brands Use Mobile Apps?

A phone's home screen is the most contested space in marketing, and nothing stays there out of politeness. When we look at how brands use mobile apps, the ones that last have usually picked a single job and done it well enough that deleting the app would cost the user something they'd miss. Everything else gets a download, a look, and a swipe into the bin.

An app that survives deletion does one job properly and quietly becomes part of someone's week.

We've spent years auditing, rebuilding and launching brand apps across wellness, drinks, property, fitness and travel, and the pattern holds across all of them. The successful brand apps aren't the ones with the longest feature list. They're the ones where someone opens the app on a Tuesday morning without thinking about it, because it has become part of how they do something they already do.

This article walks through what brand apps are for, which jobs they do best, how to get found, how to onboard people in the state of mind they're actually in, and how to tell whether any of it is working. Most of it comes from projects we've run and decisions we've had to argue for, including a few we got wrong first time.

What a Brand App Is Actually For

We were asked to audit a genetics wellness app that wasn't holding on to its users. The brand promise turned out to be sound. It offered people a real inside look at how their body works, how it's ageing, and how to improve it, and that promise was compelling and well constructed. Our audit found nothing wrong with it.

The product had been built largely by a developer team, and along the way the storytelling and narrative had been stripped out in favour of functional delivery. People got their data. They got no sense of what that data meant about them. So a product with a genuinely interesting promise behaved like a reporting tool, and nobody opens a reporting tool twice a week.

How Brands Use Mobile Apps Well

An app is the one place where a brand gets to behave like itself over and over again, in someone's pocket, without buying the attention each time. That repetition is where the commercial value sits. Narang and Shankar found that users of branded apps spend 37% more than non-users, according to Narang and Shankar, 2019, and the figure makes sense when you think about what an installed app actually is. It's permission to be present in someone's day.

The question to answer before any design work starts is simple. What does this app let someone do that they'd notice losing?

The Jobs Brand Apps Do Best

We worked on a water tracking product where the entire job was logging a glass of water. Small, repetitive, over in two seconds. That's exactly the kind of job a phone does better than anything else, because the phone is already in your hand when the moment happens.

The jobs that suit a brand app tend to share that shape. They're short, they repeat, and they happen away from a desk.

• Logging something small and frequent, like water, a workout or a meal
• Holding information that has to work with no signal, like a ticket or a boarding pass
• Reordering something the person already knows they want
• Sending something to a friend in a few taps
• Proving who you are to staff in a shop, a gym or a hotel

Our position on the travel category is firm on this point. Tickets, boarding passes and itinerary details are static information that isn't going to change, and any travel product that fails somebody in an offline situation is the product that collects one-star reviews. Those are the moments where the app either does its job or burns the brand.

What these jobs have in common is that the user already has the need. The app doesn't have to manufacture a reason to exist, which is the hardest thing for a brand app to do and the thing that sinks so many of them.

Choosing Fewer Features and Doing Them Properly

Teams usually arrive with a feature list rather than a job. The instinct is understandable, because a long list looks like value for the budget. What it buys in practice is a product where eight things half work.

Our advice on travel apps makes the trade-off concrete. Offline support sounds like a feature nobody should argue with, and we still tell teams to leave it out unless it's genuinely critical to the product, because a partial offline mode that fails at the gate does more damage than no offline mode at all. Users are far less likely to come back after a bad experience, and that single failure can define the whole brand in someone's memory.

A feature that half works is worse for the brand than a feature that isn't there.

So the better use of a budget is to pick a small number of things, build them to a standard you'd be happy demonstrating in front of a room, and keep the rest of the money for making them faster and clearer. We'd rather ship three features people trust than ten they learn to work around.

Write down the one thing your app has to do perfectly. If a proposed feature makes that thing slower, slacker or harder to find, it goes on a later list.

Getting Found: The App Store Listing as First Experience

A client came to us with a gifting and wishlist platform and was hesitant to spend anything on the app store listing. The product was social and group-oriented, so the thinking was that word-of-mouth referral would carry it. Our view was that referral and the listing do different jobs. People really need to understand what the product is about from the very first screen, and the icon, the name, the category, the copy and the screenshots are what let them self-select in or out before they ever download.

