A café sign can look clear from behind the counter while someone arriving from the car park cannot tell which door to use. Before ordering another sign, observe the actual route from a defined arrival point to the public entrance.
This is a small task trial with volunteers. It ends at the door: it does not measure passing traffic, queues, sales conversion or the accessibility of the premises. ER-029 is an original fictional example showing how to keep assistance and unfinished trials visible.
Define the route before inviting people
Choose the arrival points people actually use, such as the public pavement approach and the pedestrian arrival from a car park. Record an exact start description, the intended public entrance, date, lighting and weather, and the version of the signs being tested. A person starting in sight of the door faces a different task from someone approaching around a corner.
Walk the route with the responsible site team first. Use only routes that are permitted and appropriate for the participants. Do not put anyone into traffic, obstruct a path or hide a hazard to test whether a sign works. Questions about lawful signs, installation and accessible routes need their own site-specific review; a successful trial does not answer them.
Give a task without pointing out the answer
Recruit willing first-time visitors who resemble people likely to use the café. Explain the observation, obtain agreement and make clear that they can stop. Record an anonymous participant code and any earlier exposure to the route. Avoid names, photographs or unnecessary personal details in the shared worksheet.
Use the same neutral instruction, such as: ‘From here, please find the public entrance to Café A.’ Do not point at the sign or name the turn you want them to take. Watch where they look, hesitate, turn back or ask for help. If help is requested or needed, provide it and record what changed. Safety and participant comfort take priority over completing the observation.
Worked example: two arrival points, five planned trials
ER-029 tests sign version V1 at fictional Café A on 8 November 2026 in dry daylight. Start A is the public pavement approach; start B is the car-park pedestrian arrival. Five volunteers are planned, all unfamiliar with the route. The task finishes when the participant reaches the intended public entrance.
T3 asks for help after trying the wrong door. T4 chooses to stop before reaching the entrance. T5 is unavailable and never starts. Those outcomes must remain different in the record.
| Trial | Start | Started? | Outcome | Observation |
|---|---|---|---|---|
| T1 | A | Yes | Independent | Reached intended entrance |
| T2 | A | Yes | Independent | Reached intended entrance |
| T3 | B | Yes | Assisted | Tried wrong door; asked for help |
| T4 | B | Yes | Incomplete | Participant chose to stop |
| T5 | B | No | Not started | Participant unavailable |
Count the trials you actually observed
There are four started trials: two independent, one assisted and one incomplete. One further planned trial did not start. An assisted arrival is useful evidence, but it is not an independent arrival. An unstarted trial is not a failed route attempt, and a missing observation must not be filled in as success.
These counts describe this small convenience sample under the recorded conditions. They do not establish a population success rate, prove that one approach is generally worse, or show how many visitors became customers. Do not invent a time target or remove the difficult cases to improve a headline result.
Turn observations into specific questions
Record the decision point, what was visible and what the participant actually did. ‘T3 approached the adjacent door and asked which entrance to use’ is evidence. ‘The sign is too small’ is a possible explanation that still needs checking. Ask neutral follow-up questions after the task to understand the observation.
For ER-029, the owner proposes a clearer entrance cue for the approach from B. The proposal is not yet installed. Record the owner, permission and route-review dependencies, and the intended change before testing it. Avoid moving an obstacle or changing several unrelated cues without documenting them, because the next result would describe a different route.
Retest the changed version and retain the earlier result
Give the new arrangement its own version and record what changed. After the relevant site review and permissions, test it from both arrival points with new first-time visitors under stated conditions. Someone who has already learned the route can help inspect a change, but their repeat visit is not another independent first-time trial.
Keep V1 and its four started outcomes. A later V2 result should sit beside its own trial records, including help, stops and unstarted sessions. If the lighting, entrance position or arrival route changes, record that context before comparing observations. A route that works in daylight has not thereby been tested after dark.
Use task observation as a method, not a certification
The GOV.UK Service Manual describes watching people attempt realistic tasks, giving neutral instructions, avoiding hints and recording what happens. This guide applies that general research method to a physical entrance. The source does not prescribe café signs, dimensions, permissions or an accessibility standard.
Keep the resulting questions practical: which decision point confused someone, what help was needed, what change is proposed and what happened in the next trial. For counting crossing events, use the footfall guide; for waiting after arrival, use the queue guide.
Put it into practice
Your action checklist
Your progress saves in this browser. No sign-up needed.
0 of 10 complete
From reading to doing
Entrance Route Trials
Use Entrance Route Trials for one sign version and set of arrival conditions. It is a fully manual worksheet with scope, task, consent-process and route fields; 10 trial rows; 6 issue/action rows; and 4 retest rows. Each trial records its ID, anonymous participant code, start point, earlier exposure, whether it started, outcome, observed decision point, help or stop reason and evidence. Summary counts are typed entries and must be recounted after detail changes. Blank or unobserved results remain unresolved. There are no formulas, automated accessibility findings or statistical conclusions. Extend the record rather than omit difficult trials to fit the available rows.
Open the Entrance Route Trials tab in the launch workbook. Keep a copy for your own business.
Download the free workbook See what’s inside the toolkit →Sources & further reading
- Using moderated usability testing — GOV.UK Service Manual ↗
Updated 3 October 2017; retrieved 8 November 2026. Used only for general task-observation principles: relevant tasks, neutral instructions, avoiding hints and observing behaviour. Applying these to café entrance wayfinding and all ER-029 details are original. No sign specification, permission or accessibility certification is inferred.
Published by Keystonne Grow. Product links connect to the Keystonne ecosystem; check the linked product or service for its current scope.