How to (and How Not to) Onboard New Users to Your Mobile App

Lessons from TikTok, Houseparty, Spotify, Bunch, Calm and twenty other top consumer apps

Raoul Boström
Jan 8, 2021 · 19 min read

For your app to make it big — really, BIG — you have to figure out how to turn a flood of app installs into a long-term user base.

You need to turn a user’s often-vague intent (“this looks kind of interesting”) into a genuine belief that your app has a role to play in their life —that it will make them fitter, happier or more productive, in at least some small way.

You need to ‘onboard’ your user. And you need to do it quickly. Because you’re doing it in an environment where a user’s attention is measured in seconds, not minutes.

Looking at the top consumer mobile apps — those that already have huge user bases and those that are getting there fast — we can see common patterns in how they onboard users. In how they get the user through those first few screens. How they get the user smiling and nodding along. How they get the user to that moment where they know the app is for them.

User onboarding is, says onboarding guru Sam Hulick, about “guiding the uninitiated all the way to their own personal promised lands.”

How do the top apps do this?

Well, they …

1. … get the user thinking: this is what I’ve been looking for 💡

2. … get the user taking action, doing something valuable, in the first few screens 🎬

3. … personalise the user’s experience, with minimal work by the user 🎯

I’ll cover each of these in turn.

1. The top apps … get the user thinking: *this* is what I’ve been looking for 💡

They orient their users right from the app’s first screens, setting expectations for what’s to come.

If a user has installed and opened an app, they likely have some expectations about how the app will improve their life in some small way — how it will strengthen their connections with their friends, or how it will make them laugh, or how it will help them focus.

The first screen of the app reinforces these expectations. Done well, it generates excitement about how the app is going to deliver on their expectations. It gets the user thinking “this is what I’ve been looking for”.


Twitter, Yubo, Tasty: concise, ambitious promises

Twitter and friend-discovery app Yubo make promises directly to the user, telling them exactly how the app will improve their life in some small way. They make ambitious promises and set clear expectations. (No mention yet of the app’s features or how it works—there’s plenty of time for that later.)

Buzzfeed’s Tasty cooking app injects a little empathy and humor. It shifts to the first person as it sets expectations for what the user will get out of it: delicious food that (even) they can cook.

Left to right: Twitter, Yubo, Tasty

Reddit, Bunch, Depop: promises, animated

These apps — which feature predominantly visual content — use video to show the user what to expect as they get deeper into the app.

Reddit shows off some of its greatest hits (which just can’t be captured in words). Group gaming app Bunch shows the user the fun they’re about to have. And street fashion marketplace Depop establishes the aesthetic embraced by so many of the users and the products on the app.

Left to right: Reddit, Bunch, Depop

Fabulous, Calm: putting the user in control

Habit building app Fabulous takes a slightly different approach, offering a list of ways it can improve the user’s life, depending on their goal. It groups them all under ‘things you can control’.

Meditation and sleep app Calm adds a twist to this approach. It converts the list into a question and asks the user to select their goal on the app’s very first screen.

It goes beyond telling the user that they can control more than they thought. It actively places control in the hands of the user. The app responds to the user’s choice, tailoring the rest of onboarding to the user’s selected goal.

Left to right: Fabulous, Calm

Regardless of the format, all these apps have one thing in common.

They are, within the first few seconds of the user opening the app, making promises about how they are going to improve the user’s life. They’re re-enforcing ideas the user already has and building the user’s confidence that they’re going to get what they came for.

Sure, some users might decide ‘actually this app isn’t for me’, that it’s not what they came in search of. Users opening Calm for the first time who didn’t have any of those goals will almost certainly abandon it after the first screen. And that’s okay. Those are users that would have dropped off anyway as more of the app was revealed. Maybe they would have logged a few more screens on their first session, but the chance of converting them to active users is pretty close to zero.


Facebook, Instagram, Snapchat: taking it for granted

Some apps dispense with any attempt to orient the user, to set expectations about what’s to come — Facebook, Instagram and Snapchat among them. The first screen of each of these apps is, pretty much, a logo and two buttons. Sign up. Or log in.

Left to right: Facebook, Instagram, Snapchat

Now, these apps are so well known and so much a part of our culture that most new users already know what to expect of them. But they also passed on the chance to get the user excited about what’s coming next. (Snapchat, at least, uses that first screen to establish the tone of the app with its now-iconic visual design).

Noom, Nike Run Club, MyFitnessPal: missed opportunities

Other apps fail to make an immediate promise to their users. But they do this without having instant credibility with new users that social media giants can assume.

