Kano Model: Classifying Customer Satisfaction

Not all features create the same user reaction. Some surprise and delight users, while others just meet expectations.

Why feature categories matter

Pendo's 2019 analysis found that 80% of features in the average software product in its dataset were rarely or never used¹. Low usage is a reason to investigate a feature's purpose and value before investing further.

Who This Guide Is For

This guide is for you if you're new to product management or want a way to classify features. It explains how to understand what users value and use that information when setting priorities.

No experience needed. We'll start with the basics.

The Solution: The Kano Model

Noriaki Kano and colleagues described the model in 1984.² It distinguishes feature categories by the satisfaction users associate with their presence and absence.

The Five Kano Categories

1. MUST-BE (Basic Features)

These are features users expect to find. They won't praise you for having them, but they'll be upset if they're missing.

Think of it like: A coffee shop having cups. Nobody thanks them for cups, but imagine if they didn't have any.

Examples:

• Password reset (every app needs this)
• Search bar on Amazon
• Save button in Google Docs
• Works on mobile phones

Why this matters: Missing basics can create dissatisfaction for the users who expect them.

Strategy: Meet the relevant expectation, then decide whether additional investment is justified.

2. ONE-DIMENSIONAL (Performance Features)

With these features, the better they work, the happier users become. Users actively compare these between products.

Think of it like: Internet speed. Users often value faster performance, though the value of an additional improvement can vary by use case.

Examples:

• How fast pages load
• Phone battery life
• Storage space (like Google Drive)
• Number of app integrations

Why this matters: Users choose products based on these comparisons.

Strategy: Match what competitors offer, or do better if you can.

3. ATTRACTIVE (Delighter Features)

These are unexpected surprises that make users happy. Users don't ask for them because they don't know they want them.

Think of it like: Finding a free cookie with your coffee order.

Examples:

• Spotify's year-end music summary
• Instagram's Stories feature (when it launched)
• Netflix's "Skip Intro" button
• Amazon's "Customers also bought" suggestions

Why this matters: These features can create satisfaction beyond what users expect.

Strategy: Consider these as potential differentiators, alongside evidence of user need and delivery cost.

4. INDIFFERENT Features

Users don't care about these features. They neither help nor hurt satisfaction.

Think of it like: The pattern on a coffee cup. Most people don't notice.

Examples:

• Dozens of color themes in settings
• Complex features nobody uses
• Detailed stats for casual users
• Extra customization options

Why this matters: The feature may not improve satisfaction for the surveyed users.

Strategy: Investigate the segment and use case before further investment or removal.

5. REVERSE Features

These features actually make users unhappy. Some users want them, but most don't.

Think of it like: Background music in a library. Some might like it, but most want quiet.

Examples:

• Videos that autoplay with sound
• Too many notification pop-ups
• Complicated setup for simple apps
• Forced social features in work tools

Why this matters: These features can reduce satisfaction for some users.

Strategy: Consider whether the feature should be changed, made optional, or avoided for the affected segment.

Feature Evolution Over Time

Features can change categories over time. A feature may move from attractive to performance or basic expectations as markets and user expectations change.

Think of it like smartphones:

• 2007: Touchscreens were amazing (Delighter)
• 2010: Screen quality mattered (Performance)
• Today: Touchscreens are a basic expectation for many smartphones

Current examples:

• Dark Mode: Was a delighter in 2018, now it's expected
• Cloud Storage: May be expected in products where users need access across devices
• AI Features: Their category depends on the product, users, and market expectations

Why this matters: Revisit classifications as the product and user expectations change.

Try It Now

Analyze your features with real data:

Sample Analysis

Feature: "AI-powered meeting summaries"

1. Survey 30 target users
2. Input responses into calculator
3. Result: 60% classify as Delighter, 30% Performance, 10% Indifferent
4. Decision: Consider the result with effort, reach, and strategic context.

How to Run a Kano Survey

The Kano Model asks two questions about each feature.

Why Two Questions?

One question isn't enough. You need to understand how users feel when the feature exists AND when it doesn't exist.

