How to Analyze Jotform Survey Data: Export, Analyze and Act

Jotform is used for almost anything that fits into a form: event registrations, order forms, job applications, patient intake and surveys. That flexibility helps when collecting data and adds work when analyzing it, because a Jotform export often holds payment details, contact fields and file uploads in the same table as the ratings and comments you want to study.

This guide follows one example throughout: a conference that collects registrations and then sends attendees a feedback survey. It covers setting up the form, exporting and preparing the file, choosing statistical tests and analyzing open-ended answers, then shows where Kimola fits when answers and comparisons grow.

What Jotform Is and How It Stores Submissions

Jotform is an online form builder. Forms are created in the Form Builder, and submissions are stored in Jotform Tables, where each form has a table with one row per submission and one column per field. Tables is also where you filter, edit and download submissions.

Two other products matter for analysis. Report Builder turns submissions into visual reports with charts and tables that can be shared, embedded or saved as PDF. The AI Data Assistant, which works in Tables and in Jotform Inbox, answers questions about submission data in plain language and can summarize, categorize and label open-ended answers; it is covered in the section on open-ended answers.

Setting Up a Survey That Produces Clean Data

Start From a Decision

A survey is easiest to analyze when it is built around a decision. "Get feedback on the conference" leads to a long form and a vague report. "Decide whether to keep the online ticket next year and which sessions to change" tells you what to ask, whom to compare and what the analysis must answer. Every question should be traceable to that decision. Ask about one thing per question, keep the form as short as the decision allows, and say at the top why you are asking and how the answers will be used.

Name Fields and Keep Answer Options Consistent

The label of each field becomes its column name in Tables and in the export, so make each label distinct and recognizable on its own; three columns that all begin with "Please tell us" are easy to misread.

Answer options are what you will group and compare, so keep them stable. Use one rating scale throughout the form, for example 1 to 5 where 5 is best, and fixed choices such as dropdowns for anything you want to count, like ticket type or attendance mode. If the survey runs again next year, keep the options identical; changing "Online" to "Virtual" between editions creates two groups where there should be one.

Use Conditional Logic

Jotform's Conditional Logic, set up in the Form Builder under Settings and then Conditions, can show or hide fields, skip or hide pages, calculate values, and change the thank-you page or the email recipient. In a feedback survey, its most useful job is targeting follow-up questions: an attendee who joined online can be asked about the stream, and someone who rated a session 1 or 2 can be asked what went wrong. Answers become more specific, and the form stays short for everyone else.

Test the Form Before Launch

Submit the form several times along different paths through the conditions, and ask two or three colleagues to do the same. Note when testing ended, because those entries will need to be removed before analysis.

Exporting Submissions From Jotform

Downloading CSV or Excel From Tables

Open the form's submissions in Jotform Tables and click Download All in the upper-right corner. Choose Download as CSV or Download as Excel, and the file downloads once it is ready. The third option, Default Document, creates a PDF of each submission and emails them as a zip file; it is useful for records, not for analysis.

To export only part of the data, tick the checkboxes next to the entries you want or apply a filter first, such as a date range, and the download contains only those entries. Jotform's help article on downloading submissions describes each option.

Sending Submissions to Google Sheets

For a survey that keeps collecting, the Google Sheets integration saves repeated downloads. Set it up in the Form Builder under Settings, Integrations and then Google Sheets. New submissions are added automatically, and existing ones can be sent at setup. Edits made in the sheet do not flow back to Jotform, so treat the sheet as a working copy.

Cleaning and Preparing the File

Preparation is where Jotform exports differ most from the exports of survey-only tools. The goal is a file with one row per respondent, the columns the analysis needs and nothing that should not be shared.

Remove Test Entries and Duplicates

Delete the submissions made during testing. Then look for duplicates, such as two rows with the same email address a few minutes apart, and apply one rule consistently, for example keeping the most recent submission per person. Check suspected duplicates before deleting them, since two colleagues from the same company can produce very similar rows.

Separate Operational Fields From Feedback

Payment fields, billing addresses, signatures, file uploads and invoice numbers are needed to run the event, not to understand it. Leave them out of the analysis file, or work from a copy that has only the columns you need. Some registration fields are worth keeping because they describe the respondent, such as ticket type, attendance mode or session track. These become the groups you compare.

