View/Export Results
Manage Existing Surveys
Create/Copy Multiple Surveys
Collaborate with Team Members
OR
Sign inSign in with Facebook
Sign inSign in with Google

Team survey: 14 questions that measure team effectiveness

Two questions that describe the team, ten ratings on one line, two open items. About three minutes, written for a group of four to twelve who need each other to finish anything.

  • 14questions
  • about 3 minto answer
  • 4 to 12people it is built for
  • Every itemwritten for this page
Use this template Take the survey first
Team Survey · answer to move through it
Use this template
Question 1 of 14 The team you are describing
How many people are in the team this survey is about?
2 or 3
4 to 6
7 to 9
10 to 15
More than 15
Question 2 of 14 The team you are describing
How much of your own work depends on what other people in this team produce?
Almost none of it
A small part of it
About half of it
Most of it
Almost all of it
Question 3 of 14 Getting work through the team
When a piece of work has to move from one person in this team to another, it moves without anyone chasing it.
1
2
3
4
5
Almost neverAlmost always
Question 4 of 14 Getting work through the team
When two pieces of this team's work overlap, it is clear who finishes which part.
1
2
3
4
5
Almost neverAlmost always
Question 5 of 14 Getting work through the team
When something this team owns is going to be late, the rest of the team hears early enough to react.
1
2
3
4
5
Almost neverAlmost always
Question 6 of 14 Getting work through the team
The work that lands on this team is shared out across the whole team rather than the same few people.
1
2
3
4
5
Almost neverAlmost always
Question 7 of 14 Deciding and disagreeing
This team can name the two pieces of work that matter most to it this month.
1
2
3
4
5
Almost neverAlmost always
Question 8 of 14 Deciding and disagreeing
A week after this team settles something, people are working from the same version of what was settled.
1
2
3
4
5
Almost neverAlmost always
Question 9 of 14 Deciding and disagreeing
A disagreement inside this team ends in a decision rather than in silence.
1
2
3
4
5
Almost neverAlmost always
Question 10 of 14 Deciding and disagreeing
After something in this team goes wrong, the way the team works is different the next time.
1
2
3
4
5
Almost neverAlmost always
Question 11 of 14 What the team cannot fix by itself
Enough of the same people stay in this team for it to keep what it has learned.
1
2
3
4
5
Almost neverAlmost always
Question 12 of 14 What the team cannot fix by itself
This team can finish a piece of work without waiting on a decision from outside the team.
1
2
3
4
5
Almost neverAlmost always
Question 13 of 14 In your own words
What is the one thing that slows this team down most?
Question 14 of 14 In your own words
What does this team do well that would be missed if it stopped doing it?

Three things have to be true before the answers mean anything

A team survey averages people who need each other to finish work. Send it to a group who do not, and the average describes nothing anyone in the room can change.

The survey asks the three conditions before anything else. Questions 1 and 2 record the size of the group and how much of each person's work depends on what the others produce. The same rating means two different things in a group of four who share one deliverable and a group of twenty who share a manager.

Send it once per team, not once per department. Two teams answering one link produce a middling number that belongs to neither. If you do not yet know where one team ends and the next begins, an organizational network analysis survey asks who actually works with whom, and you draw the boundary afterwards.

If the group fails the three conditions, use a different template. For friction between two teams rather than inside one, the team collaboration survey questions are the ones people can answer. For the whole organization, once a year, use the staff engagement survey.

Isometric diagram of five blocks of different heights joined by short lines into one closed ring, with one connector picked out in coral, and a sixth block standing apart with no line to it
Five people joined by the work, and one who is not on this team.
The conditionWhat it means hereWhat this survey does about it
BoundedEveryone can name who is on the team and who is not, and gives the same list.Question 1 fixes the size, so you know how many responses to expect and can tell a quiet team from a partial one.
InterdependentThe work cannot be split into pieces people finish alone without losing something.Question 2 asks it directly. Answers at the low end mean the ten ratings describe a group rather than a team.
StableMembership holds still long enough for the group to keep what it learns.Question 11 measures it, and it is filed under what the team cannot fix by itself, because churn is a staffing decision.

The 14 team survey questions

Every item with its scale and what to do with the answer.

Plain text, one question per line, with the answer scale.

