A series ends on a major unresolved development and is then cancelled. It happens constantly and it is the outcome viewers find hardest to forgive.
It is almost never a decision to end that way. It is a timing problem, and understanding it explains why it keeps happening.
The sequence that produces it
Writing and production happen months before broadcast. A finale is written and shot long before anybody knows how the season performed.
At the point of writing, the production does not know whether it is renewed. It usually has no signal at all.
So the writers face a choice: end conclusively, which wastes an opportunity if renewed, or end openly, which sets up the next season.
Most choose openly, because the incentive structure rewards it. A show that ends conclusively and is then renewed has to manufacture a new situation from nothing, which is harder than continuing one.
The renewal decision then arrives after broadcast, based on performance data, and if the answer is no, the cliffhanger is permanent.
Why the decision comes so late
Because the data does. Viewing accumulates over weeks, and platforms want completion and retention figures rather than premiere numbers.
Costs also escalate between seasons — cast contracts have built-in increases, and a successful show becomes more expensive to renew each year.
And commissioning decisions are made against a slate, so an individual show is weighed against everything else competing for the same budget, which is not settled until late.
What has changed with streaming
The problem has arguably worsened, for a few reasons.
Data is not published, so nobody outside the platform can anticipate a decision. Under the old ratings system, everybody could see a show was in trouble, and writers frequently hedged accordingly.
Cancellation decisions appear to come faster in some cases, with shows ended after one or two seasons on metrics nobody outside can evaluate.
And the serialised format that streaming encourages produces stories that are structurally unfinished at any point other than the end, which maximises the damage when it stops.
The two-part finale trick
A practical response some showrunners have adopted, and it works.
Write a finale that resolves the season's main story satisfactorily, and place the setup for the next season in a short coda.
If renewed, the coda is the launch point. If not, the audience has an ending and a lingering question, which is a considerably better outcome than a cliffhanger.
The best-executed examples of this are ones nobody notices, because a show that ended well does not generate an argument.
What happens to unresolved shows
Occasionally something, usually nothing.
Some have been resolved in other media — comics, novels, a film — where a rights holder saw value in completing it.
A few have been revived by another platform, which happens more than it used to and remains rare relative to the number of cancellations.
And showrunners sometimes publish what was planned, which is genuinely appreciated by audiences and which some are contractually prevented from doing.
The rights position is the obstacle in most cases. A cancelled show still belongs to somebody, and completing it elsewhere requires their agreement.
What viewers can reasonably conclude
That nobody set out to leave it unresolved.
That the writers were making a rational bet under uncertainty and lost it.
And that the practice of watching a series only after it has been confirmed complete, which a lot of people have adopted, is a rational response — while also being one of the reasons first-season viewing figures are soft, which contributes to cancellations.
That is a genuinely circular problem and I do not think anybody has a good answer to it.
The contract clocks
An additional pressure that shapes the timing.
Cast and crew contracts contain option dates by which a production must commit or release them, and those dates frequently fall before viewing data is complete.
Which means a platform sometimes has to decide whether to hold a cast — at cost — before it knows whether it wants another season.
Letting options lapse and attempting to reassemble a cast later is possible and difficult, and it is why some revivals take years and arrive with recast roles.
What audiences can do, realistically
Very little, and the little there is has some effect.
Completion matters more than viewership. Platforms weight whether people finish a season heavily, so watching to the end counts for more than starting.
Watching soon after release matters, since decisions are frequently made on a window of a few weeks.
Campaigns after a cancellation occasionally work, and the successful ones have generally involved demonstrable commercial interest rather than volume of complaint.
None of that is a reason to watch something you are not enjoying, and it is worth knowing that the thing which counts is finishing rather than protesting.
The limited series alternative
The structural answer that some creators have adopted deliberately.
A story commissioned with a fixed length has no renewal question, and can be written toward an ending from the first episode.
The commercial complication is that a successful limited series creates pressure for a second, which is how several supposedly complete stories acquired continuations that nobody had planned.
The ones that resisted are generally the better for it, and the ones that did not are a reasonable illustration of why the original commitment was worth making.