The research we followed when we designed your parent's screen

Marlow has two halves, and the half your parent sees is the one people ask about. It shows today and what’s next, in big type, and nothing on it can change the schedule. That wasn’t a hunch. Most of it came from reading what people who study older adults and technology have been saying for years, and then doing what they said.

This post is the reading list, roughly, and the decision each part of it led to. The three posts that follow go deeper on vision, attention and hands.

We ran no studies. We read.

Worth saying first. Marlow is a small app built by one person, and we have not sat anyone in a lab. What we did was go through the published guidance: the CREATE center’s work and the book Designing for Older Adults by Sara Czaja, Neil Charness, Walter Boot and Wendy Rogers; Nielsen Norman Group’s usability testing with older adults; the WCAG accessibility guidelines; and Apple’s own Human Interface Guidelines. Then we held every part of the senior side up against it.

If something below sounds like a finding of ours, it isn’t. It’s theirs. And where we’ve taken a finding further than its authors did, we try to say so.

Confidence predicts use better than age does

The CREATE researchers keep coming back to one result. Whether an older adult takes up a piece of technology is predicted by their anxiety and confidence around it more than by how old they are. The fix they describe isn’t simply making the thing simpler. It’s lowering the stakes: training, and design where nothing you do can go badly wrong.

The same work is clear that older adults learn technology perfectly well when someone takes the time to teach them. So read only is not a verdict on what your parent can do. Plenty of 78 year olds run their own calendars, and if yours is one of them, a screen she can’t touch may feel like being talked down to. Nothing stops a family from inviting a parent into the circle like anyone else. The senior side is for the parent who shouldn’t have to operate anything, and for the family that would rather she didn’t have to.

For that parent, the stakes argument reshaped the whole thing. A simpler app still has a login, and a login is a thing you can fail at. So there is no account on your parent’s phone and no password. A family member pairs it once with a code shown on their own screen. After that it shows the day.

It also settled read only. The parent can’t add, move or cancel an appointment, and that isn’t done by hiding buttons. It’s enforced at the database, so the server would refuse a change from that account even if the app had a bug. There is no wrong tap on the schedule, because there is nothing on the schedule to tap.

Eyes change, and the guidelines have numbers

Presbyopia arrives in most people’s forties and is nearly universal later. Reading glasses fix it, which is worth saying plainly, because big type is not a cure for presbyopia. It’s a hedge against the glasses being on the nightstand. What glasses don’t fix is the rest: the aging lens yellows and lets through less light, contrast sensitivity falls, glare takes longer to recover from, and older eyes need considerably more light to read the same page. Cataract is common.

WCAG gives minimums: 4.5 to 1 contrast for normal text and 3 to 1 for large text. Those are floors written for a general audience. On the senior side we treat them as the line to clear comfortably, not to meet. Dark text on a light ground, and contrast well past the floor. The family side is set at an ordinary adult size, and even that is bigger than most apps, because many of the people organizing a parent’s care need reading glasses themselves. Why the type is so big walks through it, including the cases where bigger type isn’t enough.

Fewer choices, less to hold in mind

Two well worn results shaped the layout. The Hick Hyman law says the more options in front of you, the longer the choice takes. Working memory research, from the Baddeley and Hitch model through Miller’s “magical number seven” and Cowan’s argument that the real limit is nearer four, says people can hold very few things at once, and that the capacity declines with age. Salthouse’s processing speed theory adds a general slowing that accounts for a lot of the age gap on cognitive tasks. Divided attention gets harder too.

None of those results is about menus. They’re lab findings about choices and lists, and applying them to a phone is our inference, not theirs. But read together they argue against tabs, a calendar grid and anything you have to navigate back from. So the senior side is one screen. It shows the next appointment, the time to be ready, and who is coming. Nothing to remember from a previous screen.

That has a cost. The screen shows today and what’s coming next, not the month. A parent who likes to plan two weeks out won’t get that here, and should know before the phone arrives. One screen, no menus is the long version, cost included.

Small targets, unsteady hands, and every extra step

Fitts’s law, from Paul Fitts in 1954, says the time to hit a target depends on how far away it is and how big it is. Small targets take longer and get missed. Add hands that have changed, whether that’s arthritis, a tremor, or fingertips a touchscreen doesn’t always register, and a small control becomes a control that sometimes does the wrong thing. Apple’s guidelines ask for tap targets of at least 44 by 44 points; WCAG’s AAA level asks for 44 by 44 pixels.

The accessibility literature adds the part that matters more. Older adults are more reluctant to explore an interface when they are afraid of breaking something, and slower to recover once an error happens. So the answer on the senior side was not bigger buttons. It was no buttons that change the schedule. Everything you can tap is large, and none of it can move an appointment. Big targets, no mistakes is that argument in full, including the one tap on her phone that does record something.

Nielsen Norman Group’s testing found older adults slower and less successful on interfaces built for younger people. The plain reading of that is that every step between the person and the information costs more than it would for a younger user. Unlocking a phone is a step. Finding an app is a step.

So the same information appears on the lock screen and in a home screen widget. When to be ready, who’s coming. Your parent can see it without unlocking anything or opening anything.

What we can’t measure

Marlow runs no analytics and no crash reporter. There are no trackers, and that’s true of the website too. It’s a deliberate privacy choice. It also means we cannot tell you that the senior side works. We don’t know how often it’s looked at, or whether “the ride is being arranged” reassures your mother or annoys her.

What we do instead is ask families. Marlow is small, and the parents using it are the ones telling us what’s wrong. That’s slower and less tidy than a dashboard. It’s also the only kind of evidence we’re willing to collect from an 80 year old who never signed up for anything.