How to analyze Net Promoter Score results using advanced filters in Delighted

Looking at Net Promoter Score (NPS) results is easy. Actually getting useful insights out of them? Not so much. If you’re using Delighted to run your NPS surveys, you’ve got access to some pretty solid filtering tools—but only if you know how to use them without getting lost in the weeds. This guide is for people who want real answers about what their customers think, not just a feel-good dashboard.

Let’s get into how to use advanced filters in Delighted to break down your NPS results, spot what matters, and ignore the noise.


Why bother with advanced filters?

Before you roll your eyes: yes, the overall NPS score is easy to track. But if you stop there, you’re missing the whole point of collecting feedback. Advanced filters help you:

  • See how different customer segments feel (not just the loudest voices)
  • Catch problems before they become trends
  • Figure out which changes actually move the needle

If you want more than “our NPS is up 2 points this quarter,” you’ll need to go deeper.


Step 1: Get your data in shape

Advanced filters are only as good as the data you feed them. If your customer data is a mess, your insights will be, too.

Checklist before you start:

  • Tags and properties: Make sure you’re passing relevant info with each survey response (like plan type, region, signup date).
  • Consistent formatting: “Enterprise” vs “enterprise” vs “Enterprise Plan” is a recipe for confusion.
  • Enough responses: Filters can’t work magic on tiny datasets—if you have 6 responses in a segment, don’t trust the trends.

Pro tip: If you’re not already sending properties into Delighted (like customer tier or product used), talk to whoever runs your integrations. The more context you have, the more powerful your filters get.


Step 2: Find the advanced filters

Delighted’s interface tries to keep things simple, so “advanced” filters aren’t buried, but they’re not always front-and-center either.

  • Go to “Survey People” or “Dashboard”
  • Look for the filter bar near the top—this is your main tool
  • Click “Add Filter” (sometimes it’s a funnel icon)
  • You’ll see options for properties, tags, dates, response types, and more

What’s worth your time: - Properties (anything you’re passing in, like plan, location, or account manager) - Response type (Promoters, Passives, Detractors) - Date range (to see if things are getting better or worse) - Tags (if you use them for campaigns, feedback themes, etc.)

What you can skip: Filtering by things like browser type or operating system is almost always a waste unless you’re running a tech support operation.


Step 3: Slice your NPS by customer segments

Now for the good stuff. This is where filters actually give you answers.

Common segments worth checking:

  • By plan/tier: Are enterprise customers less happy than basic ones? Or vice versa?
  • By product/service: If you have more than one, see which is pulling your score down.
  • By region: Sometimes issues are location-specific.
  • By lifecycle stage: New vs. long-term customers often have very different experiences.
  • By account manager or team: If you want to see if specific reps are making a difference.

How to do it:

  1. Click “Add Filter”
  2. Choose the property (e.g., “Plan”)
  3. Select the value(s) you want to compare
  4. Note the NPS, response count, and (if you want) export the comments for each segment

What to watch out for: - Small N: Don’t trust dramatic swings if you only have a handful of responses. - Cherry-picking: It’s tempting to hunt for the “worst” group just to look busy. Focus on actionable differences, not just statistical blips.


Step 4: Zero in on the “why” with open-ended feedback

Numbers are fine, but the real gold is in the comments. Delighted lets you filter responses and then view only the open-ended feedback for those segments.

How to use this:

  • Filter by Detractors only, then read what they’re actually complaining about.
  • Compare Promoters’ comments for high-value customers vs. everyone else.
  • Tag recurring themes manually if you see patterns (e.g., “support wait time,” “feature requests”).

Don’t overthink it: Natural language analysis tools are usually overkill unless you’re drowning in responses. For most teams, skimming 50-100 comments and jotting down common themes by hand is faster and more accurate.


Step 5: Track changes over time, not just snapshots

It’s easy to get caught up in the latest numbers, but trends matter more than one-off results.

  • Use date range filters to compare this quarter vs. last, or pre- and post-launch of a big feature.
  • Combine with other filters (e.g., only enterprise customers, only in Europe) to see if specific changes helped or hurt.

Watch out for: - Seasonality: Holidays, big industry events, or even billing cycles can mess with your numbers. Don’t panic over a one-month dip unless it lasts. - Survey fatigue: If you suddenly see a drop in responses, double-check if you’re over-surveying the same group.


Step 6: Export and share—without drowning in spreadsheets

You don’t have to keep all this insight to yourself. Delighted lets you export filtered results as a CSV so you can make charts (if you’re into that sort of thing) or share with your team.

Tips:

  • Only export what you need; massive exports just create work.
  • Add a summary of your key findings (1-2 sentences) when sharing, or nobody will read it.
  • Keep it simple—focus on “what’s changed” or “what needs attention,” not just the NPS number.

Step 7: Beware of common traps

Filtering is powerful, but it’s not magic. Here’s what to skip:

  • Don’t over-segment: If you slice your data into 20 micro-groups, patterns disappear. Stick to segments that actually matter for your business.
  • Don’t ignore context: A dip in NPS might be tied to a recent change, or maybe a competitor ran a big promo. Always ask “why now?”
  • Don’t let dashboards drive decisions alone: Use filters for insight, but talk to customers or frontline staff before making big changes.

What actually works (and what doesn’t)

Worth doing: - Regularly check segments that map to business goals (e.g., high-value customers, new signups). - Use comments to spot trends before they hit your main NPS score. - Share insights, not just numbers.

Mostly hype: - Overly complex filtering logic (AND/OR chains, nested filters) rarely gives more value than a couple of basic slices. - Automated sentiment analysis—usually just noise unless you have thousands of responses per month.


Keep it simple and iterate

Advanced filters in Delighted are powerful, but only if you use them with purpose. Start with your most important customer segments, look for meaningful patterns, and don’t get lost chasing micro-trends. Most of the value comes from sticking to the basics and checking in regularly—not running every report under the sun.

If you’re always asking “so what?” after looking at your filtered data, you’re on the right track. Don’t let the tools overcomplicate things—dig in, look for what matters, and keep making small improvements. That’s how you actually move the needle.