How many things belong at the top of an app?

Picture the last time you opened your banking app. What did you want? Almost certainly one of two things: to see how much money you have, or to send some of it to someone. You didn’t open it to read the bank’s news, or fiddle with notification settings, or browse mortgages. Those things exist, and somebody needs them sometimes. But you, on a normal Tuesday, wanted your balance.

That’s the whole method, and it’s less clever than it sounds. It’s also the first thing we ask about on any app, long before anyone draws a screen. Before the team argues about the menu, we write down the handful of reasons a person opens the thing. Not the features — the reasons. “Check if the package is coming today.” “Find the recipe I saved.” “See who messaged me.” Do this honestly and the list is short; I’d be surprised if yours needs a second page. Most apps have two or three reasons people open them daily, and a long tail of reasons people open them once a month. (There’s a grand name for this part of the job, information architecture, which mostly means putting things where people expect to find them.)

The short list goes on the counter. The long tail goes in the cupboards.

How this goes wrong: say a recipe app lists its sections as Home, Discover, Recipes, Saved, Shopping list, Meal plans, Videos, Community, Shop, Profile, and Settings. Every one of those is real. But a person standing in a supermarket aisle with one hand on a trolley wants exactly one of them, the shopping list, and now they’re hunting for it among ten others.

Here’s the thing nobody says out loud in the meeting. A lot of navigation gets decided by who’s in the room. The team that runs Community wants Community at the top, because that’s their job and their numbers. The team that runs Shop wants Shop up there for the same reason. Everyone is being reasonable, and the result is a menu that looks like a list of departments. Does anyone using the app think in departments? No. They think in things to do.

I get the feeling, and I don’t think anyone’s being petty. When your section isn’t at the top, it feels like your work doesn’t count. But the top of the app isn’t a trophy shelf. It’s a counter. It holds what’s used most, and that says nothing about how good the waffle iron is. (I’d defend the waffle iron’s honor to anyone. I just wouldn’t give it the counter.)

The way through, I think, is to pin the reasons list to the wall and let it decide. It’s much easier to argue with a list of what people actually do than with the colleague whose section just moved.

Partly because of your hand. On a phone, the row of buttons along the bottom has to fit across one screen, and each button needs to be about the size of a fingertip so your thumb doesn’t catch its neighbor. Past five, the buttons get narrow and the labels start to squash.

Partly because of the people who make the phones. Apple’s guidance on tab bars — that row of buttons at the bottom of most iPhone apps — asks designers to keep the count small, and if an app insists on six or more, the iPhone quietly shows four and turns the fifth into a button called “More.” Google’s guidance for the Android version of that row says three to five. Two companies that rarely agree on anything landed in the same place. I find that oddly comforting, and it’s the thing we point at whenever someone wants seven.

And partly because of how deciding feels. Every extra option is one more thing to read and reject before you tap. With three choices, you barely notice choosing. With nine, you feel it. It’s a café with a small chalkboard versus a diner with a laminated menu the size of a newspaper. Which one gets you fed faster? The chalkboard, every time (the diner has more pancakes, sure, but you’ll spend four minutes finding them).

But I said about five. Our rule of thumb is that five is a ceiling, not a target. If your app has two things people do, put two things on the counter. Adding a third because the row looks empty is how junk ends up next to the kettle.

You’ve seen the icon. Three short lines in a corner, usually top left. People call it a hamburger menu because it looks a bit like one if you squint and you’re hungry. Tap it and a panel slides out with everything the team couldn’t agree on.

I’m not against it in every case. On a website with forty pages, you need somewhere to put the forty pages. But in a phone app it’s often a way to avoid a decision rather than make one. Everything goes in the panel, the top stays tidy, and the team gets to feel organized. Is that just my taste? Partly. But it’s also been measured. When the Nielsen Norman Group tested hidden menus against visible ones, people on a laptop used the hidden menu in only about a quarter of cases, roughly half as often as a visible one, and took at least 39% longer to get things done. On phones the gap was smaller, but it was there. So we’d take a short row you can see over a long list you can’t, almost every time.

The hamburger is the kitchen’s junk drawer. It looks fine from the outside. Open it and there are batteries, a birthday candle, three takeaway menus, and the one key nobody can identify. The problem isn’t that things are tucked away. It’s that they’re tucked away without an order. A good cupboard has a logic: baking things together, glasses together, lids near the pots. So does a good secondary menu. Group the less-used things by what they help you do, give each group a plain name, and the cupboard becomes a place people can find things instead of a place things go to disappear.

How do you tell which one you’ve built? Ask someone to find a specific setting. If they open the panel and read every item top to bottom, it’s a drawer. If they glance, spot a group name, and tap, it’s a cupboard.

