011 min
The problem Red Route Usability solves
Picture a team that runs an online shop selling headphones. They have just run a usability test, and their notes list a long run of problems. One button label confuses people. A filter hides some results. The profile page rejects certain photo files. Every person in the room has a favourite problem, and whoever argues most strongly usually decides what gets fixed first.
The team can fix only so much in a month, so the order matters more than the list. Red route usability gives them one question to ask about every problem: does it sit on an activity that many people do, or that people cannot do without? A shared answer also stops the argument, and it stops features that serve no key activity from being added to the plan without a reason. To see why the question works, it helps to know where it came from.
021 min
Where Red Route Usability comes from
The question comes from the roads of London. Red lines along the kerb mean that vehicles may not stop there, and the lines mark the key roads that must stay clear for traffic to move. London's first red lines were painted in 1991, on roads in the north and east of the city. One driver who stops on a red route can hold up every bus behind them. The rule therefore allows no quick stop to buy a newspaper.
The rule helped bus passengers. Research that David Travis later reported found that bus journeys on the red routes became 10% quicker and 27% more reliable.
In 2006, Travis, a chartered psychologist at the consultancy Userfocus, borrowed the idea for websites. His reasoning was that a site has a few key journeys, just as a city has a few key roads. A team that finds those journeys can treat each obstacle on them the way London treats a stopped car, and remove it without exceptions.
032 min
What counts as a red route
To use the question, the team first needs to know what a red route is. Travis's test has two parts: how often people do the activity, and how badly they need it.
A frequent activity shapes how people feel about the product, because they meet it on every visit. Searching is one example. A critical activity may be rare, but people dislike a site that handles it badly. Travis's example is changing the personal details stored on a site. An activity that is both frequent and critical is the most important part of the product, and getting it wrong costs the most.
At the headphone shop, buying a pair of headphones is the clearest red route. It is not one click, though. Travis lists five traits of a red route. The first is that it is a complete activity, not a simple task, and it will probably need several pages. The other four are:
- Success is obvious. Anyone can say what finishing looks like, which a hope that the site will be easy to use cannot do.
- It carries over to other sites. Buying a product works on any shop, so a team can compare itself with its competitors.
- It names a goal, not the screens used to reach it, so it does not fix one design.
- It is accurate and realistic: it reflects the most important goals of the customer and of the organisation.
Choosing is a red route too
A shop's own list of red routes starts with buying. Shoppers have another route that the shop can miss. Before they pay, they want to know that the price is fair and that the model suits them. Travis calls this activity choosing, and it needs independent reviews, comments from other buyers and prices from other shops. The seller gains nothing directly from a rival's price, but the shopper will not buy without it. A red route list therefore has to come from the customer's goals as well as the business's.
042 min
How to find the red routes in a product
Naming the traits is the easy part. The hard part is knowing which activities people really come to do, and the weakest way to learn that is to invent them.
When a red route is invented
Travis describes a client whose typical red route, reading the latest company news, was backed only by an invented story about a journalist on a slow news day. A team can believe a story like that for a while, because stories are easy to tell. It is still no evidence that any real customer follows that route.
Where real evidence is
Travis points to data that most organisations already hold. A team can read a month of search queries and group them by what people were trying to do. It can ask the staff who meet customers face to face what customers ask. It can spend a day listening to support calls and sort them.
Once the activities are listed, the team can count them. A common tool is a grid with the number of users who do an activity on one axis and how often they do it on the other. Activities in the corner where both are high are the red routes. For the headphone shop, buying and choosing sit in that corner. Changing a profile photo sits near the opposite one. A grid that ranks by frequency can hide a rare but critical activity, such as editing stored details, so the team also asks of each activity how critical it is.
A survey instead of a grid
Gerry McGovern's top tasks method reaches a similar list by asking people. The team writes a long list of tasks, 50 or more, shows it in random order to representative users and asks each person to pick five. When the votes are added up, a handful of tasks collect most of the votes and a long tail of tasks collects almost none. McGovern calls the handful the long neck. The grid answers the question of what people do, from their behaviour. The survey answers the question of what people want, from what they say. A team that has both can see where the two disagree.
051 min
Red Route Usability in real products
A route is easier to see in products that many people know. A ride-hailing app has two red routes, one for each side of the market: a rider requests a ride, and a driver accepts one. A home-booking service splits the same way: a guest books a room, and a host manages reservations. A team that serves two groups of users needs a route list for each group, because a fix that helps riders does nothing for drivers.
Amazon's one-action purchase
Amazon shows what removing steps from the buying route looks like. The company filed a US patent application for single-action ordering in September 1997, and the patent was granted in September 1999. In the patent, a shopper the system already recognises gets one action to place the order, such as a mouse click. The returning customer, who uses the route most often, skips the extra questions that other shoppers still answer.
What a hard buying route costs
A buying route that is hard to use costs sales. Baymard Institute, a research firm that studies checkouts, asked US online shoppers why they had abandoned a cart. Setting aside the many who were only browsing, 18% named a demand to create an account, and 17% named a checkout that was too long or complicated. Both are obstacles on the red route itself, and the shop put them there.
063 min
How to rank problems with Red Route Usability
With its routes named, the headphone team can go back to its long list of problems. It asks three questions about each one. Does the problem occur on a red route? Is it hard for people to get past? Does it keep happening, or does it bother people only once? The more yes answers a problem collects, the higher it ranks. A problem with three yes answers is critical, and the team fixes it first. A problem with none is trivial, and it waits. Sorting by urgency in this way is triage, the same idea used in emergency care.
