What this is
What is a peer observation?
What is a peer observation?
It is a behavioural safety observation carried out by one trained worker on another, with feedback given directly between them and no supervisor involved in the exchange. Only the counts and themes from the observation move upward into reporting; the specific individuals observed are not meant to be identifiable in that reporting.
Why does the programme rely on workers rather than supervisors?
Behaviour based safety practice consistently shows higher reporting rates and more candid at-risk answers when a supervisor is not part of the exchange, because workers are less likely to hold back for fear the observation becomes a performance record. Removing the supervisor is a deliberate design choice, not an oversight in who gets trained to observe.
Why must the observer be trained specifically for this?
An untrained observer tends to either avoid raising at-risk behaviour to keep the exchange comfortable, or turns the conversation into criticism that damages the trust the programme depends on. Training covers how to give feedback that is specific and non-judgemental, which is the skill the whole programme's credibility rests on.
Scope
When is a peer observation required?
This template is for the specific, no-supervisor, anonymised exchange that behaviour based safety practice depends on. Using it for anything supervisor-led or individually traceable defeats the reason the programme exists.
Use this template when
- A trained worker is carrying out a scheduled peer observation as part of the behaviour based safety programme
- The observation is happening peer to peer, with no supervisor present or involved in the feedback
- A new record is needed and each one takes its own ID in the form OBS-2026-000
- The counts and themes from this observation need to feed the programme's trend reporting
- You need evidence the peer programme is running as designed, without capturing who observed whom
Do not use it for
- Safety Observation, which is a short general record of a task noting safe practice and risk, open to any observer rather than restricted to trained peers.
- Behaviour Based Observation, which covers a wider agreed behaviour set and is not built around the no-supervisor, anonymised design this template depends on.
- Leadership Safety Walk, which is a senior leader spending time in the workplace talking with workers, the direct opposite of the no-supervisor design here.
- Critical Task Observation, which watches one named high-risk task against its written procedure rather than general workplace behaviour.
- Anything outside KnowSafe, which belongs in the workspace that owns that process
Compliance mapping
Which BBS programme practice requirements does this satisfy?
BBS practice is not itself a numbered code, so the map draws on the ISO 45001 provisions for worker participation and competence a genuine peer programme stands in for.
| Clause | Requirement | Where it lands |
|---|---|---|
| ISO 45001:2018 Cl. 5.4 | Workers must be consulted and enabled to participate in identifying hazards, not just have data collected about them by someone else. | Header |
| ISO 45001:2018 Cl. 7.2 | Anyone carrying out a competence-dependent safety function, including a peer observer, must be able to demonstrate the training behind it. | Header |
| BBS programme practice — Anonymity Principle | Individual worker identity must not travel with the observation record; only aggregated counts and themes move upward into reporting. | Header |
| ISO 45001:2018 Cl. 9.1.1 | The organisation must monitor the extent of compliance with expected safe behaviour through direct, first-hand observation data. | Observations |
| BBS programme practice — Immediate Feedback | Feedback on an at-risk behaviour must be given to the individual at the time of the observation, not batched or delayed. | Observations |
| ISO 45001:2018 Cl. 10.2 | Recurring themes identified through worker-level monitoring must feed a documented review process, not stop at the individual record. | Related records |
What it does not cover
- Names Recorded answered Yes, which converts an anonymised feedback exchange into a traceable personal record and undermines the reporting rate the programme is built to protect.
- Observer Trained answered No, which means the observation was carried out without the specific skill the programme's credibility depends on and should not be counted toward programme coverage.
- Feedback Given answered No on an At Risk observation with no note explaining why, which records the risk without doing the one thing that is meant to follow it.
- Response Received left unanswered where Feedback Given is Yes, which breaks the loop the field exists to close and leaves the exchange's outcome unknown.
- Themes Noticed left blank on a record that logged several At Risk observations in the same category, which discards the pattern the trend review is meant to receive.