The team you are describing

Two questions before the ratings start. Not warm-up: the size of the group and how much its members need each other decide whether the twelve answers after them can be read at all.

  1. How many people are in the team this survey is about?Pick one of 5, requiredThe denominator. It tells you how many rows to expect, and how careful to be with the two open answers: in a group of four an answer has a name on it whether or not it was signed.
  2. How much of your own work depends on what other people in this team produce?Pick one of 5, requiredThe one that decides whether the rest is worth reading. Low answers across the group mean you have a reporting line rather than a team.

Getting work through the team

Four ratings on one 1 to 5 line anchored at Almost never and Almost always, each naming something a member could have watched happen rather than inferred.

  1. When a piece of work has to move from one person in this team to another, it moves without anyone chasing it.1 to 5, anchored at Almost never and Almost alwaysThe handover. Chasing is cheap to observe and expensive to leave alone, because its cost is paid by whoever chases rather than by whoever caused it.
  2. When two pieces of this team's work overlap, it is clear who finishes which part.1 to 5, anchored at Almost never and Almost alwaysThe seam. A low score here and a high score above means the team moves work along fine and duplicates it at the join.
  3. When something this team owns is going to be late, the rest of the team hears early enough to react.1 to 5, anchored at Almost never and Almost alwaysEarly warning, worded as what the team hears rather than as what an individual admits, so the answer is about a routine and not about somebody's courage.
  4. The work that lands on this team is shared out across the whole team rather than the same few people.1 to 5, anchored at Almost never and Almost alwaysLoad. Read this one against question 1: the same score means something different in a team of three than in a team of twelve.

Deciding and disagreeing

Four ratings about what happens to a decision after it is made. A team can act on these by itself within a month, which is why they sit apart from the two below.

  1. This team can name the two pieces of work that matter most to it this month.1 to 5, anchored at Almost never and Almost alwaysDirection, made checkable. It asks whether the team could NAME two priorities, which you can test in the next meeting, rather than whether it feels aligned.
  2. A week after this team settles something, people are working from the same version of what was settled.1 to 5, anchored at Almost never and Almost alwaysWhether a decision survives the week. A team can agree in the room and leave with three readings of what it agreed, and that shows up here.
  3. A disagreement inside this team ends in a decision rather than in silence.1 to 5, anchored at Almost never and Almost alwaysDisagreement, scored on its ending rather than on its temperature. A quiet team and a settled team answer this one differently.
  4. After something in this team goes wrong, the way the team works is different the next time.1 to 5, anchored at Almost never and Almost alwaysWhether anything changes after a failure, worded as a difference in how the work gets done rather than as a reflection habit.

What the team cannot fix by itself

Two ratings that belong to whoever set the team up. Keep them in and keep them apart from the eight above: read as a failing of the team itself, they produce a workshop about something no workshop can reach.

  1. Enough of the same people stay in this team for it to keep what it has learned.1 to 5, anchored at Almost never and Almost alwaysStability. Churn is a staffing decision, and its cost lands on the team as the same lessons learned twice.
  2. This team can finish a piece of work without waiting on a decision from outside the team.1 to 5, anchored at Almost never and Almost alwaysAuthority. If the team escalates to finish ordinary work, every other score here is capped by something outside the room.

In your own words

Two open items, both optional, both about the work rather than about a person. Read them; do not count them.

  1. What is the one thing that slows this team down most?Long text, optionalThe answer to read before any average. It is where the rating that surprised you gets its explanation.
  2. What does this team do well that would be missed if it stopped doing it?Long text, optionalHere so the survey returns something to defend as well as something to fix. A team that only hears its low scores stops answering.

Two swaps before you send. Put the team's name into question 1 and into the survey title, so nobody answers about the wrong group. For a short fortnightly repeat instead of one longer read, send questions 3, 7, 8 and 13 as a team pulse survey.

Reading answers from six people

A small sample by design, and small samples are read differently from big ones.

Do not average the ten ratings into one score. Ten ratings summed is a number with no unit, and it hides the pattern worth acting on: which items disagree. High on question 3 and low on question 4 says the work moves and gets duplicated at the seam. High on 9 and low on 8 says the team settles arguments and loses the settlement inside a week.

