Observer training
How to score what you see
This guide explains how to use the field observation tool. You don't need neuroscience training. You only need to know how to look, count, and record what happens in front of you.
The only scale you will use
0
None
Not observed
Not occurring or you cannot confirm it. No one in the visible audience shows this behavior.
1
Sporadic
~5–15% of visible audience
A few people. You see it, but it's isolated and doesn't set the general tone of the room.
2
Frequent
~15–35% of visible audience
A notable group. You clearly see it, it catches your attention. No longer isolated.
3
Majority
+35% or very noticeable
Most people or something so visible it's impossible to ignore, even if it's a single person.
Room percentage calculator
Most important: you score based on the visible audience from your position, not the total room capacity. If you're at the back of a 200-person room and can only clearly see 60, those 60 are your reference. A 3 doesn't mean 70 people are yawning — it means that of the people you can see, most of them are.
Concrete examples — same indicator, all 4 levels
B1
Visible yawns
negative signal
0
None
None in the 2-minute observation period.
1
Sporadic
1–2 yawns or less than 5% of visible audience. With 100 visible: up to 4 yawns. With 20: only 1.
2
Frequent
5–15% of visible audience, or several in chain. With 100 people: 5–15 yawns. With 20: 2–3 is already a pattern.
3
Majority
Over 15% of visible audience, or contagion effect across an entire zone. With 100 people: 15+ yawns.
A1
Active visual orientation toward content
positive signal
0
None
Most are looking elsewhere. The speaker is talking to an empty room.
1
Sporadic
Only a few people look toward the screen or speaker. The rest are scattered.
2
Frequent
A good portion of the room is oriented toward the content, though with some distractions.
3
Majority
Nearly the entire room looks at the speaker or screen. Visible and sustained collective attention.
When in doubt between two scores, go with the lower one. It's better to record a 1 that might be a 2 than to inflate the data. A conservative score is always preferable. Data integrity matters more than apparent precision.
Complete catalog
What each indicator measures
Click any indicator to see what each score means and a practical example. Indicators marked in amber are ambiguous — they require a contextual note.
A · Attention and focus
Visual signals that attendees are actively paying attention to content. These are the most important indicators in cognitive blocks.
A1
Active visual orientation toward content
▼
0
None
No one looks toward the speaker or screen.
1
Sporadic
Only some people oriented. Most are scattered.
2
Frequent
Many people look toward the content, with visible distractions.
3
Majority
Nearly the entire room oriented. Clear collective attention.
Real examples
04:30 PM, post-lunch. Most people have their heads down or look at phones. The speaker is talking but no one is watching.
1Dense presentation, hour 3. About 10 out of 70 visible people still look at the screen. The rest are scattered.
2Opening keynote, 30 min in. Most look forward, though a group in the back rows is on their phones.
3Speaker shares a powerful case study. The entire room is silent, looking at them. No one checks their phone.
A2
Active posture (upright or leaning toward content)
▼
0
None
Mostly reclined posture, leaning back or slouching.
1
Sporadic
Few people with upright posture. Most have relaxed into their chairs.
2
Frequent
Balanced mix. Many upright, many reclined.
3
Majority
Most with upright torso or leaning forward. Sign of active engagement.
Real examples
0Afternoon session, hour 4. Most reclined, some practically lying on the armrest.
3Workshop with team exercise. Everyone leaning toward the table, writing or actively talking.
A3
Personal devices not linked to the event
▼
0
None
No one uses their phone or laptop in a way unrelated to the content.
1
Sporadic
1–3 people on their phones. Isolated and not noticeable.
2
Frequent
A visible group uses phones or laptops outside the event context.
3
Majority
Over a third of the room on phones without the speaker requesting it.
⚠ Ambiguous indicator — always add a contextual note: If the speaker asked them to look something up, if they're voting in an event app, or taking content-related notes on their laptop, the score should be 0 and you must explain it in the note. Don't score device usage — score usage unrelated to the event.
