False Proxy Filter
Reject easy metrics that do not measure the change you seek
- Difficulty
- Moderate
- Time to result
- ~weeks to results
- Steps
- 5
- Confidence
- 99%
The False Proxy Filter challenges metrics by asking what they actually prove. Godin compares social-platform numbers with phrenology: both can invite people to infer a meaningful trait from an unrelated visible signal. Ten million TikTok views prove that a video received ten million views, but not that trusted people changed in the intended way. The filter begins with the real job and the change the work seeks to make. Each available metric is then tested for whether it demonstrates that outcome or merely records activity on somebody else's platform. Metrics that do not connect to the job are deliberately ignored, preventing them from steering creative and strategic decisions toward reach for its own sake.
Origin
Godin used phrenology's false inference from head bumps as an analogy for treating views, reviews, and similar platform numbers as evidence of meaningful impact.
Core principles
- 01A measurable number is not automatically a useful measure
- 02Platform attention can be unrelated to the actual job
- 03False proxies redirect effort toward the wrong beneficiary
- 04Measure change in the people who trust you
How to run it
- 1
State the job
Write down what the work exists to accomplish beyond generating activity on a platform.
Pro tip Describe a change in the people who trust you.
Watch out Do not define the job with the same metric you are evaluating.
- 2
Inventory the proxies
List the views, reviews, likes, rankings, or other visible numbers that currently influence decisions.
- 3
Ask what each number proves
Interpret every metric literally, then test whether it demonstrates the intended human change.
Pro tip Use Godin's formulation: a view count first proves only that the views occurred.
Watch out Correlation or visibility alone does not establish relevance.
- 4
Ignore false proxies
Remove unrelated measures from routine attention so they cannot redirect the work.
Watch out Collecting a metric while claiming to ignore it can still shape behavior.
- 5
Watch for actual change
Use direct evidence that the trusted audience moved, learned, decided, or behaved differently in the intended direction.
In the wild
Godin says that ten million TikTok views measure only whether the content received ten million TikTok views. They do not establish that the creator fulfilled the job of changing trusted people.
→ The view count is rejected as a substitute for the intended change.
Common mistakes
Optimizing what is visible
Easy access to a number can make it feel strategically important even when it measures the wrong thing.
Working for the platform
Chasing platform metrics can turn platform satisfaction into the goal instead of serving the people who trust the work.
Is it for you?
Best for
It is best for work surrounded by abundant engagement, reach, review, or platform metrics.
Not ideal for
It is not ideal when a metric has a proven direct relationship to the intended outcome and operational decisions.
From the transcript
“Getting 10 million views on TikTok is a measure of only one thing. Did you get 10 million views on TikTok?”
“Your job is not to make YouTube happy. Your job is to do your job, which is to change the people who trust you.”
“And I refuse to waste my time looking for false proxies.”
From the episode
Big Business Growth: Strategy Marketing with Seth Godin (a.k.a. Willy Wonka)
Seth Godin