Booking a psychological safety session? Take a baseline first. Three measures to use, when to run them and how to match the session format to the result.
Link copied
Psychological safety has moved from academic journals into leadership programmes. Many HR and L&D teams now want a keynote or workshop on it. Few of them measure it before the session, which makes the session's effect impossible to judge afterwards.
Psychological safety, in Amy Edmondson's 1999 definition, is "a shared belief held by members of a team that the team is safe for interpersonal risk taking". In practice, it means people raise problems, admit mistakes and ask questions without expecting to be punished or embarrassed.
This piece is for the person who has to justify the booking. It covers three measures you can baseline, when to run them, and how the result should shape the format you book.
The definition above comes with a measure. Edmondson's original study used a seven-item team scale, rated on a seven-point scale from "very inaccurate" to "very accurate". Three of the items give a sense of it:
The scale is still the standard. The CIPD's 2024 evidence review on psychological safety notes that it can be measured with seven items adapted from Edmondson's scale. Google's re:Work guide on team effectiveness, which lists psychological safety first among five team dynamics in order of importance, prints the same seven items.
Because the measure is at team level, it tells you which teams need attention, and it tells you more than an organisation-wide average would.
Three reasons.
First, a baseline lets you judge the session. Without a before figure, the only evidence afterwards is how much people enjoyed the talk.
Second, the baseline tells you which problem you are solving. UK data points to voice and conflict as common gaps. In the CIPD Good Work Index 2026, based on 5,074 UK employees, 20% said they have no channel at all to express their views at work. A quarter had personally experienced conflict at work, and a majority of those said they simply let it go. In the 2025 edition, only 37% said managers allow them or their representatives to influence final decisions.
Third, it changes the brief. A speaker who knows the baseline can talk to the gap your teams actually have. Our guide on how to write the speaker brief covers what else belongs in it.
You do not need a new survey platform. Three existing measures cover the ground:
| What you measure | Tool | What it tells you | Effort |
|---|---|---|---|
Team climate | Edmondson's 7-item team psychological safety scale | Whether people feel safe to raise problems and take risks within their own team | Low: 7 questions, per team |
Voice | Questions on whether a voice channel exists, and on whether managers keep people informed, seek their views, respond to suggestions and allow influence on decisions (as used in the CIPD Good Work Index) | Whether views travel upwards and whether anything happens to them | Low to medium |
Risk context | HSE Management Standards Indicator Tool, 35 items, focusing on the Relationships and Support areas | Whether psychosocial risks sit behind a low climate score | Medium |
The HSE Management Standards cover six areas of work design: demands, control, support, relationships, role and change. They sit comfortably alongside ISO 45003:2021, the international guidance on managing psychosocial risk within an occupational health and safety system. If your organisation already runs the Indicator Tool as part of stress risk assessment, you may have the risk-context baseline already.
A simple rhythm works for most programmes:
Keep the survey short and anonymous, and report at team level only when the team is large enough to protect individuals.
A keynote can give people a shared language and make the topic legitimate to discuss. It is unlikely to move a team's climate score on its own, because climate is built in everyday interactions with a manager.
A half-day workshop that ends with each team committing to one specific behaviour, followed up by the team's manager, has a better chance of shifting the score.
That is why the measure should follow the people who shape those interactions. If climate scores are low in specific teams, the managers of those teams are the audience that matters most. The same logic applies to wellbeing programmes, as we set out in why a wellbeing programme is not built in one session.
Workplace culture and how teams work well together are the subject of this conference talk:
Bruce Daisley at The Economist Innovation@Work conference (YouTube)
The baseline should decide the format:
| Baseline result | Likely issue | Format that fits |
|---|---|---|
Low team climate, concentrated in some teams | Manager behaviour in those teams | Workshop series for the managers concerned, with a follow-up measure |
Reasonable climate, low voice scores | Channels upwards are missing or ignored | Leadership session on listening and acting on input, plus a visible change to how views are gathered |
Low scores across the board, poor Relationships and Support results | Wider psychosocial risk | Risk assessment first; a speaker session supports it, and should not be the first step |
Good scores overall | Maintenance | Keynote to reinforce shared language, often combined with another theme |
Our piece on four formats and four different outcomes explains what each format can realistically deliver.
The most widely used measure is Amy Edmondson's seven-item team psychological safety scale from 1999, rated on a seven-point scale. The CIPD's 2024 evidence review notes that psychological safety can be measured with items adapted from it. Run it at team level, before and after an intervention, with identical wording.
Without a before figure, the only evidence afterwards is how much people enjoyed the session. A baseline shows which teams and which issue need attention, sharpens the brief, and gives you a comparison for the budget discussion.
A keynote can create shared language and make the topic legitimate to discuss. Team climate is shaped in everyday interactions with managers, so lasting change usually needs follow-up work with the managers of the teams where scores are low.
In the CIPD Good Work Index 2026, based on 5,074 UK employees, 20% said they have no channel to express their views at work. A quarter had personally experienced conflict at work, and most of them let it go. The 2025 edition found that only 37% say managers allow them to influence final decisions.
A practical rhythm is four weeks before the session, six weeks after and three months after, using the same questions each time. The three-month result is the most useful for deciding whether to continue or change the programme.
After Dinner Speakers
Contents