The aim is for everyone who downloads to arrive already pre-sold. Nobody benefits from a download followed by the realisation that the app was never for them, and Think with Google report that 21% of users give up on an app if they don't get it right away.

On an anonymous messaging app we built, registration came before anything else, so we couldn't drip-feed the value inside the product at all. Everything we'd normally reveal through use had to be front-loaded into the listing and the screenshots instead, and that meant a lot of out-of-app work we'd budgeted nothing for at the start.

The order people read a listing in is worth designing against.

1. The icon, which says what kind of thing this is
2. The name and subtitle, which say who it's for
3. The first two screenshots, which carry the whole pitch
4. The description, which only committed readers reach

Onboarding That Matches the User's State of Mind

On a concierge app we built for residents moving into a new block of flats, we made a deliberate decision not to surface all the building information at once. We had recycling locations, emergency procedures and local area guides ready to go. Handing all of it over on day one would have been faster to build and worse to receive.

The reason was emotional. Someone who has just moved is in a highly varied state. Some of these residents had just bought their first property and were elated. Others were moving after a separation or a divorce. People in those states aren't mentally receptive to a lot of information, so we chose to unburden them and release content when it would actually be useful.

The effect was a markedly more human experience, and it cost nothing extra in build. It was a scheduling decision and a content decision, made with the person's first fortnight in mind rather than the content management system.

Designing for Habit, Not Achievement

On a fitness app for weight loss, we changed what the gamification rewarded. It had been built around achieving results, which sounds right and quietly punishes everyone whose results are slow. We moved it to showing up every day, being better than you were yesterday, and taking small steps.

That shift matters because the outcome is the thing the user controls least. Weight moves unevenly. Turning up is something anyone can do on a Wednesday, so rewarding consistent engagement and continued improvement keeps people in the product long enough for the outcome to arrive.

What Habit Design Looks Like

In practice it means the streak counts the open, not the result. It means the app has something warm to say on a bad week. And it means the reward loop fires on the small action, so logging a single glass of water or a ten minute walk feels like progress rather than a rounding error.

The gap between the average app and the ones that hold on is enormous. MoEngage report that the top 20% of apps retain users five times as long as the average app, and habit design is a large part of what separates them.

Pick the smallest action a user can take in your product and make that the thing you celebrate. Save the big milestones for people who get there.

Loyalty Mechanics That Reward More Than Spending

We worked on a product for the drinks industry that let people buy and send drinks to friends at a bar or an event when they couldn't be there in person. The loyalty mechanic was built around gifting rather than personal consumption. The more you bought and the more you gifted, the more you benefited as an individual user.

That structure does something a points-per-pound scheme can't. Generosity towards other people also rewards the giver, so the act that grows the product and the act that earns the reward are the same act. Nobody has to be bribed into inviting a friend, because the invitation is the product.

The best loyalty mechanics reward the behaviour that grows the product, not just the size of the receipt.

It's worth asking what else your app could reward. Reviewing a product, showing up to a class, referring a friend, completing a profile that makes the next visit better, turning up at a physical store. Those behaviours all have value to the brand, and most loyalty schemes ignore every one of them in favour of counting spend.

Spend-based tiers also have an uncomfortable side effect. They tell your most frequent, lowest-spending customers that they don't matter much, which is rarely what the brand believes.

Giving Data a Story and the User an Identity

Going back to the genetics wellness app, the fix wasn't new features and it wasn't a visual redesign. We brought the narrative and the storytelling back in, and we gave people a real sense of identity within the product. Retention and engagement improved on the back of that alone.

The mechanism is straightforward. A number on its own asks the user to do the interpretive work. A number inside a story tells them what it says about them, what it means for the next few weeks, and what to do about it on Thursday. Those are the same data points, doing a completely different job.

Turning a Figure Into a Sentence

So instead of a dashboard, the product speaks to the person. Rather than showing a score and a graph, it says where that score puts them, how it's moved since last time, and which single thing would move it next. That framing is what makes someone feel the product is talking to them as an individual rather than processing them.

This applies far beyond genetics. Skincare routines, sleep, training load, water intake and spending all arrive as raw figures, and all of them are more useful when the product has an opinion about what they mean for this particular person.

Take your main results screen and read it aloud as a sentence about the user. If it only reads as a list of values, the story layer is missing.

