My list had over a hundred items and I was watching almost nothing from it. Everybody I asked reported the same thing, which suggested it was not a personal failing.
The list and the mood are different systems
The core mismatch.
Things get added in an engaged, curious state — reading about a film, hearing a recommendation, browsing deliberately.
Things get watched in a tired, low-effort state at the end of a day.
The list is therefore populated by one version of a person and consulted by a different one with different appetites, and the second version reliably wants something easier than the first version chose.
This is well described in research on intertemporal choice — people select more demanding options for their future selves than they actually consume — and it applies precisely here.
The list is not sorted by anything useful
A practical failure on top of the psychological one.
Most watchlists are chronological or arbitrary, with no indication of length, mood, effort required or why it was added.
Faced with a hundred undifferentiated items at nine in the evening, the effort of choosing exceeds the effort of accepting whatever the home screen suggests.
Which is exactly what happens, and it is why the recommendation surface wins over the list every time.
The interface is not neutral
Worth stating because it explains the asymmetry.
The list is generally a menu item requiring navigation. The recommendations occupy the entire opening screen with large images and automatic previews.
One of those is designed to be chosen and the other is designed to be available.
That is not accidental. A service benefits from directing viewing toward what it wants promoted, and the watchlist is a container for intentions that compete with that.
What actually worked for me
Four changes, of which two mattered.
Recording why something was added, in a word or two. Recommended by a specific person, a director I follow, something I read about. This turned an undifferentiated list into something I could navigate by mood.
Splitting by effort. A short list of things requiring attention and a longer one of things I could watch while tired. The second gets used constantly and the first gets used on weekends.
Capping the list, which forces a decision. Adding something means removing something, and the removals are informative — anything I will not defend against a new addition was never going to be watched.
And deciding the night before rather than at the moment, which removes the choice from the tired version entirely.
Keeping it outside the service
The change that made the largest practical difference.
A list held in the service is limited to that service's catalogue, disappears when a title leaves, and is fragmented across however many subscriptions you hold.
A single list held elsewhere — a note, a document, a dedicated tracking application — covers everything, survives catalogue changes, and can be sorted however you like.
It also allows recording where something is available, which is otherwise a search every time and is a substantial share of the friction.
The honest conclusion
Most of the list was aspirational rather than intended, and pruning it was a relief rather than a loss.
What remained, at about fifteen items, gets watched. What was removed was mostly things I thought I should watch rather than wanted to, which is a distinction worth being honest about.
The general version of this applies to reading lists and saved articles equally, and the same fix works: fewer items, a note about why, and the decision made in advance rather than in the moment.
The paradox of choice, properly stated
Worth being careful here because the popular version overstates the research.
The finding that more options reduce satisfaction and delay decisions has been demonstrated in some contexts and has replicated inconsistently, and the effect appears to depend heavily on circumstances.
What does seem robust is that choosing is effortful, and that effort competes with whatever else you were going to spend attention on.
Which is enough to explain the watchlist problem without needing a stronger claim. At the end of a day, the effort of choosing from a hundred options exceeds the effort of accepting a suggestion, and the suggestion wins.
The shared household version
Harder again, and worth mentioning because it is where most viewing decisions actually happen.
Two people choosing together face a negotiation as well as a choice, and the outcome tends toward the least objectionable option rather than the best one.
The practical fix that has worked in my household is alternating — one person chooses without discussion, then the other next time.
It produces better viewing than consensus does, because consensus selects for inoffensiveness and alternation selects for enthusiasm.
The abandoned series problem
A related and more specific failure, since serialised content compounds it.
Stopping four episodes into a season creates a barrier to returning, because resuming requires remembering what happened.
The longer the gap, the higher the barrier, until returning requires rewatching, at which point most people do not.
Which is why partially watched series accumulate in a way that films do not, and why the practical advice is to finish or abandon deliberately rather than drifting away.