Weight loss app Noom lacks any promise at all or a preview of what’s to come. Nike Run Club and diet tracker MyFitnessPal ask the user to swipe through to see this, for no obvious reason. (When the user does swipe they see a lot about the features of each app. And very little about benefits the app will deliver to their lives).

Left to right: Noom, Nike Run Club, MyFitnessPal

While these are popular and well-rated apps, many users won’t know exactly what to expect when opening them for the first time. That first screen was a chance to give them something to believe in. Something to get excited about. Something to give them energy for the rest of the onboarding flow. Something to keep them going as they’re asked to fill in personal details, create an account and do a bunch of other stuff that can feel like quite a lot of effort without a clear reward at the end of it.

KEY TAKEAWAYS:DO:1. Make a concise, ambitious promise about what your app will deliver for your users. Focus on how your app will improve the user's life in some small way. (Like Twitter, Yubo and Tasty)2. Animate your promise. Use video to show what's in store for the user, especially if your app is highly visual. (Like Reddit, Bunch and Depop)3. Put the user in control, from the very first screen. Let the user select their own goal and tailor the rest of onboarding to serve that goal. (Like Fabulous and Calm)DON'T:1. Take it for granted that the user knows how their life will benefit from using your app. (Like Facebook, Instagram and Snapchat)2. Miss opportunities to get the user excited about your app, to get them believing that your app has a place in their life and to give them energy for the rest of the onboarding flow. (Like Noom, Nike Run Club, MyFitnessPal)

2. The top apps … get the user taking action, *doing* something valuable, in the first few screens 🎬

In his overview on the importance of the ‘first mile of product’, Behance Founder Scott Belsky advocate’s doing things for your users. He formulates the rule “DO>SHOW>EXPLAIN”:

the absolute best hook in the first mile of a user experience is DOING things proactively for the user.

Even better than doing things for your users is getting your users doing things for themselves.

Former Pinterest growth lead Casey Winter wrote about the importance of getting user to key actions — the action that “signifies the user is receiving enough value to remain engaged”. Facebook famously had a key action of getting users connected to 7 friends in 10 days. Pinterest focused on getting users ‘saving’ (pinning images to their board).

In a lot of apps, these key actions happen at the end of the user’s onboarding. They are a final step after the user has already spent several minutes and quite a bit of mental energy signing up, providing a whole bunch of personal info, granting the app phone permissions etc.

Today we see more and more apps getting users to take key actions as part of their onboarding.


TikTok, Shine, Etsy: do it early, do it often

TikTok gets users watching videos and swiping up through the content feed — even liking and following creators — within seconds.

By the time the user hits the sign up screen, they’ve already watched several of TikTok’s addictive short videos — and experienced what the app will bring to their lives. They’ve seen the promise that TikTok makes on it’s first screen (“Videos to Make Your Day”) and, within seconds, they’ve seen that promise kept.


Meditation app Shine gets users listening to their first meditation and writing their first journal entry during onboarding — and then acknowledges the user’s success at completing these. The user is on their way to becoming a more mindful person — and they haven’t even reached the app proper!


Etsy gets the user ‘favouriting’ products as part of their onboarding. The app starts the user off with a limited, curated selection of products across key categories (showing the app’s product range at its best).

It asks the user to favourite a few products — getting the user performing a key app action — and then carries over this action into the user’s home feed.

Landing on the feed for the first time, the user sees items they’ve already said they like, rather than one informed purely by algorithmic guesswork.


Each of these apps gets the user focused on one key action at a time, guiding their attention and mental energy toward completing that action. An action that will help the user start to get what they wanted from the app. An action that they will repeat as they become long-term users of the app.


Nike Run Club, Houseparty, Snapchat: splitting the user’s attention multiple ways

The apps above stand in contrast with those that split the user’s attention between multiple possible actions.

Nike Run Club has a prominent and inviting yellow button to start a run. But for a first time user of the app, this button is initially hidden by two modals.

The first modal asks for permission to see the user’s location. This is a needed permission, but one that could be better timed (more on notifications in the next post in this series).

The second modal introduces a concept (the ‘guided run’) that is pretty self-explanatory and likely relevant to only a small proportion of app users (and those users would likely have sought this out for themselves).

Once the user gets through those modals, there’s a further distraction — a red notification indicator in the top right. Now the user needs to start thinking … Is this something important? Something I need to read now? What happens if I just ignore it?

Simply by attracting their attention, the indicator distracts the user. And it does it just as they’re getting to that key action of starting a run — so close they can almost click it.

