Findability testing — Examples

Example 1 — Turning route tasks into target tasks

A colleague drafted the instructions for a findability round, but each one names the path to follow. Rewrite the tasks as targets, then record a first click even where the person succeeds.

1 Go to Billing, then open invoice history and find last month's invoice.
2 Click Account, then Addresses, and check the saved postcode.
3 Use the top menu to reach Returns and start a return.

Solution

1 You need the invoice for last month's order.
2 A parcel is going to an old address; you want to check what is saved.
3 A jumper does not fit; you want to send it back.

Recording, task 1 (illustrative)
P1 first click: Your plan and bills   success, 2 clicks
P2 first click: Your orders           success, 5 clicks   <- structure weak
P3 first click: Help and support      failed

A route tests whether the person can follow instructions rather than whether the structure leads them anywhere, so the rewrite leaves the path as the finding. Recording the first click separately from the outcome is what exposes P2: a correct answer after four guesses points at a weak label. The mistake is to score success alone and miss that nobody’s first instinct matched the group.

Example 2 — A first-click rate you can act on

Four tasks from a round have been counted, and the table below shows where each participant’s first click landed (illustrative). Decide which tasks need fixing, what change to make, and how the re-test should be set up.

Task                        Correct first clicks   Participants
Find the returns label              3/5               5
Check the delivery window           2/5               5
Find the invoice                    1/5               5
Update the saved postcode           4/5               5

Solution

Fix now:  Find the invoice (1 of 5 correct first clicks)
          Check the delivery window (2 of 5)
Action:   rename the group so that bills are visible in its label;
          add a "Delivery" line to the same group

Re-test:  the same three tasks, three new people, grey boxes and text
          links only, so the comments stay on the structure

A first-click rate under about two thirds says the label or the grouping is wrong for that target, so the count names the work instead of the loudest opinion in the room, and grey boxes keep attention off the styling. The mistake is to redesign from a single participant’s path: three to five people and a rate per task is the smallest unit that supports a decision.