A PM thinking skill
One of the most common reasons products fail is that teams solve what they observe instead of understanding what is actually causing it.
These are visible signals.
But visible signals are not always the actual problem.
One of the biggest differences between average and strong product thinking is the ability to separate symptoms from underlying causes before proposing solutions.
Visible signals
Intuitive explanation
Symptom
A symptom is what you observe.
Examples:
• Orders dropped
• Users are uninstalling
• Engagement reduced
• Drivers are cancelling
• Watch time declined
Symptoms are visible.
They tell you something is wrong.
But they usually don’t tell you why.
Problem
A problem is the underlying reason causing the symptom.
Examples:
• Checkout became slower
• Recommendations became irrelevant
• Pricing confused users
• Trust reduced
• Onboarding became harder
Problems explain why the symptom is happening.
Why this matters
Symptoms are urgent, visible, and measurable. That makes teams react quickly. But reacting too quickly without understanding the underlying problem often creates bigger issues.
01
Teams build solutions that don’t actually address the underlying issue.
02
The visible symptom may improve temporarily while the real issue remains unresolved.
03
The solution itself may create frustration, clutter, trust issues, or workflow problems.
04
Months can be spent optimizing the wrong layer of the problem.
05
Repeated symptom-level optimizations can make the overall user experience worse over time.
PM thinking shift
Weak PM Thinking
Users are not opening the app enough. Let’s send more notifications.
Strong PM Thinking
Why are users not returning?
Where exactly is the friction?
Did something change?
Is this a trust problem?
A workflow problem?
A value problem?
Or is this naturally a low-frequency product?
Strong PMs investigate causes before proposing solutions.
Real-world product patterns
Many product decisions that users experience every day are actually examples of teams reacting to visible symptoms instead of understanding deeper problems.
Observed symptom
Users stop opening the app frequently after active job search periods.
Product response
More notifications:
• Someone viewed your profile
• You appeared in 18 searches
• New jobs for you
• Recruiters are hiring
Possible problem
This may naturally be a low-frequency product.
• Users may only need the app intensely during active job search periods.
Potential side effect
The app starts feeling spammy or artificially engagement-driven.
Observed symptom
TikTok engagement and watch time were growing rapidly.
Product response
Instagram aggressively shifted toward:
• Reels-first behavior
• Full-screen feeds
• Heavier recommendations
• TikTok-style discovery
What happened
Users and creators pushed back heavily.
Many felt:
• friend/follow graph weakened
• original Instagram value reduced
• content reach became unstable
Possible underlying problem
Instagram’s core user value may not have been identical to TikTok’s usage behavior.
Observed symptom
Need stronger monetization and seller promotion opportunities.
Product response
Search and discovery became increasingly filled with:
• Sponsored products
• Promoted listings
• Ads inside search results
User experience impact
Users increasingly struggle to:
• identify best products quickly
• trust rankings fully
• separate ads from relevance
Possible underlying problem
The core user need in shopping is trustworthy and efficient product discovery.
Interactive experience
A food delivery app notices that daily active users dropped by 12% over the last month.
Decision moment
What is the actual problem?
Thinking framework
Step 1
Visible Symptom
Step 2
Identify where it happens
Step 3
Break down workflow / funnel
Step 4
Investigate possible causes
Step 5
Validate with data
Step 6
THEN propose solution
Quick practice
Prompt 1
Users are uninstalling a budgeting app after signup. What might be the symptom? What could be the underlying problem?
One strong response
The symptom is that users uninstall after signup. A possible underlying problem is that onboarding feels hard, trust is low, or the initial value is unclear.
Prompt 2
Drivers are cancelling rides frequently. What additional information would you investigate before proposing a solution?
One strong response
Investigate where cancellations are concentrated: geography, time of day, rider type, wait time, payout expectations, and any recent operational change.
Prompt 3
A grocery delivery app sees lower repeat orders. Why might discounts alone fail to solve the issue?
One strong response
Discounts may temporarily move the visible metric while the real issue remains. Repeat orders might be falling because selection, delivery quality, reliability, or trust is weak.
Reflection
Think about a product feature you personally found annoying, unnecessary, or confusing.
What visible symptom do you think the company was trying to solve?
What might have been the deeper underlying problem?
Takeaway
Symptoms are visible.
Problems are causal.
Strong PMs don’t just react to metrics — they investigate what is actually driving them.
Next steps