Control Vertex Hub
Method

We watch the hunt, not a scripted tour of every screen

A session is a person, a phone, a short list of tasks, and someone in the room who will not point at the gear mark. This page is the protocol we use in Petaling Garden.

People talking around a table with notebooks

Before anyone sits down

You send a build we can install without a work account we do not have. You name the controls that already generate complaints, or the ones you are about to move. We write tasks in the language of the job, not in the language of your settings tree. “Make the rain show in millimetres” is a task. “Open Units, then Precipitation” is a hint, and we will not use it.

Priya or Amira walks the tree and draws the control map: path, label, default, whether a change survives a restart, and how the previous state is restored. Empty cells in the reverse column are flagged before testers arrive. If a reverse path does not exist, the session will show the moment a person looks for it.

Who sits in the chair

We recruit in the Klang Valley for people who already use a similar utility. A flashlight study does not use testers who have never turned on a torch. Staff from your company are excluded. So are children. Six to eight people is the usual set for an Interaction Audit; recovery sessions can run with four.

Language follows the tester: English, Bahasa Malaysia, or Mandarin. Wei Liang facilitates in the first and third; Amira covers Bahasa Malaysia. The consent form is on paper. We do not film faces unless the brief asks for it and the tester signs a separate line.

In the room

The hour itself

  1. Settle and consent

    Ten minutes. The tester reads what we will ask them to attempt, how notes are stored, and that they can stop. The phone is already on the table, on a home screen that is not your app.

  2. First task without a tour

    We do not show the settings tree. The first task is always something the main screen of a utility can almost do, then a control that lives one or two screens deeper. Silence is allowed. Pointing at the corner is not.

  3. Change, then reverse

    If the person succeeds in changing a setting, the next task is to put it back. This is where most of the useful notes come from. A confirmation that appears on the way in and not on the way out is recorded as such.

  4. Short debrief with the tester

    We ask where they thought the control would live, which words they looked for, and whether they would try again on their own phone. We do not ask them to rate the app.

What the memo is, and is not

Findings are grouped by control. You will not receive eight personality portraits. You will receive, for each setting in scope, where people hesitated, which label they misread, whether they found a reverse path, and whether they looked in the system settings instead of in your utility.

We keep quotes when a sentence from the room is clearer than our paraphrase. We do not invent scores. If four of six testers never opened a nested screen, we write that. If one person loved a label that confused everyone else, that sits in a footnote, not in the headline.

After the last session

The debrief with you is ninety minutes. Amira walks the map, then the two or three controls that caused the most abandoned hunts. The conversation is about what to change first, not about whether the app is “good.”

Field Observation Days use a lighter protocol: no task cards, notes when settings are opened among other work, photographs of the physical place only with consent. Recovery sessions skip the full tree and begin with a prepared wrong state.

Choose a study type