How to Set Up a Survey in LimeSurvey: Easy Guide (2026)

Setting up a survey in LimeSurvey is a short job once you know the order: log in to the admin panel, create a survey, add a question group, put your questions inside it, write the welcome and end messages, set your dates, preview, then activate and share the link. A clean first survey takes most people about twenty minutes plus the time needed to write the questions.

The one snag every beginner hits is the same one. In LimeSurvey, every question must live inside a question group, and if no group exists yet the question toolbar sits there useless. Create the group first, then build.

Table of Contents

1. 1Quick summary: the 7 steps
2. 2What You Need
3. 3Step-by-Step: How to Set Up a Survey in LimeSurvey
4. 41. Create a New LimeSurvey Survey
5. 52. Add Questions and Choose Question Types
6. 63. Set Required Questions, Validation, and Randomization
7. 74. Add Skip Logic and Conditional Pages
8. 85. Configure Survey Settings and Privacy Options
9. 96. Preview, Test, and Check the Survey on Mobile
10. 107. Activate, Publish, and Collect Responses
11. 118. Export and Analyze Responses
12. 12Common Mistakes
13. 13Frequently Asked Questions
14. 14Is LimeSurvey free?
15. 15Do I need a question group before I can add questions in LimeSurvey?
16. 16How do I make a LimeSurvey survey open to everyone?
17. 17Why can’t I reorder questions in LimeSurvey?
18. 18How do I export my LimeSurvey responses?
19. 19What is the difference between an active and an inactive LimeSurvey survey?
20. 20Conclusion

Quick summary: the 7 steps

1. Log in to the LimeSurvey admin panel with an account that has survey creation rights.
2. Create a new survey: title, base language, page format, theme.
3. Add a question group, then add questions inside it and pick the right question type for each.
4. Mark questions as mandatory, add validation, and turn on answer randomization where it helps.
5. Set privacy, access and publication options, including start and expiration dates.
6. Preview the survey, submit a test response and check the logic on a phone.
7. Activate the survey, copy the survey link or embed code, distribute it and monitor responses.

When the survey closes, export the responses to Excel, CSV or SPSS and clean the variable names before analysis.

What You Need

You need access to a LimeSurvey instance. That is either a free LimeSurvey Cloud account or your institution’s or company’s own installation, where an administrator has given you a username with survey creation rights. Many universities run their own instance, and the login page is usually just the site address plus /index.php/admin/authentication/sa/login.

You also need the content, not just the tool. Write down what you are trying to measure and the specific questions that answer it, each with the answer options you want to see. A survey written from scratch inside the admin panel is painful to fix afterwards.

Optional extras, depending on your study:

• A participant list with email addresses if you plan an invitation-only survey using tokens.
• Consent text and your ethics or IRB reference if the survey counts as human-subjects research.
• A brand logo and colour scheme if the survey goes out to external participants.
• Extra languages if your respondents do not all read your base language.

One more thing worth deciding early: whether responses should be anonymous or linked to a participant. That choice drives the access settings in step 5, and it is much harder to switch after data collection starts.

Step-by-Step: How to Set Up a Survey in LimeSurvey

The workflow below follows the order the admin interface expects. Work through it and the survey URL at the end will accept real responses.

1. Create a New LimeSurvey Survey

In the admin sidebar, go to Surveys and click Add survey. The form asks for four things:

• Title: shown to respondents on the welcome screen, so write it for them rather than for your filing system.
• Language: the base language for the survey. Additional languages can be added later.
• Format: group by question group, group by question, or put everything on one page. For anything longer than about fifteen questions, use question groups.
• Theme: the template that controls appearance. You can switch themes without touching the questions.

Click Create and LimeSurvey drops you into the survey with the Structure tab open. You can also start from an existing survey by importing a .lss file with Tools → Import/Export, or from one of the templates if one has been added to your instance.

Set the survey status now and you will know where you stand: new surveys are Inactive, which means the link exists but returns nothing to respondents.

2. Add Questions and Choose Question Types

First, create the container. Click Add Group at the end of the structure toolbar, give the group a title (it appears as a page heading to respondents) and an optional description. Then click a question type in the toolbar above the structure area.

