Sequels have a poor reputation that is broadly deserved, and the exceptions are consistent enough to suggest what actually determines the outcome.
Why most are worse
Structural reasons rather than laziness.
The original had a reason to exist and the sequel has a commercial reason, which is not the same thing and does not produce the same decisions.
The surprise is gone. A substantial part of what worked was novelty, and novelty is not repeatable by definition.
The ending has to be undone. A first film that resolved satisfactorily must have its resolution disturbed, which frequently means retroactively diminishing it.
Escalation is the default response and it is a trap. Bigger stakes, more characters, wider scope, none of which addresses why the original worked and all of which dilutes it.
And the team changes. Directors and writers move on, and a sequel made by different people is a cover version.
What the successful ones did
The pattern is more consistent than I expected.
They changed genre or register rather than repeating. The continuation is a different kind of story with the same characters, which sidesteps the novelty problem entirely.
They took the original seriously as a foundation rather than as a brand, developing consequences that the first film implied rather than inventing new incident.
They kept the people. Continuity of director and writer is the strongest single predictor in the cases I can think of.
And they had something to say that required a second film, which is the version of the same point that matters most.
The escalation failure specifically
Worth examining because it is the most common and the most avoidable.
The instinct is that a sequel must be bigger, and this is close to always wrong.
Scale is not what made anybody care. Character and specificity did, and both are diluted by expansion.
A sequel with a larger cast gives each character less time. A sequel with higher stakes makes the personal stakes proportionally smaller.
The continuations that work frequently go smaller, which is counterintuitive and is what the evidence suggests.
The third-film problem
A specific pattern that recurs enough to be worth noting.
Second films are frequently good and third films frequently are not, and the reason appears to be that a second film benefits from having a foundation and freedom, while a third must resolve everything.
Resolution is harder than development, and a trilogy conceived after the first film succeeded has generally not planned an ending.
The ones that work were planned as a whole, which is rare and expensive and requires a commitment made before anybody knows whether the first will succeed.
The legacy sequel
The current dominant form and worth assessing on its own terms.
A continuation arriving decades later, returning original cast, and passing the story to new characters.
The good versions use the interval, making the passage of time the subject rather than an inconvenience.
The poor versions treat the interval as a gap to be papered over, and use the returning cast as reassurance while the actual story goes elsewhere.
The distinguishing question is whether the film would be worth making without the returning characters. If not, they are functioning as marketing.
The commercial reality underneath
Worth acknowledging because it explains the volume.
A sequel to a known property carries far lower marketing costs and far lower risk, which is why studios facing expensive productions prefer them overwhelmingly.
That is a rational response to the economics rather than a failure of ambition, and it is unlikely to change while production budgets remain where they are.
Which means the sensible expectation is that most continuations will be made for commercial reasons and that some of those will be good anyway, because the people making them occasionally have something to say.
The test I apply
Before watching: is the director the same, and is the story a consequence of the original rather than a repetition of it.
Those two questions have predicted the outcome for me more reliably than anything else, including reviews, and they are answerable before release.
The gap that helps
An observation about timing.
Sequels made immediately, while a production team is intact and a schedule is set, are frequently rushed and made without a story.
Sequels made after a long gap are frequently better, because they were made when somebody had an idea rather than when a slot required filling.
The counter-case is that a long gap loses the cast, the audience and the tone, which is why the successful long-gap continuations are notable rather than typical.
What both cases share is that the good ones started with a reason to exist, which is the whole argument in one sentence.
The prequel variant
Worth a note since it has become common and has a distinct failure mode.
A prequel knows its destination, which removes uncertainty about outcomes and forces the interest onto how rather than whether.
That can work well, and it requires the how to be genuinely interesting rather than merely connecting known points.
The common failure is a story that exists to explain something which did not need explaining, and which is diminished by explanation. Mystery is frequently the asset being spent.