Global
Peer Observation requirements by country
Peer-led behavioural observation is not a licensed activity anywhere, but the duties around worker consultation and how the data can be used differ enough between jurisdictions to matter.
OSHA anti-retaliation provisions, 29 CFR 1904.35
OSHA has warned that behaviour based safety data used to discipline or rank individual workers, rather than improve conditions, can amount to discouraging injury and hazard reporting.
Drift toward identifiable, traceable peer observation records edges into the reporting-suppression territory OSHA has explicitly flagged.
Health and Safety at Work etc. Act 1974, s.2(6), and the Safety Representatives and Safety Committees Regulations 1977
Employers have a statutory duty to consult workers on health and safety, and worker-led programmes like peer observation are one practical route to meeting it.
A functioning peer programme is consultation evidence only if it demonstrably runs as workers running it for each other, not management data collection wearing a worker-led label.
ISO 45001:2018, Cl. 5.4 — Consultation and participation of workers
The clause distinguishes consultation, a two-way exchange, from merely informing workers, and expects mechanisms giving workers a genuine voice.
A management-scripted or supervisor-shadowed peer programme does not meet Cl. 5.4's intent even with the form unchanged, because the clause is about the relationship, not the paperwork.
How to complete it
How to complete a peer observation, step by step
The fields are quick to answer. The programme's value rests on four judgement calls the form cannot make for the people running it.
On a large site, leaving the name field blank is usually enough. On a small crew or a night shift, area, time and behaviour category can identify the individual just as effectively as a name. Genuine anonymity on a small team sometimes needs coarser detail than the form defaults to, and that is a judgement the observer has to make.
Observer Trained should reflect current competence, not a record from whenever the programme launched. A worker trained two years ago who has never run an observation since is a different risk to the field's credibility than someone recently refreshed.
The design assumes feedback happens face to face while the behaviour is fresh. A Feedback Given answer of Yes that actually describes a delayed conversation has already lost the immediacy that makes peer feedback land differently to a supervisor's note.
Themes Noticed is free text, the easiest field to fill in vaguely and skip linking forward. The judgement is whether the person closing the record raises a Trend Review ID when the same theme has shown up before, rather than treating each observation as standalone.
What auditors find
Most common peer observation findings
These six findings turn up repeatedly wherever a peer observation programme has run for more than a few months.
| Finding | Clause | What fixes it |
|---|---|---|
| Names Recorded creeping to Yes as supervisors ask for accountability | BBS programme practice — Anonymity Principle | Lock the field default to No and require a documented programme-level exception, signed off above supervisor level. |
| Observer Trained answered Yes with training predating the current feedback method | ISO 45001:2018 Cl. 7.2 | Attach a refresh interval to the observer competency record rather than treating initial training as permanent. |
| Feedback Given marked No on At Risk observations with no reason captured | BBS programme practice — Immediate Feedback | Require a reason whenever Feedback Given is No on an At Risk observation, so the gap is visible. |
| Response Received left blank even where Feedback Given is Yes | ISO 45001:2018 Cl. 9.1.1 | Make Response Received mandatory whenever Feedback Given is Yes, closing the loop. |
| Themes Noticed filled in with generic text not matching any Behaviour Category logged | ISO 45001:2018 Cl. 10.2 | Prompt the theme field with the specific behaviour categories recorded in that observation. |
| Trend Review ID left blank across observations carrying the same recurring theme | ISO 45001:2018 Cl. 10.2 | Flag a theme once it repeats across a set number of observations, and require the Trend Review ID before closing. |
Case in point
Case in point: the peer programme that quietly re-identified everyone it promised to protect
A site running a peer observation programme for over a year had Names Recorded consistently answered No, with reporting rates genuinely higher than the supervisor-led observations they replaced. Six months in, a safety lead began adding shift, role and machine detail into Themes Noticed to make trend reporting more useful — 'night shift, packing line 2, rushing at changeover' rather than 'time pressure'.