If your screen is narrow, the toolbar collapses into a dropdown showing all available question types. Clicking a type opens a form with the question text, the question code, and options such as Mandatory and the question layout.

This table covers the question types you will use in almost any study:

Question type	Best for	Example
Single choice	One answer from a short list, anything you will code numerically	How often do you use the library?
Multiple choice	Several answers from the same list	Which services have you used? (tick all that apply)
Array / Matrix	One statement rated on the same scale several times	Rate each statement from 1 strongly disagree to 5 strongly agree
Ranking	Forcing an order between options	Rank these five teaching methods
Matrix	One scale applied across several rows or subquestions	How satisfied are you with each department?
Free text	Anything open-ended or numerical	How old are you? What did you like least?
Date	A specific date rather than a year	When did you first join?
Expression	Calculated or hidden values used in scoring and randomisation	A random number between 1 and 6

Below each question there is a Select control where you choose the answer structure: answer options, subquestions and columns, or label sets for the array and ranking types. Answer options are the value-plus-text pairs you type into the table under the question.

Keep question codes stable. LimeSurvey generates them automatically (Q1, Q2, and so on) and those codes become your column names in the data export, so changing or regenerating codes after collecting responses breaks the link between your analysis script and the data.

3. Set Required Questions, Validation, and Randomization

On the question form, the Mandatory star controls whether respondents can continue without answering. Use it sparingly; a survey with fifteen mandatory questions loses people halfway through. Mark the consent item, the demographic items your analysis needs, and leave opinion items optional.

Validation sits on the same form under Validation. Options include regular expressions, numeric range, minimum and maximum characters, and custom expressions. A numeric-range rule on an age question (set 18 to 90) prevents stray values like 250 from reaching your dataset.

Randomization is set per question and per group:

• Randomize answers shuffles answer options to reduce order bias. Turn it on for long lists and attitude scales, and leave it off when you need a stable reference order.
• Randomize questions (set on a question group) reorders the questions inside that group.

Randomising a whole survey version is different: the common workaround, documented on the LimeSurvey forums, is to place a hidden expression question that generates a random number in a first group and then use conditions to show only the group matching that number.

4. Add Skip Logic and Conditional Pages

Skip logic keeps respondents out of questions that do not apply to them. Open the Conditions tab from the toolbar, add a condition and build it with the expression editor rather than typing syntax from memory.

The concrete case everyone needs: a single-choice question asks Are you currently a student? Yes or No. The employment and course questions live in a later group, and a condition set that group to display only when the student answer is No. Respondents who answer Yes skip the entire group instead of seeing irrelevant questions.

Two things catch people out here. Conditions are evaluated on the answer value, not the answer text, so check the values column in the editor rather than assuming the label. And once questions carry conditions, drag-and-drop reordering in the structure view can be blocked, which surprises people who think the interface has frozen.

For multi-part paths, set conditions on whole question groups rather than on individual questions. One condition per group is far easier to audit before launch.

5. Configure Survey Settings and Privacy Options

Open the Settings tab and work through the settings pages. The ones that matter before launch:

• Publication and access settings: start date and time, expiration date and time, listed publicly, and whether the survey is open or invitation-only.
• General settings: whether to store the respondent IP address and user agent, whether to enable cookies, and how anonymity is handled.
• Security settings: CAPTCHA options for spam protection.
• Presentation settings: progress bar, question index, one-question-per-page display, and the end URL redirect.
• Text elements: survey description, welcome message, end message and end URL, reachable from the toolbar on the Structure tab or from the settings list.

The anonymous-versus-token decision lives here, and getting it wrong is the most common reason a public link does not behave as expected:

Setting	Open link, no participants table	Survey participants table (tokens)
Who responds	Anyone with the link	Only people you sent a token to
Access setting	Open, with listed publicly switched on	Invite by email or link with a unique token
Identification	Anonymous unless you store IP or user agent	Each response tied to a named participant record
Use it for	Open feedback, general population samples	Panels, follow-ups, reminder emails

For an open survey, create no survey participants table at all and just share the survey URL. If you do create one and leave access closed, the link will refuse respondents, and that dead end is what the official help portal article about open surveys keeps having to correct.

