How to Use Forms for SaaS Product Research and Feature Prioritization

To prioritize features with surveys, ask users about their problems before you ask about features: measure product-market fit with the "very disappointed" question, collect the jobs people hire your product for, then have users rate how much specific features would matter to them. Combine those ratings with your own effort estimates in a framework like RICE, and use logic so each user only answers about the parts of the product they actually use.
The biggest risk for a product team isn't failing to build a feature; it's building the wrong one because the loudest voices drowned out the representative ones. A steady survey program gives you the representative view.
1. The product-market fit survey
The core question, popularized by Sean Ellis, is: "How would you feel if you could no longer use [Product]?" with the answers Very disappointed, Somewhat disappointed and Not disappointed. Ellis found that companies where at least 40% of users answer "very disappointed" tend to have product-market fit, as described in First Round Review's account of how Superhuman used the survey.
Send it to users who have had time to experience the core value (active users, not people who signed up yesterday), and add:
- "What is the main benefit you get from [Product]?" Tells you what to protect and what to say in marketing.
- "What type of person do you think would benefit most from [Product]?" Shows how users describe your ideal customer.
- "How can we improve [Product] for you?" The raw material for your roadmap.
2. Feature prioritization with RICE
RICE, introduced by Intercom, scores each idea on Reach, Impact, Confidence and Effort, then divides the first three by effort. Surveys can feed the first two:
- Reach: how many users a feature would affect. A survey question like "How often do you [do the task this feature helps with]?" (Never, Monthly, Weekly, Daily) estimates it.
- Impact: how much it would matter to them. Ask users to rate each candidate feature on a 1 to 5 scale from "Not important" to "Extremely important".
- Confidence: how sure you are. More responses and consistent answers raise it.
- Effort: comes from your team's estimate in person-months, not from users.
Keep the feature list short (five to eight candidates per survey) and randomize the order of options so the first item doesn't get an unfair advantage. You can also ask "Which of these would you use most?" with a multiple choice question that allows several answers, and tell people to pick up to three.
3. Use logic to find silent pain points
A survey that shows everyone the same questions misses the nuance. Branch on what people tell you:
- If a user says they rarely use a feature, ask why: too complex, not relevant, didn't know it existed, or slow.
- If a user identifies as an advanced user, ask about integrations, API needs or bulk actions.
- If a user says they're considering leaving, ask whether the reason is a missing feature, price or usability.
Pass plan or account type as a hidden field in the survey link (?plan=pro) so you can segment results without asking.
4. Jobs-to-be-done questions
Jobs-to-be-done research focuses on the struggle that made someone look for a product. Ask about what happened, not about features:
- "What happened that made you start looking for a tool like ours?"
- "What were you using before, and why wasn't it working?"
- "What almost stopped you from signing up?"
- "When did you realize [Product] was going to work for you?"
These answers are qualitative, so read them all and group them into a handful of recurring jobs. For more on mixing numbers and words, see qualitative vs quantitative data.
5. Close the loop automatically
Product research works best when the whole team sees it:
- Slack: post each response to a
#product-feedbackchannel. - Google Sheets or your tools: send responses to a sheet for analysis, or use webhooks with Zapier, Make or n8n to tag users in your CRM or create items in your issue tracker.
- Say thank you: a short ending screen that tells people what happens next. On FlowyForm Premium, you can also send respondents a confirmation email.
Build it in FlowyForm
Start from the product feedback survey, feature request form, product idea submission form or beta tester signup form. Use opinion scales for impact ratings, multiple choice with randomized options for feature lists, hidden fields for plan or account ID, and logic for follow-up questions. FlowyForm has no ranking question, so use ratings per feature or a "pick your top three" multiple choice instead.
For more question ideas, see our product survey questions, in-app survey guide and beta waitlist guide.
FAQ
What questions should a feature prioritization survey ask?
How often users do the task the feature supports, how important each candidate feature would be to them (1 to 5), which they'd use most, and an open question about what's missing.
How many features should I include in one survey?
Five to eight. Longer lists lead to rushed ratings, and the last items get less attention.
Who should I send a product research survey to?
Active users for product-market fit and feature questions, recent sign-ups for jobs-to-be-done questions, and people who cancelled for churn reasons.