Remove Personal Data

Names, email addresses and phone numbers are rarely needed for analysis. Remove them or replace them with an internal ID, so the file can be shared without exposing respondents; data protection rules such as the GDPR expect personal data to be limited to what a purpose requires. Jotform's security page describes how Jotform itself stores data, including GDPR and CCPA compliance and HIPAA features, but the exported file is the organization's own responsibility.

Keep the Date and Segment Columns

Keep the submission date, every rating and multiple-choice answer, and every column that describes the respondent. The date shows change over time, and the segment columns show whether online attendees, students or one session track see the event differently.

Analyzing Quantitative Answers

Ratings and multiple-choice answers are the quantitative part of the survey. Analyze them first, because they show where to look when you read the comments.

Start With Descriptive Statistics

For each rating question, report the number of responses, the distribution across the scale and the median. A 1 to 5 rating is an ordinal scale: the answers are ordered, but the step from 2 to 3 is not necessarily the same size as the step from 4 to 5. The share who answered 4 or 5 and the share who answered 1 or 2 are usually easier to interpret than an average of 3.8. Then break the numbers down by segment; a pivot table with attendance mode in the rows and the rating distribution in the columns shows whether online and in-person attendees rated the event differently.

Choosing a Statistical Test

A statistical test estimates whether a difference in a table is larger than chance would produce. The right test depends on the type of data and the number of groups.

A chi-square test compares two categorical variables, such as ticket type and the answer to "Would you attend again?"; with small groups, Fisher's exact test is the usual alternative. A t-test compares a mean between two groups, such as the ratings of online and in-person attendees, but because a single rating question is ordinal, the Mann-Whitney U test, which compares rankings, is often the better choice. ANOVA compares three or more groups, such as four session tracks, and its rank-based alternative is the Kruskal-Wallis test; a significant result says at least one group differs, not which, so post-hoc comparisons follow.

Two cautions apply to all of them. With 12 students in the sample, two answers move their share by more than 15 points, so report the number of responses next to every percentage. And running many tests at once produces some significant results by chance, so an isolated difference among dozens of comparisons is a question to check, not a finding.

Analyzing Open-Ended Answers

The open-ended question, here "What should we change for next year?", is where attendees explain their ratings. Reading those answers systematically is a qualitative research task.

Thematic Analysis by Hand

Thematic analysis turns free text into themes that can be counted and compared. Read a random sample of 50 to 100 answers and note what attendees raise, such as the venue's Wi-Fi, session lengths, the online stream or networking time. Turn the notes into a codebook in which each code has a name, a one-sentence definition and an example.

Then code every answer, with a column per code and a note on whether each mention is positive or negative. "The talks were excellent but the breaks were too short to meet anyone" covers two topics with opposite sentiment, and each is coded separately. Count the codes overall and by segment, and keep a few quotes for each important theme. A second person coding a shared sample shows which definitions are unclear.

Using Jotform's AI Data Assistant

Jotform's AI Data Assistant opens beside a table in Jotform Tables. You ask questions in plain language, and it returns summaries, breakdowns and charts and can point out recurring themes and outliers. Jotform says it is available to all users.

For open-ended answers, the most relevant part is AI Columns, added with the Add button at the top of the last column. Each creates a new column from an existing one: Summarize condenses each answer, Categorize groups similar answers, Sentiment Analysis labels each answer as positive, neutral or negative, and Translate converts answers into a chosen language. A custom column follows your own prompt, and new submissions can be processed automatically.

This is a convenient first structured view of one form's answers without leaving Jotform. It works at the level of the whole answer and one table, and the documentation does not describe giving one answer several themes with separate sentiments, testing whether differences between segments are significant, or combining several forms in one analysis.

Analyzing Jotform Data in Kimola

Manual coding works for a few hundred answers. With thousands, the codebook drifts over days of work, and comparisons multiply: ten themes across four ticket types and two editions make eighty, too many to test one by one. A chat assistant can summarize a few dozen comments, but consistent classification, significance testing and a traceable source for every figure need a tool built for that work.

Kimola is an AI platform that turns customer feedback into enterprise-grade market research. It analyzes survey responses, reviews, support conversations and social discussions the way a research team would. It does not build forms or collect submissions; it starts from the file you export from Jotform.

