How to segment users based on survey responses in Refiner

You’ve got survey data coming in through Refiner, but now what? If you’re like most SaaS or product teams, you want to actually do something with those responses—whether that’s sending smarter emails, prioritizing features, or just understanding who’s happy (and who’s not). This guide walks you through how to segment users in Refiner based on survey answers, so you can stop staring at spreadsheets and actually get value from all those pop-ups and emails.

If you’re new, don’t worry—you don’t need to be a data scientist. If you’ve been using Refiner for a while, we’ll skip the fluff and get straight to what matters. Let’s get into it.

Why segment users by survey response?

You might be tempted to blast everyone the same follow-up, but that’s a quick way to annoy people and miss out on insight. Segmenting users by how they answer surveys lets you:

  • Send relevant messages to the right people (not everyone cares about your latest beta).
  • Spot trends, like which features delight your power users—or drive others nuts.
  • Prioritize feedback from your best customers, not just the loudest voices.

But here’s the honest bit: Over-segmentation is real. Don’t make 17 segments for every possible answer. Start simple, see what’s useful, then refine.

Step 1: Make sure your survey data is clean (and useful)

Before you start slicing and dicing, check your survey setup:

  • Are your questions clear and actionable? If you’re asking vague stuff (“How do you feel about our product?”), you’ll get vague answers.
  • Is your data coming in reliably? If users aren’t identified properly in Refiner, your segments won’t mean much.
  • Do you have enough responses? Don’t bother segmenting if only three people answered your survey.

Pro tip: Standardize your questions where possible—using dropdowns or scored answers is way easier to segment than free-form text.

Step 2: Find your survey responses in Refiner

Log in to Refiner and head to the “Survey Responses” or “Audience” section—names can change, but you’re looking for the area with individual user profiles and their survey data.

  • You should see a list of users, each with survey responses attached.
  • If you’re not seeing recent answers, double-check your survey’s targeting and triggering settings.

Got multiple surveys? Make sure you know which survey you want to segment by. Mixing data from different surveys can get messy fast.

Step 3: Build a new segment based on survey answers

Here’s where the magic happens. Refiner lets you create segments (think: groups of users) based on just about any data it’s got—including survey responses.

  1. Go to the Segments section.
  2. This is usually on the left sidebar, sometimes under “Audience” or “Segments.”
  3. Click “Create Segment” (or similar).
  4. Give your segment a clear, no-nonsense name. “Promoters—Q2 NPS” beats “Segment 5.”
  5. Set up your conditions:
  6. Choose the survey property you care about (e.g., “NPS Score” or “Feature Request: Dark Mode”).
  7. Pick how you want to filter. For example:
    • NPS is greater than or equal to 9 (to find promoters)
    • Feature Interest equals 'Yes'
    • Churn Reason contains 'Pricing'
  8. Stack conditions if needed.
  9. You can combine: e.g., “NPS < 7” AND “Plan is Pro.”
  10. But don’t get carried away—if you add too many filters, your segment might end up empty.
  11. Save the segment.
  12. Most tools let you preview how many users match before saving. If your number is weirdly high or low, double-check your filters.

What works: Stick to one or two key questions per segment—like “Would recommend” or “Interested in beta.” That keeps things manageable.

What doesn’t: Don’t try to build micro-segments for every possible combo (“Left negative feedback on Monday and uses iOS and is on trial”). You’ll end up with noise.

Step 4: Put your segments to work

A segment does nothing on its own. Here’s how to actually use them:

  • Trigger messages or emails: Most teams use segments to send follow-up emails (“Thanks for your feedback!” or “Want to try our new feature?”).
  • Sync to other tools: Refiner can usually push segments to CRMs, email tools, or Slack. Use this to keep your sales/support teams in the loop.
  • Analyze trends: Look at what percentage of each segment is converting, upgrading, or churning. This helps you spot patterns, not just anecdotes.

Pro tip: If you’re syncing segments to other tools, double-check the mapping. You don’t want to send a “We’re sorry you’re unhappy!” email to your biggest fans.

Step 5: Iterate (and don’t overcomplicate it)

Segmentation isn’t one-and-done. Here’s what to watch for:

  • Check segment sizes over time. If a segment gets too small, it’s probably not worth maintaining. If one’s huge, maybe your filters are too broad.
  • Tweak questions and filters as needed. If you realize nobody is picking a certain answer, consider rewording or removing it.
  • Review actions tied to segments. Are your follow-ups actually helping? Or just annoying folks? Look at open rates, replies, and unsubscribes.

What to ignore: Fancy scoring models or “AI-powered” segments sound cool, but if you can’t explain them in one sentence, they’re probably not worth the hassle—at least until you’ve nailed the basics.

Tips for better segmentation (from people who’ve messed this up)

  • Start with big, obvious buckets. Happy vs. unhappy, interested vs. not interested, active vs. inactive.
  • Don’t reinvent the wheel. Most teams find NPS, feature interest, and churn reasons are the most actionable segments.
  • Keep segment names human-readable. You’re not impressing anyone with “seg_2024q2_promoter_beta_v2.”
  • Document what each segment means. Future you (or your teammates) will thank you.
  • Review segments quarterly. Old segments tend to pile up and get ignored.

Common pitfalls (and how to avoid them)

  • Too many segments: If you can’t remember what half your segments are for, you’ve gone too far.
  • Data decay: If surveys change, segments based on old questions will break. Set reminders to review.
  • Assuming segments are static: People change—someone who was unhappy last month might love you after a new feature drops.

Wrapping up: Keep it simple, keep it useful

Segmentation in Refiner isn’t rocket science, but it’s easy to overthink it. Start with the questions that matter, build a few clear segments, and see what actually drives action. If a segment doesn’t help you make a decision or trigger something useful, it’s just clutter.

Iterate, prune, and—most importantly—use your segments. The goal isn’t fancier dashboards; it’s better conversations with your users.