On a three-person night shift, that detail was enough for anyone reading the trend report to work out who had been observed. Workers noticed within weeks, and at-risk reporting on that line dropped by more than half the following quarter, even though Names Recorded had never changed from No. The anonymity the form was designed to protect had been undone by a well-meant improvement to a different field entirely.
The template
The template, field by field
The form exactly as it installs. Every field, option, score and conditional rule is editable, and the links to other templates come with it.
4 sections
- Reference
- SAF-046
- Archetype
- Inspection
- Record ID
- OBS-2026-000
- Scoring
- Percent safe
- Direction
- High is good
- Singleton
- No
- Basis
- BBS programme practice
- Links
- Feeds trend review
- Tags
- Observation, Behaviour, Culture
- Sections
- 4
- Fields
- 24
- Follow up fields
- 1
- Repeating sections
- 1
- Links out
- 3
Header
11 fieldsObservation ID*
Auto sequence. Format OBS-2026-00000.
The record's own ID. Other templates point at this value.
Status*
Drives who this goes to next.
- Planned2 pts
- In progress2 pts
- Complete3 pts
- Deferred0 pts
- Open0 pts
- Closed3 pts
- Overdue0 pts
Peer To Peer
No supervisor involved. Feedback goes directly between colleagues and stays there. Only the counts and themes are reported upward.
Date and Time*
Observer*
Observer Trained*
- Yes3 pts
- No0 pts
Names Recorded*
Should be no. Peer programmes work because they are not personal records.
- No2 pts
- Yes0 pts
Site*
Site ID*
Format SITE-000.
Links to FDN-001 Site ID
Area*
The area within the site.
Exact Location
Drop a pin for anything hard to find.
Observations
Repeats5 fieldsBehaviour Category*
- Bypassing a control0 pts
- Not using PPE0 pts
- Line of fire0 pts
- Shortcut under time pressure0 pts
- Improvised tool or method0 pts
- Unsafe posture1 pt
Observation*
- Safe2 pts
- At risk0 pts
Element Tag
Feedback Given*
- Yes2 pts
- No0 pts
Response Received
Related records
1 fieldObservation Programme Review ID
The programme review this feeds.
Links to SAF-123 Review ID
Result
7 fieldsObservations Made*
Percent Safe*
Feedback Rate Percent*
Themes Noticed
Trend Review ID
Links to SAF-047 Review ID
Observer*
Signature*
SAF-046 · record IDs look like OBS-2026-000 · Feeds trend review
Open in KnowellaRun it with agents
From a document you fill in to a programme that runs itself
Running one observation is simple. Keeping the programme's anonymity intact as it scales, and making sure a repeated theme actually reaches a trend review, is where peer programmes tend to erode quietly.
Holds the peer observation library separately from supervisor-led records, aggregates counts and themes for reporting without exposing individual observations, and links recurring themes forward to the trend review.
Tracks which workers currently hold live peer-feedback training and flags a lapsed refresh, so the observer roster only draws from people actually competent to run the exchange today.
Keeps the evidence a consultation or worker-participation duty can point to — programme coverage, feedback rates, trend actions — separate from anything that could re-identify an individual observation.
Coordinates the crew, rolls completion and exceptions into one view, and holds every write for your approval before it touches a record.
This template lives in KnowSafe — safety and compliance. Incidents, hazards, permits, inspections and the critical controls behind them.
Meet KnowSafe→Glossary
Peer Observation definitions and key terms
- BBS
- Behaviour based safety — a family of observation and feedback practices aimed at reducing at-risk behaviour, dependent on trust between the people observing and being observed.
- Peer observation
- A behavioural observation carried out by a trained colleague rather than a supervisor, with feedback given directly and no individual identity carried into upward reporting.
- At-risk behaviour
- An observed action that increases the chance of harm compared with the safe alternative, recorded without attributing it to a named individual.
