Agile Retrospectives: The Safety Check

This is post #3 in a series dissecting agile retrospectives, see the introduction here

What is it?

The safety check, typically performed at the start of a retrospective, is a tool often used by a facilitator to gauge the safety of the participants, to see how comfortable they feel in expressing their opinions and getting involved in discussion and is often used to make a judgement call on whether or not the retrospective should continue (I’ll give examples of how this is done later in the post)

Before I go into any more specifics about how to best perform a safety check (and if it’s even necessary) I want to first talk a little about why we do one – because in order to know if something is necessary, you must first know why you’re doing it.

This really is a case where the destination is far more important than the journey. I don’t care how you achieve the goal, that is not important, what is more important is that we understand the objective the safety check is attempting to fulfil and that the goal is achieved. If you find another way to fulfil the same objective, then that’s great!


The safety check has two main objectives; the first, most obvious objective of the safety check is, as I mentioned above, a way for the facilitator to get a handle on how safe everyone is feeling, in order to make a decision whether the retrospective should continue or not. If the safety is too low, well then it indicates that people aren’t really comfortable in expressing their thoughts, meaning that you can’t really expect any real value to come of the retrospective. Continuing with the retrospective under such conditions would truely make the whole thing empty ceremony – if you have information that tells you that the process will not be valuable, then the only logical response is to not do it. Low safety indicates there are other more pressing issues that the team needs to resolve.

There is another, less understood reason for the safety check. It is just as valuable to the participants as it is for the facilitator. When I facilitate, once the safety check is complete, I like to take a step back and direct the teams attention to the results. The team can use this information to gain awareness of the people around them. By looking at the results of the safety check, we remind ourselves that these are real people around us, with real feelings and emotions. We remind ourselves that while we are talking and discussing, that there are others around us that may be happier just to listen and might prefer not to talk.

I can think back to quite a few retrospectives that I’ve participated in, where either the facilitator or even another participant has put people on the spot and called them out – “You’ve been very quiet there, what do you think?”, or “Do you have anything to share with the group?”. Don’t do this. and tell the participants not to do this too.


The most common safety check is performed by handing everyone a post-it note and a pen. Then ask everyone who is participating to write a number between 1 and 5 (in private) on the post-it to show how safe they feel. Once done, fold the paper in half and then hand it to the facilitator. The facilitator then proceeds to plot all toe post-its up on the wall, visible for all to see.

A general rule that I try to follow for all communication is try to leave as little ambiguity as possible. To this effect, its often helpful to have a slide or poster or some other mechanism to make it clear what the numbers mean. Here is an example of a slide I’ve created and used in the past, the wording is my own, feel free to change it to whatever suits you if you disagree with my description: the important this is that everyone is on the same page as to what the numbers mean.

Just to keep things interesting, here’s a (conceptually similar) alternative safety check format that I’ve used in the past.

The ESVP safety check is very similar to the 1-5, just with a different twist.The participants put one of the letters E,S,V or P on the post-it:

E (Explorer). here to explore both their own and others thoughts and feelings. Happy to talk about anything

S (Shopper). Happy to observe and interested in what people have to say about all topics. If something that they care about is discussed then they will get involved and talk.

V (Visitor). Not keen on talking, just happy to sit back and observe.

P (Prisoner). Not comfortable being here at all. Only here because they have to be, if they had a choice they would be somewhere else.

A few extra notes on performing the safety check:

1. Give everyone the same colour post-it and same color pen
2. Just because you’re facilitating, doesn’t mean you have the right to peek at people’s safety checks! – as shameful as it is, I have actually seen someone do this!
3. collect all post-its first before you start to put them up – It sounds crazy but I have personally had experiences where the facilitator would take each post it one by one and put them up on the wall one by one. totally defeats the purpose.


It’s funny how after we’ve done something a few times, our brain almost switches off and goes into autopilot. I feel the same thing happens with retrospectives, and I’m just as guilty of this as the next person.

It was only after I decided the actually stop and think about this in details that all the small details started to dawn on me, like when to do the safety check. Typically I’ve seen it performed right at the start of the retrospective. I now think this is a bad idea. How can I accurately tell you how safe I think I will feel over the next 1 hour, if I don’t yet know what I’m going to be doing for the next hour?

Before you do the safety check, its essential you first explain to the group how you plan to run the rest of the retrospective so they can accurately give you a safety reading. Remove that ambiguity!

To drive the point home, take the following totally fictitious (and exaggerated) scenario. You do a safety check and all 10 participants give you a safety of 4. “Fantastic!” you think to yourself, and proceed. A 4 means that safety is pretty high, and apart from a highly sensitive topic or two, people will talk about anything.Now imagine that every single person has the same topic they are not comfortable discussing, team communication. Safety check done, you proceed to tell the group that this week you’ll be running a focused retrospective (something that I have been asked to do in the past – a retrospective concentrating on a particular hot-topic) on …you guessed it…team communication. All of a sudden all those 4’s turn to 1’s.

What if safety is too low?

