Product Mental Models
How to build products customers love.
TL;DR
- Jobs to be done - What progress do customers want?
- Working backwards - Start with press release
- Opportunity cost - Yes to X = No to Y
- MVP - Ship smallest valuable thing
- Reversible fast, irreversible slow
- Customer development - Talk to users constantly
- Kano model - Basic, Performance, Delighters
- First principles - Break down to fundamentals
- 80/20 rule - Focus on what matters
Jobs to Be Done
What it is: People don't want products, they want progress.
How to use it:
- What job is the customer trying to do?
- What progress are they trying to make?
- Build the solution that helps them make that progress
Example:
- Bad: "People want a faster drill"
- Good: "People want a hole in the wall. Ship pre-drilled shelves."
When to use: Understanding customer needs, designing features.
Working Backwards
What it is: Start with the customer experience, work backwards to build it.
How to use it:
- Write the press release for launch day
- Write the FAQ customers will ask
- Build the product that makes both true
Example: Start with: "Today we launched instant transfers. Send money in 2 seconds, not 2 days." Then build what makes that true.
When to use: Planning new products or features.
Opportunity Cost
What it is: Saying yes to X means saying no to Y.
How to use it:
- For every "yes", ask: "What am I saying no to?"
- Make the tradeoff explicit
- Choose the bigger opportunity
Example: "If we build feature A (2 months), we can't build feature B (2 months) until Q3. Which matters more?"
When to use: Prioritizing features or projects.
Minimum Viable Product (MVP)
What it is: Ship the smallest thing that creates value.
How to use it:
- Identify core value proposition
- Remove everything else
- Ship and learn
Example:
- Bad: "Let's build 10 payment methods before launch"
- Good: "Ship with credit card only. Add more based on requests."
When to use: Launching new products or features.
Reversible vs Irreversible Decisions
What it is: Type 1 (one-way door) vs Type 2 (two-way door) decisions.
How to use it:
- Type 2 (reversible): Decide in <1 hour, move fast
- Type 1 (irreversible): Take days/weeks, gather data
Examples:
- Type 2: Feature prioritization, UI changes, pricing experiments
- Type 1: Platform choice, database migration, brand repositioning
When to use: Every product decision.
Customer Development
What it is: Talk to customers before, during, and after building.
How to use it:
- Before: "What's the hardest part of X for you?"
- During: "Does this solve your problem?"
- After: "What would make this 10x better?"
Example:
- Bad: Build for 6 months, launch, hope customers love it
- Good: Talk to 10 customers, build MVP in 1 week, iterate based on feedback
When to use: Every stage of product development.
Kano Model
What it is: Features fall into 3 categories: Basic, Performance, Delighters.
How to use it:
- Basic: Must-haves (login, payment). Customers expect them.
- Performance: More is better (speed, accuracy). Competitive advantage.
- Delighters: Unexpected wow moments. Differentiation.
Example:
- Basic: App loads
- Performance: App loads in <1 second
- Delighter: App predicts what you need before you ask
When to use: Prioritizing features, allocating resources.
First Principles Thinking
What it is: Break problems down to fundamental truths.
How to use it:
- Identify assumptions
- Ask: "What must be true?"
- Build solution from there
Example:
- Assumption: "Food delivery needs drivers"
- First principle: "People want food from restaurants at home"
- Solution: Partner with restaurants who have their own delivery
When to use: When conventional approaches aren't working.
80/20 Rule (Pareto Principle)
What it is: 80% of results come from 20% of efforts.
How to use it:
- Identify the 20% of features that deliver 80% of value
- Ship those first
- Kill or delay the rest
Example: "90% of users only use 3 features: send, receive, check balance. Ship those. Hold the rest."
When to use: Feature prioritization, resource allocation.
Inspired by Untools and product leaders.