A4
Active note-taking or marking
▼
0
None
No one takes notes. Notebooks closed, pens still.
1
Sporadic
2–4 people writing. In a 60-person room, that's less than 7%.
2
Frequent
A visible group is taking notes. Sign that the content generates active engagement.
3
Majority
Most are writing or marking. The speaker just shared something truly valuable or actionable.
Calculation example
1You see 80 people, of which you can clearly observe about 60. Of those 60, you estimate 12 are writing → 20% → score 1 (not 2, because 20% is right at the lower boundary).
B · Cognitive fatigue
Signs that the brain is reaching its processing limit. More common in long blocks, dense content, or post-meal segments.
B1
Visible yawns
▼
0
None
None in the 2-minute observation.
1
Sporadic
1–2 yawns or less than 5% of visible audience. Isolated, not a group pattern.
2
Frequent
5–15% of visible audience, or several in rapid succession. Already a room-level pattern.
3
Majority
Over 15% of audience, or visible contagion effect across an entire zone.
B2
Head drop / face resting on hands
▼
0
None
No one rests their head or passively leans down.
1
Sporadic
1–2 people with face in hands or head dropped toward the table.
2
Frequent
3–5 people in that posture. Clear sign the block is too long or dense.
3
Majority
More than 5 people or people practically asleep. Maximum alert.
B4
Off-topic side conversations
▼
0
None
No one chats with their neighbor about unrelated topics.
1
Sporadic
A pair whispers about something unrelated. Isolated.
2
Frequent
Several visible parallel conversations. Background noise is rising.
3
Majority
General murmur. The room is split between those listening and those chatting.
⚠ Note: If the speaker proposes a pair exercise or asks a discussion question, conversations are positive, not negative. Score as 0 and note it. It's only B4 when the conversation is unrelated to the block's content.
D · Activation state
General energy and alertness level of the room. Present in all block types.
D2
Visible drowsiness
▼
0
None
No one appears drowsy. Eyes open, active postures.
1
Sporadic
1–2 people looking drowsy: drooping eyelids, nodding head.
2
Frequent
A clearly drowsy group. Coincides with post-meal or late afternoon segments.
3
Majority
The entire room looks like a lost battle. Massive drowsiness.
When to note D2 instead of B1?
!B1 (yawns) is an active, observable signal at a specific moment. D2 (drowsiness) is a general state: the person isn't yawning but has half-closed eyelids, drooping head, or the room's overall pace is slow. You can score both if both are happening.
D3
Positive emotional expressions
▼
0
None
Neutral or serious faces. No visible emotional reaction.
1
Sporadic
Some isolated smiles or nods.
2
Frequent
Visible group smiling or nodding. Emotional connection with content.
3
Majority
General laughter, spontaneous applause, visible expressions of positive surprise throughout the room.
When to observe
Checkpoints for each block
A checkpoint is the moment within a block when you record an observation. You don't observe all the time — you observe at specific moments with specific indicators.
A complete checkpoint observation should take 60 to 90 seconds: scan the room, estimate visible audience, score the active indicators for that checkpoint, and add a note if something is ambiguous or noteworthy.