When a Brand App Becomes a Community

On a concierge app project for a property developer, we ran focus groups with social housing tenants to work out what actually creates a sense of community. The answer was familiarity. People feel they belong when they accumulate shared personal knowledge about each other, including names, occupations, children, pets and hobbies. The sense of belonging turns up naturally the minute you start surfacing those data points between people.

That finding changed the product. What started as something close to a building directory and a maintenance logger became more like a social network, with a social feed at the heart of it. Residents posted content, asked neighbours for help, and followed updates from the concierge team and local charities. The building management features stayed, as a secondary layer underneath.

The design decision underneath all of that was to encourage the conversations that surface those personal details rather than automating the interactions away. Digital self-service is easy to build and it quietly removes every reason for two neighbours to speak. So we put the mundane building tasks in the app and used the app to push people towards each other face to face.

Community features work when the app is the reason people meet. They fail when the app becomes the place people go instead of meeting.

Beyond the Phone: Watches, Wearables and Offline Moments

The water tracking product we worked on started as a phone-only app, and then came the addition of an Apple Watch component. The native app was tactile and engaging, and it took full advantage of a large screen. On the watch we had very limited real estate, so we had to rethink the navigation and rethink how we were handling the interactions from the ground up.

We didn't achieve the same fluidity on the wrist, and we were honest about that. We still got meaningful features into the watch interface, and the lesson we carried forward is that a watch needs its own design logic rather than a shrunken version of the phone layout.

Offline moments deserve the same rethink. There's a very strong correlation between getting offline functionality right, particularly in travel, and the retention, engagement and brand perception that follow. A boarding pass that won't load in a basement is the moment a user decides what your brand is.

Physical spaces matter too. MobiLoud, 2025 report that 74% of consumers worldwide use retail apps while shopping in physical stores, which makes in-store behaviour a design problem rather than a nice extra.

The Metrics That Tell You It Is Working

Downloads tell you about your marketing. They say nothing about whether the product works. The numbers we care about start the moment someone opens the app for the first time and continue for about a month.

The first is whether people reach the thing the app is for. On the water tracking product, that means logging a glass. On a loyalty product, it means a first scan at the till. Define that one event, then measure how many new users reach it in the first session, because everything downstream depends on it.

Downloads measure the marketing. Day 30 return rates measure the product.

The second is rhythm. For a daily logging app, four sessions a week is health and one is a warning. For a travel app, the right rhythm is a cluster around a trip and silence in between, and judging that against a daily app's pattern will tell you the wrong story entirely.

The third is what people say in the store. Review text is the cheapest qualitative research available, and the one-star reviews in particular tend to name the exact moment the product failed somebody. We read them before we read a dashboard.

Where Brand Apps Usually Go Wrong

On that same concierge project for the property developer, the client wanted two things. One was a full-service building management utility. The other was a community-building social tool. We recognised early that those two goals were at odds with each other, because the first removes human contact and the second depends on it.

Holding both ambitions without resolving them is the most common failure we see in brand apps. A product that's trying to be two things makes a worse version of each, and the user feels the confusion as soon as they open it. On that project we resolved it by running the focus groups, choosing community as the heart of the product, and demoting building management to a supporting layer.

The Other Recurring Faults

The rest of the pattern is familiar. Apps that repeat the website with a bottom nav bar. Products where the storytelling got engineered out during the build, like the genetics wellness app before we audited it. Store listings treated as administration. Onboarding that empties the whole content library onto someone in their first thirty seconds.

All of these are decisions rather than accidents, which is the good news. Every one of them can be reversed by the same team that made it.

Conclusion

The brand apps that work have a few things in common, and none of them are expensive. They pick one job that happens often and away from a desk. They do a small number of features to a standard the team is proud of. They say clearly, in the store listing, who the product is for. They release information at the pace a real person can take it in. And they reward showing up rather than only rewarding results.

The patterns across the projects in this article point the same way. Our audit of the genetics wellness app found the promise was fine and the storytelling had gone. The concierge app got better when we stopped automating away the conversations that create familiarity. The fitness app held on to people when the gamification started rewarding a Wednesday rather than a finish line. In each case the change was about what the product understood about the person using it.

If you're planning a brand app, rebuilding one, or looking at a retention curve you don't like, we're happy to look at it with you and say what we'd do. Let's talk about your brand app.

Frequently Asked Questions