Accessing Surveys
Super Admins can navigate to the Surveys section from the left-hand menu. Surveys have three statuses:
Scheduled (upcoming)
Running (currently active)
Ended (completed)
Admin Permissions
Survey permissions are split into two separate controls, which Super Admins can assign per Admin Role:
Survey creation – build and structure surveys.
Survey launch – set participants, launch surveys, and track submissions.
These can be granted independently or together, so one group of admins can build surveys while a different (or overlapping) group is responsible for setting participants and launching them.
Creating a New Survey
Click "Create New Survey" at the top of the page.
Build a survey using AI (enter a prompt and receive a complete, ready-to-use survey tailored to your needs), choose a blank survey or duplicate an existing one (useful for onboarding check-ins.)
Importing Survey Results
If you ran a survey outside Shapes, you can import its completed results from an Excel (.xlsx) or CSV file. The file needs one row per respondent and one column per question.
Navigate to Surveys and select Import.
Upload your file.
Enter the survey name, start date and end date.
The survey is created as Running or Ended, depending on the end date, with its responses already submitted.
Good to know: If your file has an email column, the imported survey is identifiable. If it has no email column, it is anonymous. For file layout, permissions and how unmatched rows are handled, see Importing Survey Results into Shapes.
Customizing Your Survey
You can now configure:
Name and description
Privacy: Choose between identifiable responses (linked to employees) or an anonymous survey. See Anonymous vs. Identifiable Surveys below for how each option works.
Anonymous vs. Identifiable Surveys
The privacy setting controls whether responses are linked to the employees who submitted them. Choose it carefully before launching — it can't be changed afterwards.
Anonymous (Confidential)
Responses are not linked to any employee — Shapes stores them with an anonymous token, not an employee ID.
Admins cannot identify who submitted what — even Super Admins.
Results are only shown once a minimum number of responses has been submitted (3 by default). This prevents re-identification in small teams.
The anonymous label is shown to employees before they complete the survey, so they know their response is confidential.
Imported surveys with no email column are automatically treated as anonymous.
Identifiable (Visible)
Responses are linked to the specific employee.
Admins can see who submitted what and filter results by individual.
Useful for onboarding check-ins or surveys where follow-up is needed.
Important: The privacy setting is defined at the survey level and applies uniformly — you can't mix anonymous and identifiable responses within the same survey.
Note: The privacy type cannot be changed after the survey has launched.
Minimum Response Threshold
To protect confidentiality, results of an anonymous survey are withheld until a minimum number of employees have responded (3 by default). Until that threshold is reached, results stay hidden — so no one can work out who gave which answer in a small group.
Keep this in mind when running anonymous surveys in small teams: if fewer employees than the threshold respond, results will not be released.
The threshold is checked on each result separately, not once for the whole survey:
Breakdowns: a breakdown by job, team or office with fewer responses than the minimum is hidden and shows "Waiting for 3 responses…". Breakdowns that reach the minimum are shown as usual.
Questions: a question with fewer answers than the minimum, such as an optional question, doesn't show its results. Other questions that reach the minimum are still shown.
A survey with many responses overall can therefore still hide some results. The minimum is 3 by default and can't be set lower.
Confidentiality Rules
Three confidentiality rules apply everywhere survey data can be read in Shapes — survey insights, exports, AI Chat and Vibe views:
Anonymous = no identity: an anonymous survey never reveals who gave which answer.
Below threshold = blocked: when too few people have responded for results to be released, results are not shown.
Draft = no data: a draft survey returns no data at all.
Exporting Anonymous Results
Exporting anonymous survey results follows the same confidentiality rules. Per-employee attributes are not attached to anonymous responses in exports, to prevent re-identification.
Building the Survey
Add your logo, title and description
Use the question bank or create your own:
Long text
Rating scales
Single or multiple choice
Add content blocks to organize by topic
Mark questions as required, if needed
To remove a question, just click and delete it
Assigning Participants
Choose who should receive the survey:
All employees
Specific offices (e.g., London)
Selected teams (e.g., Marketing, or everyone except Marketing)
Individual employees
A minimum tenure requirement can also be set when launching the survey — any employee below the threshold is excluded from that survey.
Launching the Survey
Click Preview to review the form from the employee's point of view.
Once ready, click Create.
Set the start and end dates by clicking on Launch Survey.