- Feedback loop
- The pairing of an observed behaviour with confirmation that feedback was given and how it was received, tracked as two separate fields rather than one combined outcome.
- Trend review
- The periodic review that aggregates themes across many individual observations to look for patterns a single record cannot show on its own.
FAQ
Frequently asked questions about peer observation
Should a supervisor ever be involved in a peer observation?+
No. The design specifically excludes the supervisor from the observation and feedback exchange itself. A supervisor can see the aggregated counts and themes coming out of the programme, but should not be present for, or informed of, an individual exchange.
Why is Names Recorded a field on the form at all, if it should always be No?+
Having it as an explicit field forces a conscious answer rather than a silent default, and makes any drift toward recording names visible in an audit of the programme rather than buried in how people happen to fill in a comment box.
What counts as a trained peer observer?+
Someone who has completed training specific to giving direct, non-judgemental peer feedback, ideally refreshed periodically rather than completed once at programme launch. Observer Trained should reflect current competence, not a historical certificate.
What happens to the themes this template surfaces?+
Recurring themes are meant to feed the Observation Trend Review, which looks at patterns across many individual observations rather than any single one. A theme left sitting in a free-text field on one record achieves nothing on its own.
Does a low percent safe score mean the workforce is behaving unsafely?+
Not on its own. It can also mean the peer programme has enough trust for people to report honestly, which is the outcome the programme is designed to produce. A percent safe score should always be read alongside the feedback rate and the trend, not in isolation.
Can this template be used for a supervisor-led programme instead?+
It can be adapted, but doing so removes the design feature — no supervisor, no names — that behaviour based safety practice credits with the higher reporting rate. A supervisor-led equivalent is better served by the Behaviour Based Observation template, which does not carry the same anonymity assumptions.
Keep going
Related templates and programmes
Industries this is written for
Programmes this belongs to
Used together in Safety Culture and Engagement
Improvement Suggestion Record
Captures an improvement idea from anybody, with what it would change and who could own it
Safety Observation
A short record of how a task was actually done, noting both safe practice and risk
Behaviour Based Observation
Looks at a specific set of behaviours agreed with the workforce, such as line of fire, body position and tool use
Leadership Safety Walk
A senior leader spends time in the workplace talking with workers about how the work really goes
Safety Coaching Record
Records a coaching conversation after an at risk practice was seen
Critical Task Observation
Watches a high risk task from start to finish against its procedure
More in Observations
Safety Observation
A short record of how a task was actually done, noting both safe practice and risk
Behaviour Based Observation
Looks at a specific set of behaviours agreed with the workforce, such as line of fire, body position and tool use
Leadership Safety Walk
A senior leader spends time in the workplace talking with workers about how the work really goes
Safety Coaching Record
Records a coaching conversation after an at risk practice was seen
Critical Task Observation
Watches a high risk task from start to finish against its procedure
Observation Trend Review
Pulls together observation results over a period to show what is really happening across the site
Written and reviewed by
Siddarth Singh
Founder & Chief Executive Officer, Knowella
Certified Safety Professional and industrial and systems engineer with more than a decade inside food supply chain, freight and manufacturing operations. This page was written against the current text of the standards it cites, not against secondary summaries of them.
- Certified Safety Professional (CSP), Board of Certified Safety Professionals
- MBA, University of Chicago Booth School of Business
- MS and BS, The Ohio State University, Industrial and Systems Engineering
- Six Sigma Black Belt
Sources and last review. Reviewed 16 August 2026 against:
- ISO 45001:2018 — Cl. 5.4, 7.2, 9.1.1, 10.2
- OSHA — 29 CFR 1904.35, employee involvement and anti-retaliation
- Health and Safety at Work etc. Act 1974 — s.2(6), consultation duty
- Safety Representatives and Safety Committees Regulations 1977 (UK)
This page is general guidance, not legal advice. Confirm requirements with your jurisdiction’s regulator.