Skip to content

GRW Data Collection · for L&D and HR

Stop asking.
Start measuring.

Your survey data depends on who chose to reply. GRW Data Collection measures what a group did in the session, with no form to fill in. It reports on the group, not on one person. Anyone can opt out with a sticker.

The trouble with survey data

Six weak spots in a feedback form.

Surveys ask what people think, and only people can tell you that. But the numbers they return have known weak spots.

Who replies

The people who answer may not be like the people who stay quiet. Published guides put a good response rate for staff surveys at about 65% to 85%, higher in small firms. A high rate shrinks the gap. It does not tell you who is in it.

Source: Culture Amp; CustomInsight

Memory

A smile sheet asks what people remember once the session is over. Studies of training have found that how people rate a course is a weak guide to what they learned.

Source: Arthur and others, 2003, citing Alliger and others, 1997

Who reads it

People know the trainer or their manager may read the answers. That can pull ratings up.

The ceiling

Ratings bunch near the top and stop telling sessions apart. One speaker feedback platform reports average ratings of 99.1 to 99.4 out of 100, almost the same whether a talk drew few replies or many.

Source: Talkadot, State of the Speaking Industry 2026

One number

A form often asks for one rating for a session, or for a day. It does not show how things changed from one part to the next.

The asking

Each survey costs people time. Gallup names survey fatigue as a risk of asking too often, and warns that asking without acting can backfire.

Source: Gallup

Nobody is asked to fill anything in. That does not mean everyone is counted. Each report says how many people the camera could see clearly, and treats that count as a floor, not a headcount. It cannot say what share of the group that was.

How it works

Three steps. No forms.

  1. Scope it

    Tell us what you need to know and what you are allowed to measure. Together we agree the measures, the notice people see, and who gets the report.

  2. Run the session

    We set up a camera and measure the session live. People are told before it starts. Boost ships opt-out stickers for anyone who wants out.

  3. Get the report

    A report on the group: what was measured, what was not, and why.

The opt-out

Wear the sticker. You are left out.

Anyone who does not want to be measured wears a small red sticker on the upper chest. The system looks for it and leaves that person out of the numbers. Boost ships the stickers with each engagement.

Size
2 inches by 2 inches.
Colour
Red, with a matte finish. Not glossy.
Border
A thick black frame just inside each edge.
Centre
A black circle, 30% to 40% of the sticker width, with a light logo inside.
Where
Near the collarbone or upper chest, where the camera can see it.

The camera has to see it clearly. A jacket, hair, a lanyard or a badge can hide it, and at the back of a large room it may be too small to pick up. A home-printed copy may not be picked up either. Use the stickers we ship. The sticker is not the only protection: reports cover the group, not one person.

The Boost opt-out sticker. A red square with a black border. A black circle in the middle holds a white dragonfly logo.
The sticker print artwork, enlarged. The real sticker is 2 inches square.
What you can measure

Scoped to your question.

For each engagement, we agree which measures answer your question and how the session is split up. These three examples show the shape of the work. They are illustrations, not customers, and show no results.

Learning and development

A two-day leadership programme

The question
How the group behaved in each module, and whether day two looked like day one.
What would be measured
Which way the group faced, module by module, across both days.
What the report would show
A module-by-module view of both days, with the stretch where the fewest people faced the work marked for the next run.

Learning and development

A required compliance session

The question
How the group behaved through a required session, section by section.
What would be measured
How steadily the group faced the material, section by section.
What the report would show
The sections where the fewest people faced the material, so those get looked at first.

HR and people teams

A quarterly cycle run on a pulse survey

The question
A second source beside the pulse survey, without sending one more form.
What would be measured
The same group measures at each quarterly company meeting, taken the same way each time.
What the report would show
Quarter by quarter, what the group did, set beside what they said in the pulse survey.
Measuring people at work

We measure behaviour. We do not infer what anyone feels.

The EU AI Act bans AI that works out how people feel at work. GRW does not infer emotions, mood or morale. It measures what people visibly do, such as which way they face and how many are in view, and reports that behaviour. We do not run workplace reads in the EU or the UK.

That is why this is not an engagement score. Engagement as a feeling is something only people can tell you. We show what the group did, so you can set it beside what they said.

Source: EU AI Act, Article 5

Limits

What it does not do.

  • It does not score or name a single person. Reports cover a group of four or more.
  • It does not infer emotions, mood, morale or sentiment. It does not measure engagement as a feeling either. It measures what people visibly do.
  • It does not count everyone. Some faces are missed, so the count is a floor, not a headcount.
  • It does not guess. When the footage is too thin for a measure, the report says "not measured".
  • It does not predict performance, retention or results, and it does not measure what anyone learned.
  • It cannot rule out that people act differently because they know a camera is on.
  • It records no video. We keep the numbers.
  • It is not offered in the EU or the UK. Workplace reads are refused there.
How sure are we?

Read the evidence, including the gaps.

Each GRW score is preliminary. The validation dossier, written by a skeptic, says what is established, what is not, and what we do not claim yet.

Two products, two jobs

GRW Business or GRW Data Collection?

GRW Business

How a manager and a team work together, week to week. It is about people and how they work.

See GRW Business

GRW Data Collection

Retiring a measurement form across a programme. It is about the instrument, and it reports on groups.

Start here

Tell us what you need to know, and what you are allowed to measure.

Each engagement is scoped with you. There is no price list and nothing to sign up for. A real person reads your note and replies.

Built for organizations and teams. By submitting you agree to our Terms and Privacy Policy.

Common questions

Questions, answered.

Is this an engagement survey?

No. A survey asks people what they think. GRW Data Collection measures what a group did in a session. It does not measure engagement as a feeling, and it does not ask anyone anything.

Does it read how people feel?

No. The EU AI Act bans AI that works out how people feel at work. GRW does not infer emotions. It measures behaviour, such as which way people face and how many are in view, and reports that behaviour.

Can a manager see one person’s results?

No. Reports cover a group of four or more. They do not name or score a person.

How do people opt out?

They wear the opt-out sticker on the upper chest, where the camera can see it clearly. The system looks for it and leaves that person out of the numbers. Hair, a jacket or a badge can hide it. Boost ships the stickers with each engagement.

What does it cost?

Each engagement is scoped with you, so there is no price list. Tell us what you need to know and we will quote it.

Is this the same as GRW Business?

No. GRW Business looks at how a manager and a team work together, week to week. GRW Data Collection is for retiring a measurement form across a programme.
Next step

One conversation. No forms.

Tell us what you need to know, and what you are allowed to measure.