Skip to main content

Running Unmoderated Studies in Lookback

Learn how to plan, run, and review unmoderated studies in Lookback using Tasks and SelfTest - with a focus on data quality, participant clarity, and efficient analysis.

Henrik Mattsson avatar
Written by Henrik Mattsson
Updated today

Unmoderated studies in Lookback are designed to capture qualitative “why” data, not just task completion.

Participants complete sessions on their own time, but researchers still retain strong control over structure, intent, and evidence quality.

This article explains how to run unmoderated studies operationally and effectively.


What unmoderated means in Lookback

In Lookback, unmoderated research means:

  • participants complete sessions without a live researcher present

  • sessions are recorded (screen, audio, and often camera)

  • recordings stream live as they happen

  • findings and notes can be created immediately

Unmoderated does not mean:

  • purely quantitative testing

  • silent interaction

  • delayed analysis


Unmoderated study types

Lookback supports two unmoderated modes:

Tasks

Structured, step-by-step studies where you define:

  • tasks or questions

  • expected participant actions

  • follow-ups (including AI-assisted follow-ups)

Tasks are best when you want:

  • consistency across participants

  • clearer task boundaries

  • higher data quality at scale


SelfTest

Lightweight, instruction-driven studies where participants:

  • follow one simple written instruction

  • explore more freely

  • narrate their actions out loud

SelfTest works well for:

  • exploratory research

  • diary and longitudinal studies

  • early-stage concepts


Before you invite participants

Successful unmoderated studies are decided before the first participant joins.

Make sure you have:

  • chosen the right mode (Tasks vs SelfTest)

  • written clear, unambiguous instructions

  • explicitly asked participants to think out loud

  • confirmed device and app requirements (mobile requires Participate)

  • tested the flow using a Preview Session

If participants misunderstand the task, there is no live moderator to correct them.


AI moderation in unmoderated studies

Unmoderated Tasks can use AI moderation to reduce common failure modes.

AI moderation can:

  • prompt participants to elaborate

  • ask clarifying follow-up questions

  • remind participants to answer fully

AI follows researcher intent - it does not invent goals or conclusions.
This helps preserve qualitative depth without adding live moderation.


During live unmoderated sessions

Even though the study is unmoderated:

  • sessions stream live to the dashboard

  • stakeholders can observe in real time

  • researchers can start taking notes immediately

You don’t have to wait for all sessions to finish to begin analysis.


After sessions complete

Once sessions are complete:

  • recordings are immediately available

  • findings can be created from key moments

  • themes can be built across sessions

  • stakeholders can review evidence asynchronously

Lookback encourages analysis to happen close to the raw material.


Common unmoderated pitfalls

  • instructions that assume shared context

  • forgetting to remind participants to speak out loud

  • tasks that combine too many goals

  • skipping preview and precheck steps

  • assuming unmoderated = hands-off

Good unmoderated studies require more upfront thinking, not less.

Did this answer your question?