Audience studies with cluster analysis

You publish a study and uxspire groups the answers using machine learning. The result is segments with their share, their defining traits and one persona set card per segment.

No new snippet: audience studies run on the single uxspire tag you already have.

Start free How it works
23 Industry templates, ready to use
100% GDPR-compliant · hosted on servers in the EU
2 bis 5 Personas per study, no statistics knowledge needed
Cluster map · 4 segments
L Power user "Lena" · 28%
M Evaluator "Markus" · 31%
S Occasional "Sara" · 24%

Personas from study answers.

Many personas are created once in a workshop and never updated afterwards. An audience study builds them from the answers your users actually gave.

● Before

Scattered answers, missing patterns.

Answers arrive as a table. Anyone looking for the patterns in it merges the fields by hand.

● With audience research

Clear segments, instantly readable.

The study groups the answers by response pattern. You get segments with size, defining traits and persona set cards.

From the answer to the persona, in four steps.

The study works with the answers you collect in uxspire. Exporting them into another tool is not required.

Study responses 4 question types
SV
Single Choice · Needs
1,284 responses
CS
Rating · Usage value
812 responses
UE
Likert · Expectations
624 responses
EV
Free text · Motivations
438 responses
+
Use a template
Blank · Template · Questions
Cluster detection k = 4 · adaptive
Detected segments 4 of 4 · editable
PU
Power user "Lena"
Daily usage, high value, drives referrals. Wishes for deeper automation.
28%Share
EV
Evaluator "Markus"
Actively compares with competitors, fails at the permissions model.
31%Share
GE
Occasional user "Sara"
Monthly usage, high benefit. Onboarding is too long for her use case.
24%Share
CH
Hesitant decision-maker "Jonas"
High comparison intensity, a missing key feature. Purchase barriers become clearly visible.
17%Share
Persona set cards PDF ready
→
Strongest differentiators per segment
Cluster · instantly visible
Signal
→
Above- and below-average answer patterns
Persona · data-driven
Profile
→
Targeting cautions for messaging
Marketing · more precise
Note
→
Refine cluster names editorially
Research · in the product
Editable
→
Set cards as PDF to stakeholders
Export · 1 click
Ready

23 industry templates.

uxspire ships with 23 prepared audience study templates, from hotels and pharmacies to insurance and e-commerce. Sections, scales and logic are pre-configured and fully editable.

23 industriesEnglish · instantly availableSections, scales & logic preconfiguredFully editable

Four views on the same clusters.

Comparison, matrix, radar and focus are four analyses at different levels of depth. Comparison shows the biggest differences between clusters, matrix every single question including open-ended answers, radar the profiles across all dimensions, focus one individual segment in detail. All of them run on the same data set.

analysis · comparison
The biggest differences between the clusters.
  • Strongest deviations. Shows what truly separates audiences, sorted by discriminative power.
  • For first hypotheses. Which patterns jump out, which segments behave differently from the average?
  • Deliberately shortened. Top findings only. Open-ended answers are in matrix and radar.
Top findingsby largest cluster spread
Segment 1Segment 2Segment 3 Dot = cluster value · Bar = range
The complete question analysis.
  • Every question in study order. How each cluster performs per answer option, nothing is cut short.
  • For detailed work. Systematically review every question, every segment, every answer.
  • Open text included. Word cloud, cluster filter and raw answers show wording per segment.
Questionnaire6 questions · study order
Visual pattern recognition.
  • Similar or different? Shows per dimension where clusters align and where they diverge.
  • Quick segment pictures. Shapes, gaps and dominant axes per dimension.
  • The web doesn't fit? Short scales or text questions appear as a bar or answer card.
Profile radar9 dimensions · 3 clusters
Hover over an axis or area for cluster values and the related question.
One cluster at the center.
  • What makes the segment special? Where it responds more strongly and more weakly than all the others.
  • For persona work. Read one segment, sharpen it and translate it into language for marketing, product or research.
  • Purely quantitative. Measurable values only. Open-ended answers stay in matrix and radar.
In focus:

Three teams, three applications.

Product, marketing and sales read the same study differently.

Product & UX

Back up roadmap decisions.

Which audience needs more guidance? Which one rates the benefit highest? You can back up your argument with the share and the response values per segment.

Cluster comparison by question
Persona set cards per segment
Onboarding hypotheses with evidence
Marketing & Lifecycle

Sharpen the message per segment.

You see which answers a segment gives more or less often than average. From there you derive language, hook and offer per segment.

Messaging signals per persona
Targeting cautions
PDF set cards for campaign planning
Sales & CS

Understand objections earlier.

Which segments compare more with competitors? Where is a feature missing, where is clarity missing? Sales and customer success go into conversations with these answers.

Objection patterns per segment
Cluster-based talking points
PDF export for stakeholders

Common questions

No. The calculation runs inside the product. You see segments, shares, differentiators and persona set cards, and you do not have to set the method up yourself.
You can start an analysis from 100 received responses onward. The more reliable responses you collect, the clearer the clusters, differentiators and persona set cards become.
Yes, via the questions in your audience study. You decide which answers, scales and free-text fields feed into the segmentation. Workspace-wide events and custom properties are currently not automatically adopted as cluster dimensions.
All study data stays in your workspace and is processed on servers in the EU. User identifiers are handled via pseudonymized profile IDs; external LLMs are not used for cluster computation. A data processing agreement is part of enterprise contracts.
Audience research is included in Demo, Starter, Growth and Enterprise. Current limits: Demo 1, Starter 1, Growth 5 and Enterprise 10 audience studies. Enterprise teams receive individual contract and data protection options.

Create your first audience study.

Segments and persona set cards from your study answers. 23 industry templates, analysis in four views, export as PDF.

Persona set cards 23 industries Cluster method GDPR-compliant
Get started In 3 minutes
Configure your study Request a Demo
Or explore
In-Product Surveys Feedback in the product UX metrics NPS · CSAT · UEQ