What Every YouTube Tracker Got Wrong

Before building Gazenest I spent about six months trying every tool that existed for managing YouTube consumption. I used screen time dashboards, extension-based feed hiders, time-tracking apps, and a couple of outright blockers. Each solved something. None solved the specific thing I needed. Understanding why required articulating what that thing actually was.

The tools I tried and what they measured

YouTube's built-in Take a Break reminder sends a notification after a configurable interval. It is easily dismissed. It does not log anything, accumulate data, or affect recommendations. It is a clock, not a tracker.

Screen Time on iOS and Digital Wellbeing on Android report total minutes per app. This is useful as a rough baseline and nearly useless for behavior change. Knowing you spent 3h 12m on YouTube this week does not tell you whether those were sessions you chose or sessions that happened to you. A week with 3h 12m of deliberate documentary watching and a week with 3h 12m of compulsive Shorts scrolling look identical in the report.

Unhook, DF Tube, and similar extensions hide UI elements: the recommendation sidebar, the home feed, the Shorts shelf, autoplay. This is genuinely useful friction design. The feed hiders reduce the most aggressive recommendation surfaces. Their limitation is that they record nothing and score nothing. You can spend two hours on YouTube with Unhook active and leave with no information about how those two hours went or whether the pattern is changing over time.

RescueTime and Toggl track application usage, not content. They know you had a browser open; they do not know you were on YouTube, let alone whether you were watching what you intended. Domain-level blocking in RescueTime is available but it is the same blunt instrument as a dedicated blocker, with the same limitations.

Blockers (Freedom, Cold Turkey, StayFocused) prevent access above a threshold or during a schedule. Their strengths and weaknesses are covered separately, but the core limitation relevant here is the same: no data, no scoring, no session quality signal. The block fires or it does not.

The question none of them answered

Every tool I found answered some version of "how long were you on YouTube." None answered the question I actually cared about: were you in charge of this session, or was the platform in charge?

That distinction maps onto something concrete. If you opened YouTube with a declared purpose and watched what you came for, the session served you regardless of duration. If you opened YouTube reflexively, followed three recommendation chains, and closed the tab confused about where forty minutes went, that session did not serve you even if it was short.

Duration is the easy metric. It is quantifiable, unambiguous, and familiar from screen time conversations. It is also insufficient. A tool that measures duration and nothing else is giving you the GPS coordinates of a problem without telling you what caused it or how to address it.

What was actually missing

The gap was a quality signal: a score that reflects whether a session was intentional or was driven by the algorithm. Not just time, but whether you were steering.

This requires two things that none of the existing tools had: session intent logging and behavioral scoring derived from the match between stated intent and actual behavior.

Session intent means capturing, before a session begins, what the user is there to watch. Without that, there is no baseline against which to measure. The session ends and you have a duration and a list of videos. You do not know whether those videos were what you came for.

Behavioral scoring means deriving a number from the relationship between intent and behavior over time, not just for one session but as a rolling metric. A single low Clarity score is a bad session. A month of declining Clarity scores is a pattern worth addressing.

That is the scoring layer Gazenest is built on. Self-Control reflects how often you arrive at YouTube with a purpose versus how often you open it reflexively. Clarity reflects how closely you followed through. Diversity reflects whether your content range has narrowed without your noticing. Together they answer the question that duration alone cannot: how much of your YouTube consumption is actually yours.

The honest version

Existing tools are not wrong. Feed hiders reduce a real source of unintended viewing. Time trackers establish a baseline. Even blockers serve a purpose for compulsive-open patterns that need an external interrupt. I still recommend Unhook to people who want minimal-setup friction without tracking.

Gazenest sits in a different category: not reduction for its own sake, but legibility. The goal is that you can look at your week and know, accurately, how much of your YouTube time you chose. That knowledge is what changes behavior durably. Everything else is either a wall or a number without context.