Read the spread, not the mean. Six people who all answer 3 and six who split three and three between 1 and 5 produce the same average and describe opposite teams. The split is the finding; the open answers say which part of the team it belongs to.

Say out loud that it is not anonymous. Six people means six rows, and an open answer can be recognised from how it is written whether or not it carries a name. This template collects no name and the platform adds none; neither fact makes an answer unattributable at this size. The choice between an anonymous, a confidential and an identified send is on the anonymous staff survey page.

What you collectWhat it is forWhat not to do with it
Questions 1 and 2, the size and the dependenceDeciding whether to read the rest, and how carefully to handle the open answersDo not skip them to save the team ten seconds. They are the only two items that can invalidate the other twelve.
Questions 3 to 10, eight ratings the team can act onPicking one thing to change this month, from the items that disagree with each otherDo not add them up, and do not rank teams against each other on them: the size and the dependence differ.
Questions 11 and 12, churn and outside sign-offTelling whoever set the team up what is capping itDo not hand these back to the team to work on. Neither is inside the team's control, and treating them as if they were is how a survey becomes a workshop about nothing.
Questions 13 and 14, two open itemsExplaining the rating that surprised you, and finding what the team would defendDo not code or count them. In a team of six there are at most twelve of them; read them.

Sending it, and what to do the week after

One link, one team, one change picked in front of the people who answered.

Load the fourteen questions, name the team, send one link. They are already in the online survey builder; the text below goes in front of question one and needs two edits, the team name and the date. Nothing chases anyone who has not answered, so that date is the whole reminder.

Bring the numbers back within a week, or do not send it. A team of eight can see that eight people answered, and answers that go quiet cost more than a survey nobody ran. Show the ratings with no names attached, name the two items that disagree, and leave with one change and one owner.

Two adjacent things this is not. If the subject is one recurring hour in the calendar rather than the team that holds it, use the meeting effectiveness survey. If a sprint has just closed, the team retrospective survey is the shorter instrument for that moment.

Do not run this twice in a quarter with nothing changed in between. The second send measures what the team thought of the first send. A fortnightly reading is a different length of instrument and a different promise.

What to send with the link

Before you start

This is about [TEAM NAME] and how work moves through it, not about any one person in it. Nobody is named or rated.

Fourteen questions, about three minutes.

It is not anonymous. There are [N] of us, [YOUR NAME] can open every response, and in a group this size an answer can be recognised from how it is written. Answer accordingly, and say less than you mean rather than nothing.

What happens next: I will bring the numbers to [DATE], with no names attached, and we will pick one thing from them to change.

Please answer by [DATE].

Team survey FAQ

What should a team survey ask?

Things a member could have watched happen: whether work moves between people without chasing, whether it is clear who finishes which part, whether a decision still means the same thing a week later, and whether a disagreement ends in a decision. Ask about the team as a unit of work rather than about the individuals in it, and keep churn and outside sign-off separate, because neither is the team's to fix.

How many people do you need for a team survey?

This one is written for four to twelve. Below four there is nothing to average, so have the conversation instead. Above about fifteen the group is doing more than one job and the answers blend two teams into a middling number: send it once per team and read the sets side by side.

Is a team survey anonymous?

No, and a template that says otherwise is selling you something. Whoever sets the survey up can open every response, and in a team of six an open answer can be recognised from its handwriting even without a name on it. Say that in the invitation rather than promising confidentiality the tool cannot deliver.

Why is there no question about psychological safety?

Because the measure of it is somebody else's. The seven-item scale is printed in the appendix of Edmondson's 1999 paper and is copyright Cornell University,[1] and the construct is identified with that scale closely enough that an item written to avoid it would be a reworded copy. Name it and cite it; do not paraphrase it into a template and call it original.

If the unit is not one team

Six templates with a different unit or a different moment. The full employee survey templates list sits on one page.

