Two Percenters Rule

Kill features used by only 2% of customers—they create complexity without moving metrics

Gibson Biddle
The art of product strategy and prioritization

Two Percenters Rule

"If you find an idea that only works for 2% of your customers, now you're creating complexity, one more thing to choose. What happens when I hit this button? ... We called it scraping the barnacles—just get rid of it." - Gibson Biddle

What It Is

The Two Percenters Rule is a product simplicity heuristic: kill features that only 2% of customers use. These low-adoption features create disproportionate complexity without meaningful business impact.

Gibson Biddle developed this rule at Netflix, where they would regularly audit feature usage and eliminate "two percenters" in a practice they called "scraping the barnacles."

The logic is straightforward:

  • Low-usage features confuse the 98% who don't use them
  • They add cognitive load ("What does this button do?")
  • They get forgotten during major updates
  • They consume engineering resources to maintain
  • They rarely improve core metrics like retention

How It Works

The Netflix Party Example

When evaluating whether Netflix should launch a "Netflix Party" feature (watch together with friends), Biddle's analysis:

Key question: What percent of Netflix members would use Netflix Party if launched?

Historical data:

  • Xbox Party (2008-2009): Biddle guessed 2%, it barely hit 5%
  • Because only 5% used it, they killed it

The math that matters: If only 5% of customers use a feature, can it possibly improve retention enough to justify the complexity?

At Netflix's scale:

  • 220 million members × 5% = 11 million users
  • Even if those 11M loved it, would it move the 2% monthly churn rate?
  • Answer: Almost certainly not enough to justify the complexity

The Profiles Disaster

When Netflix launched streaming, they forgot about the profiles feature from DVD:

"In the old days, there was a profiles feature. When we launched streaming, we forgot about the profiles feature for DVD. Like, oh, crap."

This illustrates why two percenters are dangerous: they're easy to forget because few people use them, but when you break them, those few users are angry.

How to Apply It

  1. Audit feature usage regularly: Instrument your product to track what percentage of users engage with each feature.

  2. Set a threshold: 2% is Biddle's rule of thumb, but calibrate to your context. For a product with 100 users, 2% is 2 people—that's probably too aggressive.

  3. Calculate impact potential: Even if a feature improves outcomes for its users by 50%, multiply by the percentage who use it. A 50% improvement for 2% of users is a 1% total improvement.

  4. Factor in complexity costs:

    • Engineering time to maintain
    • QA coverage required
    • Documentation and support burden
    • Cognitive load on all users
    • Risk of breaking during updates

  5. Scrape the barnacles: Once identified, actively remove low-usage features. Don't let them accumulate.

Usage Threshold Guidelines

Metric Threshold Action
< 2% of users Two percenter Strong candidate for removal
2-10% of users Low adoption Evaluate if it could grow or should be cut
10-30% of users Moderate adoption Keep but don't over-invest
> 30% of users Core feature Invest in making it excellent

When to Use It

  • Product audits: Quarterly review of feature usage
  • Before major launches: Clean up low-usage features that might break
  • Resource allocation: Identify features consuming maintenance effort without ROI
  • Simplicity initiatives: When your product feels cluttered

When NOT to Use It

  • Essential features with low frequency: Security settings might be used rarely but are critical
  • New features: Give features time to gain adoption before killing them
  • Differentiated features: Some features serve positioning even if usage is low
  • Compliance/regulatory requirements: Usage doesn't matter if it's legally required

The Retention Math

At Netflix, the core metric was monthly retention (inverse of churn). For a feature to matter:

Impact = (Retention improvement for users) × (% of users who use it)

Example:
- Feature improves retention by 0.5% for users who use it
- Only 2% of users use it
- Total impact = 0.5% × 2% = 0.01% retention improvement
- At 2% monthly churn, this is nearly unmeasurable

For scale comparison:

  • Netflix's churn moved from ~10% (early days) to ~4.5% (2005) to ~2% (today)
  • A feature needs to move the needle meaningfully to justify existence
  • Two percenters rarely can

Source

  • Guest: Gibson Biddle
  • Episode: "The art of product strategy and prioritization"
  • Key Discussion: (23:45-25:45) - Two percenters concept with Netflix Party example
  • YouTube: Watch on YouTube

Related Frameworks

Back to Product Strategy