Marginal User Focus

"For me, it's a person who is just on the cusp of taking the action you want to take... When you go to the extreme and you find that person who's the worst, you see everything that's wrong with the product." - Adriel Frederick

What It Is

A growth optimization methodology that focuses on users who are on the verge of converting but face specific barriers preventing them from completing key actions. Rather than optimizing for average users or best-case scenarios, this framework identifies the "marginal user"—someone who wants to use your product but can't quite get there.

The framework also uses an extreme variant: finding the worst-case user to reveal all potential issues, then working backward to prioritize fixes that help the marginal user convert.

How It Works

Finding the Marginal User

Identify high-intent, low-conversion segments - Look for places where you have strong signals of interest (traffic, engagement) but poor conversion rates.
Go to the worst case - Find users in the most challenging circumstances to reveal everything that's broken.
Observe and talk, don't just analyze data - Data shows you how bad things are; observation and conversation show you why.

The Worst-Case Reveal Method

When Adriel worked on Facebook registration, he found value in examining the worst possible user experience:

"That's the marginal user. This is the person who is just on the cusp of coming in, wants to come in as you can see by the traffic, but we can't get them in."

In Facebook's case, this was: "A person on a feature phone on edge, trying to access Facebook in a country that was far from one of our data centers."

This extreme case revealed issues with:

Language detection
Phone number formatting
Connection latency
Data center proximity

From Worst Case to Priority

Once you've identified all issues from the worst case, you can prioritize by asking: "If this person had the best phone and internet connection, what would still be wrong?" This isolates the fixable problems from the structural ones.

How to Apply It

Find high-traffic, low-conversion segments - Filter your data to find countries, devices, or user segments with high intent signals but low completion rates.
Identify your worst-case user - Who is the person with the most barriers? Find them.
Observe them directly - Watch them use your product. Don't rely on data alone—you'll miss problems orthogonal to your funnels.
List every barrier - Write down everything that's wrong with their experience.
Prioritize by removing structural constraints - Ask: "If they had ideal conditions (best device, best connection), what problems remain?" Start there.
Repeat the observation - After each improvement, go back and watch marginal users again. New issues will emerge.

When to Use It

Optimizing registration or onboarding flows
Expanding into new markets or geographies
Improving conversion at any stage of your funnel
Before major redesigns to understand current pain points
When data analysis alone isn't revealing why users drop off

Key Insight: Data Shows How Bad, Not Why

"Don't use the data alone to figure out who the marginal user is. It'll give you a clue where they are and what might be wrong, will give you some hints. It's not going to give you the answer. You have to go watch them to find the answer."

Adriel shares a powerful example: watching a user in India sign up for Facebook and put their "full legal name" that nobody actually calls them by. This upstream decision sabotaged downstream actions (friend requests not accepted, people can't find you). The funnel data would show friend acceptance problems, but only observation revealed the root cause.

Source

Guest: Adriel Frederick
Episode: "Humanizing product development | Adriel Frederick (Reddit, Lyft, Facebook)"
Key Discussion: (~48:00-53:00) - Adriel explains the marginal user concept and shares the India registration example
YouTube: Watch on YouTube

Related Frameworks

Growth Competency Model - Builds on understanding who can execute this kind of analysis