(What’s behind the indicator? Another permission request (this time asking to send the user app notifications) and a pretty vanilla welcome message. Nothing valuable to the user.)

Nike Run Club

While Nike Run Club steps users through a series of distractions one-by-one, Snapchat and Houseparty kind of just drop the user onto a screen with multiple indicators and callouts competing for their attention.

Snapchat historically made some pretty bold ‘first mile’ product choices. But the app did, in the early days, try to focus the user’s attention on one or two key actions. Snapchat circa 2014 landed a user on a screen with a camera button, directions to “Tap to take a photo. Hold to take a video” and at most one other attention grabbing indicator (leading to a welcome message from the Snapchat team).

By 2020 we see an explosion in the number of clickable elements on that first screen. There are dot indicators on 4x different buttons, in 3x different colours. What should the user click first? What’s most important? Which one gets them closer to sending hilarious/inappropriate messages to their friends?

Houseparty has a similar explosion of clickable elements but without Snapchat’s subtlety. The first screen features three written instructions competing for the user’s attention. None of them get the user closer to joining an actual party.

Left to right: Snapchat, Houseparty

In both cases the user’s attention gets fragmented. It starts to feel like it’s going to be work to figure out the app and to get through to the thing they came for in the first place.

(And keep in mind that this is at the end of an onboarding flow that already involved work. The user has, by now, created their profile, added friends, given the app permission to access their contacts, camera, microphone etc.)

Medium: directing at the wrong moment

Like Snapchat and Houseparty, Medium introduces multiple actions to the user upfront, then adds in an extra layer of friction by referring to a key action that isn’t immediately accessible.

Saving — the first of those actions — is not possible from the home feed that the directions appear on. (A new user might think that clicking on the star would save an article. It doesn’t.) Even worse, when the user clicks into an article for the first time, where they can actually save, the app has forgotten about that action. Instead, it refocuses the user’s attention on claps.


The app seems two or three steps out of sync with the user. It’s breathlessly explaining everything that’s possible for the user, rather than getting the user doing what’s valuable to them.

KEY TAKEAWAYS:DO:1. Get the user doing something in your app, as soon as possible. Start delivering on those big promises about what your app will bring to the user's life. (Like TikTok, Shine and  Etsy)2. Focus the user on one key action at a time. Ideally, an action that carries over beyond onboarding, an action that they'll do again and again as they become long-term users of the app. (Again, like TikTok, Shine and Etsy)DON'T:1. Split the user’s attention multiple ways, leaving it to them to figure out the most valuable thing they could be doing next. (Like Nike Run Club, Houseparty and Snapchat)2. Direct users to actions that they can't take at that moment, giving instructions that are only valuable at some (unspecified) future time. (Like Medium)

3. The top apps … personalise the user’s experience, with minimal work by the user 🎯

There’s a whole separate post to be written on when and how much apps should personalise the user’s experience. There are categories of apps where a personalised experience is clearly valuable to the user (social media, for example). And others where personalisation has more limited value (utility apps, such as maps or weather, for example).

In this article, I’m focusing primarily on consumer apps, with a skew towards social media, chat and wellness apps — some of the buzziest categories of 2020. Many of the apps I’ve looked at make a big effort to personalise the app for users.

I’ve already written above about how TikTok and Etsy personalise the experience for users. They get users doing the key action that drives personalisation as part of their onboarding flow.

Other apps take a more linear approach to personalisation. They step the user through a series of questions that get them to an experience that feels relevant and valuable. The best of these personalise the app without making the users feel like they’re doing a whole lot of work to get there.


Spotify, Reddit: responding instantly to the user’s choices

Spotify quickly and efficiently gets the user to select 3 musicians that they want to listen to, from a seemingly infinite list of options.

The app’s onboarding features a list of musicians with some rough ordering by popularity, across a broad range of genres.

But the list isn’t static. As soon as a user selects a musician they’re interested in, the app updates the list — suggesting musicians similar to the one just selected. The app is responding to the user’s choices, giving the user a helping hand by making the assumption (powered by loads of historical data) that a user who, for example, likes The National, might also like Arcade Fire.

The app makes the initial selection of musicians fast and easy, then rewards the user for their selection, clearly personalising the app to the user’s preferences. When the user’s home feed loads, the app gives visual prominence to the same musician images that the user just selected. It places the user’s personalised ‘Daily Mixes’ — which are driven by their musician selection — at the top of the home feed. It uses up all the visual space above the fold (at the expense of the app’s other content and features).


Reddit can’t offer users the immediate reference point that Spotify has. Users are less likely to have existing preferences for subreddits, the way they are with musicians.