Two problems show how it works. The address step on the buying route clears what a shopper typed whenever they go back one step, and the profile page rejects some photo files. The address step is on a red route, it is hard to get past because shoppers must type everything again, and it happens on every visit that goes back a step, so it collects three yes answers. The photo upload is off the route and easy to get around by choosing another file, so it collects none. The team fixes the address step first and leaves the photo upload for later.
Fixing is not the only claim on the team's time. Frameworks such as MoSCoW, which sorts work into must, should, could and will not, or RICE, which scores reach, impact, confidence and effort, help compare usability fixes with new features. The red route answer is one input to them. The same logic keeps a new product small: its first version only has to make the red routes work.
Where the three questions come from
The last two questions are older than red routes. In 1994, Jakob Nielsen wrote that the severity of a usability problem combines how often it occurs, how hard it is for people to overcome, and whether it keeps bothering them or only bothers them once. The red route question takes the place of the first of these. Red routes are the activities people do often or cannot do without, so a problem on one is met by many people.
Checking the route with a completion rate
After the fixes, the team needs to know whether the route improved. The simplest measure is the task completion rate. In a usability test, each person gets a task with a clear finish line, such as buying a pair of headphones. A person who finishes scores 1, a person who fails scores 0, and the average is the completion rate. A grid of tasks against participants, shaded by success, shows at a glance where the route breaks. Jeff Sauro of MeasuringU gathered 1,189 tasks from 115 tests and found that the average completion rate was 78%. He notes that a consumer site that people use without training often aims for 70%. He also warns that a test only estimates the real rate: if 10 out of 10 people finish, the team can be 95% sure that between 75% and 100% of all users could.
071 min
Red Route Usability and accessibility
A route that works for most people can still shut out some. The web accessibility standard WCAG 2.2, published by the W3C, has two success criteria that bear on the buying route of almost any shop.
Signing in
Success criterion 3.3.8, Accessible Authentication (Minimum), applies to the sign-in step that many buying routes include. It says a step that relies on a memory or puzzle test must come with another way in, or with support such as password managers and pasting. Success criterion 3.3.8 exists because people with certain cognitive disabilities may be unable to solve a puzzle, remember a login or retype a one-time code. A shopper who is locked out at sign-in never reaches the payment page, however well the payment page is built.
Entering details twice
Success criterion 3.3.7, Redundant Entry (Level A), covers the address and payment steps. Information the shopper already gave in the same process must be filled in for them or offered for selection, with limited exceptions such as when the earlier entry is no longer valid. A familiar case is a checkbox that says the billing address is the same as the shipping address. Success criterion 3.3.7 aims to help people finish processes with several steps, such as buying something.
081 min
Where Red Route Usability falls short
Ranking by routes is a tool for choosing, and it has limits.
- Narrow focus. A team that looks only at its red routes can miss needs that sit elsewhere. An activity that serves a small group well can matter a great deal to that group, even though it never reaches the top corner of the grid.
- Routes change. What customers want shifts, and so do the products around them. A list made once and never checked slowly describes a shop that no longer exists.
- A route is not a page. When the list turns into the checkout page, the team starts perfecting screens and loses the goal. The route is the goal the shopper has, and the same goal can be reached through different screens.
- Saying differs from doing. A survey records what people pick from a list, and analytics records what they did. A team that relies on one of them can be misled by it.
091 min
Red Route Usability vs. nearby concepts
Red route usability sits close to several other ideas, and they are easy to mix up. Each answers a different question.
Red routes vs. user flows
A user flow draws the screens and decisions one person passes through to finish a task. It describes one path in detail. Red route usability does not draw any path. It decides which paths deserve the team's attention first, and a flow is what the team then draws for each one.
Red routes vs. journey maps
A journey map follows a person's whole experience, including what they think and feel, often across weeks and channels. A red route is a single activity inside such a journey, chosen because of how often and how badly it matters.
Red routes vs. the Pareto principle
The Pareto principle says that a small share of causes produces most of the effects, often written as 80/20. Red routes rest on the same intuition, that a few activities carry most of the value. But Pareto only counts how much. A red route is also chosen for how critical it is, which a count alone does not measure.
?4 questions
Questions people ask
How many red routes should a product have?
Does red route usability work for mobile apps?
What happens to problems that are not on a red route?
Is a red route the same as a conversion funnel?
§11 sources
Sources
David Travis, "Red route usability: The key user journeys with your web site", Userfocus, 25 March 2006 (Internet Archive snapshot of 13 February 2014):
"Red route", an encyclopedia entry on the London road markings:
Jakob Nielsen, "Severity Ratings for Usability Problems", Nielsen Norman Group, 1994:
Baymard Institute, "50 Cart Abandonment Rate Statistics", 2026 edition:
Show all 11 sourcesShow fewer sources
Amazon.com Inc, "Method and system for placing a purchase order via a communications network", US patent 5,960,411:
W3C, "Understanding Success Criterion 3.3.8: Accessible Authentication (Minimum)", WCAG 2.2:
W3C, "Understanding Success Criterion 3.3.7: Redundant Entry", WCAG 2.2:
Jeff Sauro, "What Is A Good Task-Completion Rate?", MeasuringU, 2011:
Jeff Sauro, "How to Conduct a Top Task Analysis", MeasuringU, 2013:
Userbrain, "How to Prioritize Usability Issues":
The Decision Lab, "Red Route Usability":
