TV's Loyalty Crisis Is a Misdiagnosis

The real threat to serialized television is the twenty-month gap. Here's what thirty years of scribbling tell me about the Netflix audience attrition problem.

Netflix’s internal research is making the rounds. Their biggest shows can lose between thirty and seventy percent of their audience between Season One and Season Two. Beef. Avatar.

The media are calling it a loyalty crisis. After thirty years scribbling for television. My read is different.

The Myth of the Loyal Audience

People remember the 2000s as appointment TV’s golden age. All episodes watched. Every fan locked in. But inside, our POV was different. The Networks tracked weekly Nielsen numbers on Lost, Heroes, and Alias. The numbers dropped. They always dropped.

A hit broadcast drama might open to fifteen million viewers. Settle around ten or eleven. Finish the season a few million lower. That was normal. It was expected. The only question was whether enough people would stick with us to stay on the air.

Lost premiered to 18.6 million viewers. Season One averaged 16 million. The Season Two premiere drew 23 million, a series high, and the show went on to rank second only to CSI: Miami in global popularity surveys of the era. Heroes averaged 14 to 15 million during its breakout first season and peaked near 17 million. Alias across five seasons ran 9.7, 9.0, 8.2, 10.3, then 6.7 million.

Every one of those seasons lost viewers from its peak. Every one of those shows was considered a hit. Attrition aint failure. It’s inevitable.

Three Eras, One Career

My résumé tracks across three different economic models of TV.

Alias and Lost in the network era, where success meant weekly Nielsens, advertiser demographics, and tentpole scheduling. Heroes arriving as social media started fragmenting the water-cooler conversation. Hannibal in the prestige era, where critical conviction outweighed raw ratings. That show drew two to four million live viewers and still ran three seasons. American Gods in premium cable. Star Trek: Discovery crossing into streaming, where the CBS premiere drew 9.5 million viewers but Paramount measured our success in subscriber acquisition, not overnight ratings.

I watched the business shift from one model to the next and then to the next. Measurement metrics transformed. The audience didn’t.

What Changed Was the Clock

We had time on Network TV.

Twenty-two episodes. One per week. Summer reruns. DVD box sets. Syndication. Word of mouth had months to compound. Somebody might catch your show mid-season, binge the back half on DVR, and become a super-fan by the finale. Find it in syndication the following year and work backward through the catalog.

The audience wasn’t built on a binge over a launch weekend. It was accumulated over years, and across platforms.

The Office, Parks and Recreation, Breaking Bad, Buffy the Vampire Slayer. None of those shows were immediate hits. They grew. The business model had room for that kind of slow build. Then streaming compressed the entire process.

Eight episodes arrive on Friday morning. The platform knows most of what it wants to know within two weeks: completion rate, return rate, whether subscribers stayed or churned. Then the algorithm moves on to the next recommendation.

The conversation window on a Netflix drop runs roughly ten to fourteen days. A successful broadcast drama could sustain eight months of active discussion.

The Mega Gap

There’s a second problem that barely existed in broadcast. The wait between seasons. The average gap grew from about twelve months in 2020 to roughly twenty-one months by 2025.

Think about your own relationship with a show you loved two years ago. Can you name every supporting character? Do you remember the Season One mystery that was supposed to pay off?

It’s been three years and I’m still waiting for season two of 3-Body Problem and Blue Eyed Samurai. When they return, will I remember to tune in?

The audience disappears because these shows disappear from the lives of the audience long enough for them to fill the space.

The HBO Anomaly

There is a counterexample. Despite equally long production cycles, White Lotus and The Last of Us increased their audiences betwixt seasons.

Analysts point to two factors: weekly release schedules that sustain discussion rather than burning the conversation in a single weekend, and franchise identity strong enough that the IP becomes is appointment, not the plot of any specific season.

Shows with a strong enough brand, where audiences came back for what the show fundamentally is, retained viewers across the long gap. Shows built primarily around plot mechanics struggled.

Characters carry further than plot twists. Atmosphere carries further than mysteries.

Serial Scribes

Every episode must justify the next click. Every season must survive a twenty-month absence. Every finale must leave enough emotional residue that a viewer who watched it nearly two years ago still feels the pull when Season Two drops.

The question shifted: Can Season One compel enough viewers to finish fast, feel the loss when it’s over, and still care twenty months later?

That favors exceptional hooks, fast momentum, and specific characters who stay vivid across a multi-year gap. Relationships that sting when they’re over. Emotional payoffs that hit harder than plot reveals.

Plot twists fade. Relationships linger. That’s the structural work serialized television demands of its scribblers.

The More Things Change, The More They Stay The Same

Audiences return for curiosity, character, and conflict. Mysteries worth solving. People whose next decision matters. Stories that made them feel something they want to feel again and again.

The same engines that kept readers turning pages through Dickens serials in the nineteenth century, brought families back to weekly television in the 1980s, and compel binge-watchers clicking through one more episode at midnight.

I think the most critical factor hasn’t changed that much since the first storyteller convinced a comrade to stay by the fire for one more yarn. Technology and Distribution might evolve. Business models might adapt. Attention spans might collapse.

But human curiosity remains insatiable. So if Netflix can solve for that element, eliminate the mega-gaps, keep the quality at good enough, and maybe not cancel a promising show after one season, maybe they can arrest the attrition of their audience.