Think about how you hold your phone on a train. One hand, probably. The other is holding a coffee or a rail. Your thumb sweeps comfortably across the bottom half of the screen and has to stretch, awkwardly, to reach the top corners. (Try it now. The top left corner is a small act of yoga!)

That’s why the most-used things belong at the bottom of a phone screen, in that row of buttons where the thumb already rests. So what does “the top of the app” actually mean on a phone? The most reachable place, and that’s the bottom. On a laptop, where you’re pointing with a mouse, the top of the window is fine. This is also, quietly, one more argument against the hamburger in its usual spot. The icon that holds everything sits in the one corner the thumb has the hardest time reaching.

Everyday things on the counter at hip height. Occasional things in the high cupboard. That’s not a rule someone invented for software. It’s just how kitchens work, and hands.

Here’s the part teams forget once the menu looks right on the big screen in the meeting room. Ever stayed with a friend who “helpfully” reorganized your kitchen while you were out? Everything’s tidier. And for a week, you open the drawer where the spoons used to be.

People who use your app every day have that kind of memory in their thumbs. They don’t read the bottom row anymore; they reach. Move a button and you haven’t just changed a screen, you’ve changed a habit, and habits take a while to catch up.

How this goes wrong: say you swap the second and third buttons because the new order looks more logical to the team. Nothing breaks. But everyone who opens the app ten times a day now taps the wrong thing ten times a day, for about a week, and to them the app feels broken even though it isn’t.

So we’d change the top rarely, and when you do, change it once and properly instead of nudging it every month. Say what moved, in plain words, the first time they open it. And never move the one thing people use most. That’s the kettle. Nobody moves the kettle.

How do you know the counter is right? You watch someone cook. Not literally. But you can do the software version in an afternoon, and it’s the most useful afternoon in the whole project. I mean that; I’d trade a week of meetings for it.

Here’s the setup, as a hypothetical. Say you have a rough version of the app on a real phone. Sit down with five people who aren’t on your team, one at a time, and give each of them a few small errands:

• “Find the recipe you saved last week.”
• “Add milk to your shopping list.”
• “Change the email your receipts go to.”

Then stop talking, and watch. You’re not watching for whether they succeed. Most people eventually will. You’re watching for the hesitation: the little pause, the thumb hovering, the tap on the wrong button followed by a quick retreat. Every pause is someone standing in your kitchen, opening the wrong cupboard.

Sitting on your hands is the hardest part. You’ll want to help! Don’t. The moment you point, you’ve learned nothing except that you can point. (I think this is the hardest habit in the whole craft to keep, and I’m not great at it.) If three of five people look for the shopping list under Meal plans, the shopping list probably belongs near Meal plans, whatever the team decided on the whiteboard. The people are right and the whiteboard is wrong. That’s not a failure. That’s the afternoon working!

One last thing, and it’s the one that sneaks up on teams. The names on the buttons matter as much as how many there are. Teams tend to name sections after what they are internally — “Hub,” “Workspace,” “Discover” — and then wonder why nobody taps them. What’s in the Hub? What gets discovered? A person can’t tell from the outside. I can’t either, and I’d guess the team that named it can’t say for sure.

Name the button after the thing it holds or the thing it does. “Saved.” “Orders.” “Messages.” “Search.” I’ll take short, boring, obvious words here every time, because the reader’s job is to not think. It’s the label on the kitchen jar. A jar marked “Rice” is more useful than a jar marked “Staples,” even if “Staples” sounds more elegant on the shelf.

Want a quick test? Read the row of button names out loud to someone who’s never seen the app and ask what each one opens. Four out of five right and you’re there. Two shrugs and you rename those two before you move a single thing.

You, with the app that grew to eleven sections: you don’t need eleven spots at the top. You need the two or three things people reach for every day, placed where the hand already is, with names a stranger would understand. The rest belongs in cupboards with a logic to them, not in a drawer with a birthday candle.

And when someone on the team pushes to put their section on the counter, the answer isn’t “no.” It’s “show me the kettle.” If people reach for it every day, up it goes. If not, it goes somewhere good, and it’s still in the kitchen. (Yes, this is a long answer to a question about having fewer things. I noticed too.)

In one breath: list the reasons people open the app, put the top two or three in plain sight, stop at five, give everything else a cupboard with a logic, keep the everyday things where thumbs rest, move things rarely, watch five people run errands, and name every button after what it holds.

That’s all of it. I hope your next meeting about the menu is short, and that someone leaves it happy to see their section in a good cupboard.

Hope this helps.