AI Dynamic Questions Enterprise tier only

Surveys now adapt in real time. AI Dynamic Questions (utilizing the Field Agent) automatically generate relevant open text question based on what each respondents prior questions/responses

• Add an AI Dynamic Question anywhere in your study. You can also add a custom (optional) prompt to each question.
• AI reads all prior questions and responses, then generates a short, neutral follow-up using NLP signals like sentiment, intent, and keywords. If it can't generate safely or quickly, the question is skipped.
• Results appear in Reports alongside your standard responses.
• Question styles: Clarification · Depth · Context · Impact · Sentiment · Keyword
• Safeguards: Moderation filters block PII, bias, and unsafe content. Skip logic and quotas are always respected.
• Available on: In-product surveys, long form surveys, and feedback button: Enterprise plans only. No limit on the number of AI Dynamic Questions per study.

What AI uses to generate questions

• Team context (if available)
• Agent context pka Study Goals (if provided)
• All prior questions and responses
• Question prompt (if entered)
• NLP signals: intent, sentiment, topics, and entities

💡 Start with 1–2 AI Dynamic questions per survey to add depth without increasing respondent fatigue.

For those of you who use Sprig to test new ideas and concepts, we have two enhancements to share:

1. You can now include published Figma Make and Lovable prototypes in Long-Form Surveys. Get feedback on these highly dynamic prototypes just like you can for more traditional prototypes by placing them within a broader survey that might include multiple ideas. Learn more.
2. For those who share AI-generated prototypes directly with users, we've validated that you can install the Sprig SDK into the prototypes themselves so that Feedback studies or In-Product Surveys can be part of the experience. Results show in Sprig just like when you launch studies in your live apps. Now you no longer have to rely on manual, in-person, or ad-hoc processes to get feedback on these vibe-coded ideas. Learn more.

Happy testing!

AI Follow-Up automatically generates personalized, open-ended probes based on a respondent’s specific answer, capturing richer detail without manual programming.

Key Features

• Supports Open Text questions; currently unavailable for Voice/Video. More question types to have AI Follow-up questions soon.
• Users must choose between AI Follow-Up or Skip Logic per question.
• Skips automatically if a relevant or safe question cannot be generated quickly.

Why use AI follow-up questions?

• AI follow-up questions help turn vague feedback into specific, actionable insights by probing for clarity, examples, and root causes.
• They mimic a live interviewer at scale, improving response quality without adding researcher workload.

How to Use

• Enable via the toggle in Study Setup for any Open Text question.
• Context used: Uses study goals, previous responses, and NLP signals (sentiment, intent) to tailor probes.
• Generates questions focused on clarification, impact, sentiment, or depth.
• Results: View both the AI-generated probe and the respondent’s answer in the Results tab.

Best Practice

• Use 1–2 AI Follow-Up questions per survey to balance deeper insights with respondent experience.

Display Logic is now available for Long-Form Surveys, In-Product Surveys, and Feedback studies for customers on the Enterprise plan 🎉

Now you can easily configure questions to show when a respondent meets specific criteria based on their prior responses in the survey or user attributes you've stored in Sprig, like their plan or region. This allows you to ask customers a question with options that are relevant to them (Mets, Jets, or Nets for those in New York and Bears, Cubs, and Bulls for those in Chicago) which will lead to higher engagement and richer data.

Check out the documentation to learn more about setting up Display Logic.

Generic questions are out; tailored experiences are in. Starting today, you can use the attribute data you already store in Sprig to customize your survey questions and descriptions in real-time.

Why this is great: more context for respondents means higher completion rates and higher quality responses. Instead of asking, "How much value do you feel like you're getting from your paid plan?" ask, "How much value do you feel like you're getting from your Enterprise plan, Sarah?"

How it works: Just like Response Piping, reference any attribute stored in Sprig directly within the question or description text.

Availability: Attribute Piping is available for customers on the Enterprise plan across In-Product Surveys, Long-Form Surveys, and Feedback studies.

👉 See how to set it up

With this release customers on the Enterprise plan gain an additional configuration option when randomizing the choices offered to respondents in Multiple Choice: Multi-Select and Matrix questions.

Now when setting choices to display in a randomized order in these questions there will be an option to limit the number of choices displayed. For example, if you have a list of 20 themes that you'd like respondents to select from but you don't want them to become overwhelmed with the full list, you can choose to limit the number of choices displayed to only 6 or 8.

See our Help Center documentation to learn more.

Do you ever wish you could export survey response data for a specific timeframe?

Well, now you can! You can now choose a timeframe when requesting a CSV export so you can jump into analysis faster. Choose between preset or custom timeframes. If your survey has more responses than are supported for CSV exports (950,000) you can reduce the number of responses using the new filters and continue with your work. Learn more.

MaxDiff questions are here! 🎉 🥳

Gather detailed insights from respondents as you consider how to prioritize improvements or refine brand messaging with intuitive researcher and respondent experiences. Use the recommended setting to configure the question quickly or define your own experiment structure with the custom setting. Get a sense for how respondents feel about what you're asking them to rate using the built-in best-worst summary or download the CSV to run your own analysis. Check out the documentation to learn more.

We have introduced Conversational Surveys: a new, chat-style survey experience that makes responding feel more natural and engaging.

Instead of displaying multiple questions on a single page, Conversational Surveys present questions one at a time, creating a smooth, dialogue-like flow. Respondents interact as if they’re in a friendly conversation, leading to a more human, immersive, and enjoyable experience.

Check out Conversational surveys to learn more about using this new functionality in your surveys!

We've added new validation options for Multiple Choice Multi-Select questions so that you can capture exactly the data you need from respondents. There are now three settings for validating responses to these questions:

• Unlimited - Respondents can select as many choices from the list as they'd like.
• Maximum - Set an upper limit for how many choices respondents can select.
• Range - Set both lower and upper limits for how many choices respondents can select.

Check out the documentation to learn more about using this new functionality in your surveys!