EmployeeEmployee Engagement SurveyWhen the unit is the whole organization and the reading is done once a year.20 questions, 6 min EmployeeCross-Team Collaboration SurveyWhen the friction is between two teams rather than inside one.8 questions, about 3 min EmployeePost Meeting SurveyWhen the thing being judged is one recurring meeting rather than the team that holds it.8 questions, 2 min EmployeeAgile Retrospective SurveyWhen a sprint has just closed and the team is about to talk about it.9 questions, about 3 min EmployeeEmployee Morale SurveyWhen the question is how people feel rather than whether the work moves.12 questions, 3 min SatisfactionCustomer Satisfaction SurveyThe general satisfaction read: one score, its drivers, and a worked scoring example.12 questions, 3 min

Methods and sources

Michael Hodge

Survey methodology and questionnaire design, SuperSurvey

Survey methodologist and editor since 2003: wording, response scales, reliability checks, reporting that maps to a decision.

  • All fourteen questions written for this page, then diffed by hand against the Gallup Q12 frames, the Edmondson psychological safety and team learning items, and all 38 items of the Lencioni Team Assessment
  • Psychological safety deliberately not measured, because its scale is somebody else's and a reworded version of it would still be theirs
  • The two questions outside the team's control kept in the instrument and separated from the eight inside it, so a low score is not handed back to the wrong people
  • No claim of anonymity anywhere on the page, in the invitation text or in the survey's closing message

Updated 9 September 2026 · How templates are reviewed

What this template is based on

Fourteen questions: two that describe the group, eight ratings about how work moves through it and what happens to its decisions, two ratings about conditions the group does not control, and two open items. The design has three rules. Every rated item names something a member could have watched happen, so nobody is asked to score a colleague's character. The items the team can act on are kept apart from the items only whoever set the team up can act on, because mixing them produces a workshop about staffing. And the two framing questions are asked first, because the size of the group and the degree to which its members depend on each other decide whether the ratings describe a team at all.

Question licensing

Nothing here is reprinted, and one thing is deliberately missing. The best-known measure of team psychological safety is a seven-item scale printed in the appendix of Edmondson (1999), which is copyright Cornell University and reached under terms permitting personal, non-commercial use only.[1] The construct is identified with that scale closely enough that an item written around it would be a reworded copy, so this template does not measure it and this page says so instead. The Lencioni Team Assessment, 38 items on a 1 to 5 scale, carries "Copyright (c) 2007 Patrick Lencioni. All rights reserved." on the assessment itself,[2] and permissions run through the publisher. Google's re:Work material on team effectiveness, including Project Aristotle, is governed by terms that state the Website Content "is intended for non-commercial use only" and forbid selling derivative works,[3] which does not reach a commercial template site however freely the material is published. CATME is proprietary to Purdue under an end-user licence, the Denison and OCAI instruments are proprietary, the Group Development Questionnaire is proprietary to its owner, and the NHS Staff Survey team items need express permission from NHS England. Each of those may be named and cited; none of them may be reproduced, and none of them is.

Limits and disclosure

This template collects ratings and computes nothing. There is no score, no index and no benchmark: the numbers mean something next to each other and next to the same team three months later, and nothing at all next to another team whose size and dependence differ. It cannot be anonymous. Whoever sets up the survey can open every response, and at the group size this instrument is built for, four to twelve, an open answer can be recognised from how it is written even with no name attached; the page and the invitation text both say so rather than promising a confidentiality the platform does not provide. Ten of the fourteen items are self-report from inside the team, so they describe what the team believes about itself, which is the thing a team can act on and is not the same as its output. Edmondson's study of 51 work teams in a manufacturing company found that learning behaviour mediated between team psychological safety and team performance;[1] that is one field study in one company, quoted here for what it says about the construct this template leaves out, not as a result about your team.

SuperSurvey builds the editor this template opens in. Nothing here depends on using it.

References

  1. Edmondson, A. C. "Psychological Safety and Learning Behavior in Work Teams." Administrative Science Quarterly 44, no. 2 (1999): 350-383. Copyright Cornell University. doi.org/10.2307/2666999
  2. Lencioni, P. The Five Dysfunctions of a Team: Team Assessment. San Francisco: Jossey-Bass, 2007. Copyright (c) 2007 Patrick Lencioni, all rights reserved; permissions handled by John Wiley & Sons. wiley.com
  3. Google re:Work, Privacy and Terms of Use: "The Website Content is intended for non-commercial use only." rework.withgoogle.com

Name the team, send the link

Fourteen questions, about three minutes, one change picked within the week.

Use this template