Uploading the Export

From Home in Kimola, upload the cleaned Excel or CSV file. Choose the column that holds the open-ended answers and, optionally, a Date column such as the submission date. Every other column is kept, including ratings, ticket type, attendance mode and session track. Kimola reads each column's name and values to understand how it relates to the feedback, and the columns become filters and comparison groups in the Report, which is why preparation keeps them.

A Report analyzes one text column, so a survey with several open-ended questions needs a Report for each, starting with the one closest to the decision. Before creating the Report, you set its output language and see how many queries it will use, as Kimola's guide to uploading a dataset explains.

Themes, Sentiment and Dimensions

Kimola finds themes in the data itself without a predefined list and names them after what respondents say, so the conference survey would show themes such as break length or stream audio rather than a broad label like "Logistics". With multi-label classification, an answer that praises the talks and criticizes the breaks gets both themes, each with its own sentiment, which is called aspect-based sentiment. Answers in 30+ languages can be analyzed together, which helps with international events.

Dimensions ask a research question of every answer. Pain Points show what went wrong, Motivations why people attended or would return, and Unmet Needs what they expected and did not get. Experience Stages can separate comments on registration from comments on the event itself. Dimensions can be added when the Report is created or later, use additional queries and do not change the original dataset.

Signals Across Your Own Columns

Signals cross every theme and Dimension with the columns in your file and with time, and keep only statistically significant results. A Difference might show that complaints about audio are concentrated among online attendees, or that one session track is mentioned more negatively than the others. A Trend shows a theme becoming more or less common over time, or across editions if the file combines them. Signals are scored with deterministic methods, so the same data gives the same result every time. They need a large enough dataset and selected Dimensions; a Report without Signals means nothing stood out, not that the analysis failed.

Action Plans and Sharing

On the Actions screen, Kimola can write an Action Plan for a Profile that describes your organization, such as the event organizer. It forms hypotheses from the Report, checks them against the feedback, drops recommendations the evidence does not support, and shows the findings, Signals and quotes behind each action with a confidence level. Reports can be exported to PowerPoint, PDF and Excel.

For a survey that runs after every event in a series, Kimola's Enterprise package offers Spaces, where each new file is classified with the same models and Dimensions and Signals run again after every upload. Organizations whose policies require feedback to stay in-house, such as healthcare providers, can use Kimola On-Premises.

What Kimola Does Not Do

Kimola does not create forms, collect submissions or clean the export; Jotform collects the data, and the preparation above is still worth doing. If a survey receives a few dozen answers, reading them directly is often enough. Kimola is most useful with hundreds or thousands of answers, several segments or several survey waves to compare. Kimola has a free Starter plan with no credit card required, and paid plans start at $49 per month, as listed on the pricing page.

Frequently Asked Questions

Can Jotform analyze survey responses by itself?

Partly. The AI Data Assistant in Jotform Tables can summarize answers, build breakdowns and, through AI Columns, categorize answers and label their sentiment. For several themes per answer, significance tests across segments or analysis across several forms, export the data to a spreadsheet, statistical software or a dedicated analysis tool.

How do I export Jotform submissions to Excel?

Open the form's submissions in Jotform Tables, click Download All in the upper-right corner and choose Download as Excel. To export only some entries, select them or apply a filter before downloading. Download as CSV works the same way.

Which statistical test should I use for survey ratings?

It depends on the data and the number of groups. A chi-square test compares categorical answers between groups, a t-test or the Mann-Whitney U test compares a rating between two groups, and ANOVA or the Kruskal-Wallis test covers three or more. Because single rating questions are ordinal, the rank-based tests are often the safer choice, and small groups should always be read with caution.

Can I analyze several Jotform forms together?

Yes, if they ask the same question. Download each form, align the columns, add a column that identifies the form or edition, and keep the answers to the shared question in one text column. In Kimola, that column becomes a comparison group, so Signals can show which themes differ between forms.

Conclusion

Jotform data rewards preparation. A form built around a decision, consistent answer options and an export cleaned of test entries, operational fields and personal data, but with its dates and segment columns intact, make every later step easier, from testing ratings to coding comments.

When the answers run into thousands, or when the question is which differences between groups and periods are real, the same file can go into Kimola. An organization that already collects feedback through Jotform can upload it as it is and turn it into enterprise-grade market research, without designing a new study. You can create a free account and start with a form you already have.