Employees will receive an email and find the survey in their assigned tasks.
Start time of a fixed-date survey: A fixed-date survey starts on its start date, between 00:00 and 01:00 UTC. Participants receive the email and the task in Assigned Tasks at the same time. The time is not set by the account's time zone, so participants far from UTC may receive it at a different local hour. If you launch a survey on or after its start date, it starts right away.
Launch Type: Fixed Dates or Dynamic Anchor Date
When launching a survey, choose between two ways of setting participant timelines:
Fixed dates – the same start and end date apply to every participant.
Dynamic anchor date – each participant's survey launches automatically based on their own lifecycle date (e.g. Start Date or Date of Birth), instead of one fixed date for everyone.
For a dynamic anchor date survey, set:
Launch trigger – how long after (or before) the anchor date the survey should launch for each participant, e.g. "1 day after Date of Birth."
Completion time – how long each participant has to complete the survey once it launches, e.g. "1 week."
Example: with a launch trigger of 1 day after Date of Birth and a completion time of 1 week, each participant's survey starts the day after their birthday and stays open for a week.
Recurring Surveys
A fixed-date survey can be set to repeat automatically on a schedule instead of being recreated manually each time. Toggle on Repeat this survey and set how often it repeats (e.g. every quarter).
Choose how the recurrence ends:
After a set number of occurrences
On a specific date
Ongoing, with no end date
Each occurrence appears as its own entry in the Surveys table, labeled with its occurrence number, so you can track the survey's history at a glance.
Good to know: Recurrence is only available for fixed-date surveys.
Extending a Survey's Timeline
When a survey needs more time to collect answers, move its end date yourself with Extend Timeline. It works for a survey that's still running and for one that has already ended, which you can re-open with a new deadline. It's the same option performance review cycles have.
Who can use it
Super Admins.
Admins whose admin role has the Surveys permission People & responses management. Admins who only have Survey operation management don't see Extend Timeline.
It's available for surveys with fixed dates. Surveys with a dynamic anchor date don't have one end date to move, so the option isn't shown for them.
The survey must be Running or Ended. A draft or scheduled survey hasn't started yet, so you change its dates with Edit Survey instead.
How to extend or re-open a survey
Go to Surveys and open the survey, or find it in the surveys table.
Open the survey's actions menu (...) and select Extend Timeline.
In the Extend timeline window, the Start date is shown for reference and can't be changed. Pick a new End date.
Select Save.
You'll see Survey timeline updated. and the survey shows its new end date. A running survey stays Running with the later deadline, and an ended survey goes back to Running until the new end date. Participants can respond until the new end date.
Good to know:
The new end date must be later than today and later than the current end date. Save stays unavailable until you choose a later date. An end date can't be brought forward.
Only the end date changes. The start date, questions, participants and settings stay as they are, and submitted responses aren't touched.
A running or ended survey can't be edited. Edit Survey is locked once a survey starts ("Survey is locked. This survey has started and cannot be edited now."), so Extend Timeline is the only way to change its timeline.
Export as CSV is available only once a survey has ended. If you re-open an ended survey, the export becomes available again when it reaches its new end date.
Participants aren't sent a new message about the later end date. Let them know yourself if you'd like to.
Each occurrence of a recurring survey is its own survey and can be extended like any other fixed-date survey. Shapes doesn't stop the new end date from running past the start of the next occurrence, so check the schedule before you pick a date.
Completing the Survey
Participants can:
Click into their Assigned Tasks
Fill in the survey
Submit their responses, which will then be recorded
Managing Responses & Insights
Super Admins can track who has submitted vs. who hasn't.
You can access insights that provide an AI-generated summary of responses.
Insights also show breakdowns in terms of:
Submissions
Average rating by job, team and office
Free text answers
If the survey is anonymous, results are masked and follow the confidentiality rules described above.
Reading Results with AI Chat and Vibe Views
Survey results can also be read without opening the survey itself. Ask the AI assistant to find a survey, see who was asked and who responded, and read the answers. Vibe views can read survey data as well, so a results dashboard with a per-question breakdown can be built the same way as any other Vibe view.
The same confidentiality rules apply here. When too few people have responded for results to be released, the view says so and explains why, instead of showing an empty chart.
Reading results this way is available to anyone with access to surveys who can also use Vibe views.