Checkpoints by block type
| Checkpoint | Block type | When to do it | What you look for |
|---|---|---|---|
| Baseline | 🧠 Cognitive | First 3-5 min of the block | Energy level on entry. Does the room start well or carry fatigue from the previous block? |
| Mid-early | 🧠 Cognitive | At 1/4 of block duration | First signs of attention or dispersal. Has the speaker hooked the audience? |
| Mid | 🧠 Cognitive | At the midpoint of the block | Central state. The most representative moment of the entire block. |
| Late | 🧠 Cognitive | Last 10-15 min of the block | Did attention hold or degrade at the end? Signs of accumulated fatigue. |
| Post-break | 🧠 Cognitive | First 5 min after a break | Did the break help recover attention? Was the restart effective? |
| Transition | 🧠 Cognitive | Between blocks | Energy in movement. Does the room reactivate or drag the fatigue? |
| Block start | 🍽️ F&B | First 5 min of service | Does it start with active conversation? Does the service pace allow relaxation? |
| Block end | 🍽️ F&B | Last 5 min of service | Did energy recharge? Is there rush or discomfort from the pace? |
| Opening | 🤝 Social | First min of the activity | Does the room start with active participation or wait to be pushed? |
| Development | 🤝 Social | Midpoint of the activity | Are isolated subgroups forming? Is there genuine connection or just performance? |
| Close | 🤝 Social | Last min of the activity | Energy leaving the block. Do they leave energized or neutral? |
| Spot check | 🧘 Break / Snapshot | Whenever you see fit | Quick read of general energy. Not diagnostic, just directional. |
You don't have to do every checkpoint in every block. If the block is 20 minutes, a Baseline + Late may suffice. The platform shows which checkpoints are available for each block type — choose the ones that make the most sense given the time available.
Practical exercises
What would you score here?
Real room situations. Read each description and choose the correct score for the indicated indicator. You'll see feedback immediately.
Question 1 · Indicator B1 — Yawns
It's 4:15 PM. Content block after lunch. The speaker has been talking for 35 minutes without a break. In the last 2 minutes you counted 4 yawns: one in row 2, two in row 5, one more on the side. The speaker's pace is uniform with no interaction.
What score do you assign to B1 (Visible yawns)?
Question 2 · Indicator A3 — Devices
Presentation on digital strategy. The speaker just said: "Open your phones and search for hashtag #strategy2026 to vote." 40 out of 55 visible people take out their phones and start typing.
What score do you assign to A3 (Devices not linked to the event)?
Question 3 · Indicator A4 — Note-taking
Presentation on team management, 11:00 AM. You have a clear view of 70 people. The speaker just shared "5 keys to talent retention." You count approximately 18 people writing in notebooks or laptops.
What score do you assign to A4 (Active note-taking)?
Question 4 · Indicator D2 — Drowsiness
Hands-on workshop in groups, 10:30 AM. Attendees have been working in teams of 4 for 20 minutes. You see nearly everyone leaning toward the table, talking and writing. Only one person in the right corner has elbows on the table and face resting on hands, looking tired.
What score do you assign to D2 (Visible drowsiness)?
Question 5 · Indicator C1 — Room departures
Round table, last session of the day, 5:45 PM. Session is scheduled until 6:30 PM. In the last 15 minutes you've seen 8 people leave a room with 65 visible attendees. Some returned, others didn't. The moderator continues without mentioning the departures.
What score do you assign to C1 (Room departures)?
Before entering the room
The rules that never change
Five principles that always apply, with any indicator and in any block type.
01
Score what you see, not what you think
If you can't visually confirm a behavior, the score is 0. Don't estimate "they're probably tired." The tool measures observable external signals, not internal states.
02
Your denominator is the visible audience, not the total
You only evaluate people you can see from your position. In a 200-person room where you can clearly see 60, those 60 are your reference. Always log the visible audience count before scoring.
03
If in doubt between two scores, go lower
Conservative scoring protects data integrity. A 1 that might be a 2 is preferable to an unfounded 2. Inflated data makes it impossible to compare events.
04
A good note is worth more than an exact score
If context makes an indicator ambiguous (event-linked phone use, content-related conversations, etc.), write the note and score 0 or the most conservative score. The note is data; an incorrect score is noise.
05
Log the exact checkpoint time
The platform does this automatically, but verify the time is correct before saving. The timestamp is what allows cross-referencing observations with the event program to detect exactly when a signal appeared.
→
You are the eyes of the report
What you record here is what the final client will read. There is no other access to what happened in the room during the event. Your observation, well made and well recorded, turns the event into comparable and actionable data.
Before your first real event: practice in an internal meeting, a presentation, or any low-stakes environment. The goal is for the flow (scan room → estimate audience → select checkpoint → score indicators → add note → save) to fit comfortably in 60-90 seconds. That's what you'll have available at a real event.