The Two Questions to Ask

For each feature:

1. If we HAVE this feature: "How would you feel if this feature was included?"
2. If we DON'T HAVE this feature: "How would you feel if this feature was not included?"

Give users these answer choices for both questions:

• I like it
• I expect it
• I'm neutral
• I can tolerate it
• I dislike it

Understanding the Results

Don't worry - you don't need to memorize a complex table. Here are the most common patterns:

Most Common Patterns:

1. Like it if present + Dislike if missing = Performance Feature
2. Like it if present + Neutral if missing = Delighter
3. Expect it if present + Dislike if missing = Basic Feature
4. Neutral both ways = Indifferent

The Full Classification Table (Reference Only)

Once you're comfortable with the patterns above, you can use this table for all combinations:

What the letters mean:

• A = Attractive (Delighter)
• O = One-dimensional (Performance)
• M = Must-be (Basic)
• I = Indifferent
• R = Reverse (makes users unhappy)
• Q = Questionable (conflicting response)

Selecting survey participants

Choose enough participants to represent the decision you need to make, and report the segments and limitations. Different user types may have different opinions, so analyze relevant segments separately.

Real-World Examples

Tesla (Electric Cars)

Tesla uses the Kano Model to stay exciting:

Delighters: Self-driving features, software updates while you sleep Performance: How far it drives on one charge, speed Basic: Safety features, air conditioning, connects to your phone

What they do: Tesla keeps adding new surprise features through software updates. This keeps customers excited even with an older car.

Slack (Work Chat App)

Slack shows how features evolve over time:

Started as Delighters (2013): Fun emoji reactions, GIF search Now Performance Features: Search quality, app connections Now Basic Expectations: Mobile app, notifications, file sharing New Delighters: Instant audio calls, AI summaries

Lesson: What was exciting in 2013 is now expected.

Zoom (Video Calls)

The pandemic changed everything for Zoom:

Before 2020 (Delighters): Fun virtual backgrounds, beauty filters During 2020 (Performance): Call quality, number of participants After 2020 (Basic): Screen sharing, chat, breakout rooms

What happened: Features that were "nice to have" became table-stakes overnight.

Notion (Note-Taking App)

Notion grew through delighters:

Delighters: AI writing help, web clipper Performance: How fast it loads, templates available Basic: Save your work, share with others, works on phone

Growth strategy: Users loved the delighters so much they told friends about it.

Common Beginner Mistakes

1. Thinking All Users Want the Same Things

The mistake: Assuming everyone sees features the same way. The fix: Remember that power users and beginners want different things. Survey them separately.

2. Doing This Once and Forgetting

The mistake: Running one survey and using it forever. The fix: Features change categories over time. Check again each year.

3. Making Basic Features Too Perfect

The mistake: Spending months perfecting a login page. The fix: Basic features just need to work well. Save your time for delighters.

4. Treating a category as a fixed allocation

The mistake: Allocating effort to a category without considering the product strategy or research. The fix: Use the results with delivery constraints and other prioritization inputs.

5. Trusting Surveys Too Much

The mistake: Users say they want something but never use it. The fix: After building, check if people actually use the feature.

AI Prompts for Kano Analysis

1. Convert User Feedback into Kano Categories

Analyze this user feedback to classify features using the Kano model:

Product: [your product name and description]
Target users: [describe your users]

Feedback data:
[Paste support tickets, app reviews, user interviews, feature requests]

For each feature mentioned:
1. Extract the feature and user sentiment
2. Classify using these patterns:
   - Basic: Complaints when missing, no praise when present
   - Performance: Direct comparisons, "faster/better" language
   - Delighter: Surprise, excitement, "love" language
   - Indifferent: "nice to have", neutral mentions
   - Reverse: Complaints about presence

Output as table:
| Feature | Category | Evidence Quote | Confidence | User Count | Action Needed |

Then provide:
- Top 3 broken basics to fix immediately
- Top 3 delighter opportunities
- Features needing more research

2. Create Executive Presentation for Kano Results

Create a presentation to explain Kano analysis to executives:

Audience: [CEO/VP/Board]
Meeting length: [15/30 minutes]
Business priority: [growth/retention/competition]

Kano results:
[Paste your feature classifications]

Business context:
- Current NPS: [score]
- Main competitors: [list]
- Engineering capacity: [team size]

Create 6 slides:
1. Executive Summary (business impact in one sentence)
2. The Problem (customer pain in dollars)
3. Simple Kano Explanation (with known company examples)
4. Our Analysis (key findings with quotes)
5. Action Plan (phased approach with timeline)
6. Expected Outcomes (specific metrics)

Include talking points and objection handling.

