The screen is easy to count. The audience is the hard part.
Digital signage and DOOH live or die on who actually sees them — yet "audience" hides a chain of soft assumptions. This is the working explanation of how the people in front of a screen get measured, and how not to fool yourself.
Why measurement is the whole game
A screen network's value rests entirely on the audience it reaches, and "audience" is a number built from assumptions every step of the way. Counting people is not the same as counting people who looked; people who looked is not the same as people who acted. Each layer is a reasonable estimate and each can be stretched or shrunk depending on who is doing the counting and why. The operators and advertisers who get value from screens are the ones who understand what each number actually measures — and the ones who get fooled are the ones who treat an "impressions" figure as a meter reading when it is really the end of a chain of soft adjustments.
The honest version
Good audience measurement is honest about its own uncertainty. A door counter or a camera-based people-counter gives a defensible count of bodies; attention estimation (did they face the screen, for how long) is softer and more easily overstated; outcome measurement (did the screen change behaviour) is the most valuable and the hardest to isolate from everything else happening. The skill is not pretending the soft numbers are hard — it is using each at the confidence it deserves: counts for reach, attention for relative comparison, outcomes for the few cases you can genuinely measure. A measurement practice that knows the difference is worth far more than one that quotes a single confident figure built on sand.
What this primer covers
These pages take audience measurement in the order the assumptions stack up. Counting versus attention separates the two things most often conflated. Privacy and anonymous measurement covers how to count people without surveilling them — the line that keeps the whole practice legitimate. And from data to action is the part that justifies the rest: turning the numbers into screens that actually work better. None of it requires exotic technology; all of it requires intellectual honesty about what a number means.