The apps solves this problem by starting on the ground that’s familiar to the user: asking the user to select general topics they are interested in. It offers users a finite list of topics, with topics on one screen, organised alphabetically.

Like Spotify, it responds to the user. It suggests sub-topics based on the user’s initial choices, helping the user get a bit more specific.

Only after selecting these topics/sub-topics is the user presented with a list of recommendations for subreddits to join.


Fabulous, Noom, Tasty: teeing the user up

Personalisation can be driven by far more than following users or topics or communities, of course.

Many apps need disparate pieces of information about the user to personalise the experience. This is especially true of health and wellness apps, the best of which step users through a series of questions to craft their profiles and goals.

They keep the user engaged by using a tone that’s crafted to empathise with users. The vary the question format and the UI. Most importantly they ask questions that are easy for users to answer. The effort the user needs to put in to answer the question is far less than the expected reward.

Fabulous uses a spectacular array of visual layouts as it builds a profile of the user. It asks questions that are quick, easy and feel relevant to the journey the user’s about to embark on.

Like Fabulous, Noom asks the user a pretty extensive list of questions, varying the visual format to keep the user engaged. For example, it asks users to provide their height by dragging a ruler rather than keying it in.

And it only asks questions the user can easily answer right now, without a huge amount of thought and calculation — and certainly without the expertise that the makers of the app possess. It asks, for example, for preferred weight loss speed in colloquial terms. The options range from slow to fast (or tortoise to hare on the visual scale). It doesn’t ask the user to provide a weekly target in kgs/lbs, as some other apps do. (Because: how would the user realistically know what that should be? And if they didn’t, would that become an obstacle for them continuing?)

Tasty asks only a single question, but an important one. The consequences of the answer (filtering — or not — of recipe suggestions) is built into the answer itself.

Left to right: Fabulous, Noom, Tasty


Medium: disorganised information

Unlike Twitter and Reddit, there’s no obvious rhyme or reason in the topic selection on Medium. It’s a finite list. Users can see all the topics with a couple of scrolls. But it’s not ordered alphabetically or by popularity. And there’s no grouping and obvious hierarchy. TV exists alongside Media, Javascript alongside Programming.

The user is forced to either read down through the whole list, organising the list in their mind and finding the topics that most interest them. Or they can pick a few that catch their eye and move on. Two options, each of which is unsatisfying in its own way.

Compounding this, there’s only a vague relationship between the topics selected and content that then appears on the user’s feed. There’s no obvious reward for those who have carefully selected their topics and are looking to the app to provide (as it promises in its app store listing) “ideas that find you”.


Pinterest: discarding the user’s work

Like Medium, Pinterest starts the personalisation process with a list of topics that lack clear ordering or hierarchy. The grid layout makes it a bit easier to read and visual cues make it more scannable. But it can still feel like work, scrolling and scrolling to find 5 topics that pique the user’s interests.

Once topics are selected the initial feed does reflect the user’s choices and feels relevant. It shows images from each of the topics the user selected on the previous screens. At least until … the user saves their first image. Then, all of the work the user did in selecting topics is thrown out the window. The feed, once refreshed, is populated only with images similar to that first saved image. This results in a homogenous feed that feels relevant in only a very narrow way.


Sure, the ‘work’ that was discarded was the product of just a single screen — and a minute or two of the user’s time. But it’s still something the user has done that has now been taken away from them. Rather than engaging with what’s in front of them and clicking on the next beautiful image, they’re left wondering how to get back to what they had, then lost.

KEY TAKEAWAYS:DO:1. Respond instantly to the user’s choices. Adjust your app's onboarding based on the user's initial selections and input. (Like Spotify and Reddit)2. Reward the user for the work they put into personalising the app. Clearly connect what the user sees after personalisation to the choices they've made and the answers they've given. (Like Spotify)3. Make every question you ask your users easy for them to answer right now, without effort or expertise. (Like Fabulous, Noom and Tasty)DON'T:1. Require the user to organise information that the app hasn't been able to / hasn't bothered to. (Like Medium)2. Discard the work the user has put in during onboarding, taking something away from them (however small the effort they put in to creating it). (Like Pinterest)

And … that’s a wrap. For now.

If that wasn’t enough, you can read the follow-on article, which, covers how the top apps keep users engaged through the whole onboarding process (incl. the boring parts 😪).

I’m a product consultant working out of cold, wet, grey London, UK. Follow me here on Medium for more on onboarding and user activation (plus some other stuff I have knocking about in my head). Or you can get in touch by email:

The Startup

Get smarter at building your thing. Join The Startup’s +751K followers.