3. Identify Feature Evolution Timing

Analyze when features will change Kano categories:

Current features:
[List features with current categories]

Market context:
- Competitors: [list their features]
- Industry trends: [what's changing]
- User sophistication: [beginner/advanced]

For each feature, identify:
1. Current category and evidence
2. Signs it's transitioning (competitor adoption, user expectations)
3. Predicted category in 6, 12, 24 months
4. Confidence level and trigger events

Output timeline:
| Feature | Current | 6 Months | 12 Months | 24 Months | Action |

Strategic recommendations:
- Features to stop investing in
- Emerging delighters to build
- When to launch for maximum impact

4. Make Decisions with Limited Resources

Prioritize Kano features with real constraints:

Kano analysis:
[Paste feature classifications]

Constraints:
- Dev capacity: [weeks available]
- Budget: [amount]
- Must ship by: [date]
- Tech debt: [percentage of time]
- Key metric to improve: [metric]

For each feature calculate:
Score = (User Impact × Kano Weight) - (Effort + Risk)

Kano weights:
- Broken Basic: 3x
- Performance Gap: 2x
- Delighter: 1.5x
- Indifferent: 0.5x

Recommend:
1. Must do now (critical basics)
2. Quick wins (easy delighters)
3. Next quarter (major features)
4. Backlog (nice to have)
5. Don't build (indifferent/reverse)

Include trade-offs and risks.

5. Handle Different User Segments

Create strategy when segments disagree on Kano categories:

Segments:
[Describe each segment: size, value, growth rate]

Feature classifications by segment:
| Feature | Segment A | Segment B | Segment C |
[List features and how each segment sees them]

Business context:
- Revenue per segment: [amounts]
- Growth strategy: [which segments matter most]
- Development capacity: [constraints]

Analyze:
1. Conflicts (delighter for one, reverse for another)
2. Universal basics (everyone expects)
3. Segment-specific needs

Recommend:
1. Core features for everyone
2. Optional features by segment
3. Pricing tier strategy
4. Development priority order
5. How to handle conflicts

Output implementation plan with timeline.

Quick Start Guide for Beginners

Start Here (Your First Kano Analysis)

1. Pick 5 features from your backlog
2. Survey 20 users with the two questions
3. Use the common patterns to classify
4. Focus on one delighter and one basic to fix
5. Check back in 3 months to see if categories changed

Simple Planning Framework

This Quarter:

• Fix any broken basics (users are complaining)
• Build one delighter (to stand out)

Next Quarter:

• Match competitor performance features
• Add another delighter

Always:

• Keep basics working
• Watch for features changing categories
• Listen for user excitement or complaints

Combining with Other Frameworks

Kano + RICE

Use Kano as multiplier:

• Delighters: 1.5x (differentiation value)
• Broken Basics: 2.0x (urgency)
• Indifferent: 0.5x (deprioritize)

Kano + NPS

Detractors complain about basics. Promoters rave about delighters.

Kano + OKRs

• Increase Market Share: Prioritize delighters
• Reduce Churn: Fix broken basics
• Beat Competitor: Match their performance features

Key Takeaways

Kano classifies how the surveyed users respond to a feature's presence and absence. Revisit classifications as expectations and segments change.

Use Kano alongside evidence about reach, effort, and strategy. It identifies feature categories, not a complete delivery order.

Next Steps

Keep going:

1. Analyze features with our Kano Model Calculator
2. Prioritize with RICE Scoring
3. Track satisfaction with NPS Calculator
4. Quick prioritization with ICE Framework

Sources