Finally, I want to end with a very brief set of instructions on what to do if the safety is ever too low. This probably doesn’t happen very often, but every good facilitator needs to be prepared for this eventuality, and you’ll be surprised by just how few actually know what to do in such a scenario.

step 1: Don’t ignore it. The worst thing you can do is brush it under the carpet. Be transparent and show the group that you think the safety is too low. Don’t allow the retrospective to turn into empty ceremony.

step 2: Gather suggestions on how to improve safety. Once you’ve made it clear that safety is low, its time to veer away from your planned schedule. Tell the group that you will do another exercise and ask everyone to take a post-it (again, same colour and same colour pens) and to write down any suggestions they have to improve their own personal safety. This is important: tell them not to assume what will make others safe, this should be personal and they must only write down things that will make them feel safer. Another tip to make everyone feel safe while they do this: tell everyone to write something down. Even if they have no suggestions or their safety is high, tell them to write something to that effect. This way every person writes something down and no-one stands out.

step 3: Redo the safety check. Once you’ve collected all the suggestions, go through them. If there is anything there that is in your power to change, then do it. Once you’ve executed the suggestions, re-do the safety check – don’t assume that by simply fulfilling some suggestions that everyone is feeling safe.

step 4: Be prepared to stop – If after this safety is still too low, then just stop. While retrospectives are an invaluable tool, they’re also a tool to be used by a relatively high functioning team. Low safety indicated that there are bigger issues that the team needs to address first before they can be ready to achieve the benefits of retrospectives.

note: I have to confess I’ve avoided the topic of how to judge what exactly low safety looks like. This is something I’m still trying to find the answer to and as yet I don’t feel confident enough giving you an answer. Right now I’m relying on a gut-feeling, but I can’t articulate it. If you have any ideas for me, I’d love to hear them :)

Final thoughts

Is a safety check always necessary? I’m not entirely sure. I’m not going to say that you must always perform one. I can understand that there could be situations out there that don’t warrant the use of a safety check. As a general rule, I would always start with doing it, and then over time perhaps it becomes less necessary. Personally I would still veer on the side of keeping it, because if someone is not feeling safe, then they’re unlikely to speak up and ask for a safety check is one isn’t being performed. It’s just safer to do it in my opinion.

The aim of this post was to highlight the reasons for doing a safety check. Once you know what purpose the safety check fulfils, I leave it to you to decide whether your team would benefit from it or not, or perhaps you know another way to achieve the same objectives. What matters is that the objective is met, now how it’s met.

About the Author: akash


  1. Pingback: Agile Retrospectives Part 1: Useful Tool or Empty Ceremony? | blog

  2. Pingback: Agile Retrospectives: The Facilitator | blog

  3. Reply Ilias Bartolini

    Adding a note on “when”.
    I also think is useful to add that the safety check should be after the prime directive and after ground rules about being open to collaboration are set.

    In the book “The Skilled Facilitator” there are a couple of chapters on “discuss the undiscussable” and “manage emotions”.

    Before stopping is also important to ask to the group if instead of continuing the retrospective they want to discuss the issue or issues that are impeding the team to function and if they want to change the goal of the meeting.

    If I already know the team, I don’t see any potential malfunction in the group and I know that the stakes are not high I skip the safety check most of the times.


  4. Reply Jeremy lu

    Hi Akash,

    I think a safety check is really great. Nothing beats the anonymous factor don’t you think?

    @ Ilias, looks like you have a good team if there is no potential malfunction. I guess the keyword is potential. I often think everything is going okay, but this is just my perception and assumption.

    If you are interested, the tool I use is here

    • Reply akash

      Never looked at groupmap before, personally I would recommend one of the free tools out there for dealing with distributed retrospectives.

      I’ve worked on a number of projects where the team has been distributed across multiple locations, in these situations I’ve used idea ideaboardz – (combined with a video call so everyone can see each other)

  5. Reply Eddie

    Very helpful insights, thanks! I’ve found that it’s important to treat the safety check score as non-linear, i.e. below a certain threshold you have to take a complete different approach, and as you say it’s also contextual, some topics people may feel less safe discussing than others. In one instance, a team scored very low so we couldn’t even proceed with diagnosing how to improve safety; they were caught in a Catch-22 situation. We had to back out of the retro and take a completely different approach — I set up an anonymous survey and solicited feedback that way, this enabled me to start addressing unsafe feelings in a safe way.

    • Reply akash

      Thanks Eddie, I’m glad you found it useful :)

      I agree, for example I’ve worked in small teams of 2 or 3 people, and in these situations we’ve still gotten the benefits of considering what lessons we can learn from our past actions without the need for some of the heavier process like the safety check, and instead just a more informal conversation.

      The worst thing you can do is run the safety check, see that the safety is low and then just ignore the issue and move on. I think that you backed out of the retro and attacked the safety issue directly is a great example of how to deal with low safety!

  6. Pingback: Safety Check in a Retrospective: How to handle a low result? | Johannes Thönes

  7. Pingback: Samenwerken en vertrouwen | Ik ben

  8. Pingback: Hany Elemary – Should you retro the retro?

Leave a Reply

Your email address will not be published. Required fields are marked *