Set the expiration date if the survey is genuinely time-limited. A survey left active with no expiry keeps accepting responses until you deactivate it yourself, which is fine for open feedback and wrong for a data collection window with a deadline.

6. Preview, Test, and Check the Survey on Mobile

Use Preview survey (the eye icon, also at the bottom of the structure page) rather than activating the real survey. Preview lets you walk the whole thing and lets you check conditions by answering the questions the way a real respondent would.

Do one complete test response and check four things:

1. Every question shows the text and options you intended, with no missing answers.
2. Mandatory questions block progress until answered.
3. Skip logic takes respondents the intended route in both directions of every condition.
4. The welcome, end messages and thank-you page appear where you expect them.

Then open the survey link on a phone. Most respondents on a mobile panel will never see your desktop view, and wide matrix questions with four scale columns are the ones that break first. Split them into arrays or switch to one question per page.

After testing, delete the test response from the data entry screen so it does not end up in your dataset.

7. Activate, Publish, and Collect Responses

When you are happy with it, go to Publication and access settings and click Activate survey. The status in the survey overview changes from Inactive to Active, and the start date must already have passed. If the date is in the future, the survey stays in a pending state and the link will not work yet.

Copy the survey URL from the overview or from the sharing panel, where you can also generate a shortened link, an embed snippet for a website or paper, and per-language links for multilingual surveys. Distribute it through whatever channels you approved: email, a course page, social media, or a QR code on a poster.

Watch participation from Responses and statistics on the survey sidebar. That screen shows submission counts, response times and partial responses, which is where you spot a broken question that is causing people to abandon the survey.

Close the survey when collection ends, either by setting an expiration date in advance or by deactivating it manually. Remember that expired and inactive surveys stop accepting responses but keep the data.

8. Export and Analyze Responses

From Responses and statistics, open the results or survey results view and choose Export results. The export dialog offers several formats, and three of them matter for analysis:

• Excel or CSV for quick inspection and for handing data to a colleague who does not use statistics software.
• SPSS (.sav) for direct import into SPSS with variable labels, value labels and missing values already mapped.
• Tab or semicolon delimited text when you need to fix column names in a spreadsheet before loading.

Expect two annoyances in any export. Question codes become column names rather than question text, so keep a code-to-question lookup table alongside your analysis. And respondents who skipped a question leave an empty cell or a coded missing value rather than a zero, which changes how you handle the variable in SPSS or R.

Export a copy as soon as the survey closes, while you still remember the layout of the data file.

Common Mistakes

Nearly every beginner problem traces back to one of five causes:

Problem	Cause	Fix
No questions can be added	No question group exists yet	Use Add Group on the structure toolbar, name it, then add questions
The survey link returns an error	Survey is inactive, access is closed, or the start date is in the future	Check the survey status and the start date under publication and access settings
The public link asks for a token	A survey participants table exists and access is invitation-only	Delete the participants table and set access to open with listed publicly on
Questions cannot be reordered	Conditions are applied to the affected questions or groups	Open the conditions tab, note what depends on the current order, remove the conditions, reorder, then rebuild them
Exported columns have meaningless names	Default codes such as Q12X1P1 were used	Edit the question code before launch, or keep a code-to-text lookup for analysis

A few habits keep a survey in good shape. Name question groups after the topic they cover, not as page one and page two. Test on a phone before you send anything. Add a consent question as the first item if this is human-subjects research, because institutional guidance typically expects the consent text before any other content. And write every question in neutral wording, since leading phrasing is what makes a survey unusable rather than a LimeSurvey setting.

Frequently Asked Questions

Conclusion

Setting up a survey in LimeSurvey is a fixed sequence: create the survey, add a question group, build the questions inside it, write the text elements, set the access and privacy options, preview, activate, and share the URL. Every step after that, including analysis, depends on the data you protected by doing the order properly.

Start with your research objective and a written list of questions and answer options, since nothing in the admin panel works without content to put into it. Then build, test one full response on a phone, and only activate once the logic takes the paths you intended.