Rearranging the timeline changes what the audience has to do

There is a short answer about non-linear storytelling and a useful one, and they are not the same. What follows is the useful one.

Order determines which question the audience is asking

Chronological telling naturally generates the question of what happens next, which is the strongest and simplest engine a story has. Revealing an outcome early destroys that question and replaces it with how and why, which is a different and often more interesting engine.

This trade is the entire justification for non-linear structure, and a film that makes it without gaining a better question has simply become harder to follow. Stories about inevitability, memory or investigation are the ones that usually profit, because their real subject is causation rather than outcome. Where the pleasure of a story is genuinely in surprise, rearranging it is close to self-sabotage.

Flashbacks are interruptions and must be paid for

Leaving the present line always costs momentum, so the departure needs to answer something the audience is actively wondering about. Placed at the point of maximum curiosity, a flashback reads as a satisfaction; placed arbitrarily, it reads as the film changing the subject. The strongest ones alter how the audience understands the scene they just left, so returning to the present feels different rather than merely resumed.

Length matters as much as placement, since a departure long enough to become its own story makes the return feel like an intrusion. Writers often find that a flashback carrying only information can be replaced by a line, whereas one carrying a changed meaning cannot.

Orientation has to be constant and cheap

Audiences tolerate complex chronology as long as they always know where they are, which requires reliable and unobtrusive signals. Hair, costume, location, colour and aspect ratio all do this work, and the strongest films establish the code early and then use it consistently.

Explicit captions with dates are the bluntest option and are unfairly despised, since they cost a second and prevent minutes of confusion. Confusion is only productive when the film intends it and resolves it, and unintended confusion is indistinguishable from incompetence to a viewer. The reliable diagnostic is whether a viewer who briefly looks away can re-orient within a few seconds of looking back.

Repetition with new information is the sharpest tool

Showing the same event twice from different vantage points converts a scene the audience has already understood into one they misread. The technique depends on the first version being genuinely complete and honest within its own limits, rather than withholding through obvious tricks.

Where the first version cheats, audiences feel manipulated rather than surprised, and the distinction is precise: withholding is fair, misrepresenting is not. Structures built entirely from this move risk diminishing returns, because the third revelation of the same event is rarely as strong as the first.

Used once at the right point it can reframe an entire film without a single new plot development.

Parallel timelines have a rhythm problem

Two periods running alongside each other must be cut so that each departure lands on a question in the strand being left. Uneven weighting is the usual failure, where one strand is plainly the real story and the other is felt as an obligation to be endured. A useful discipline is ensuring each strand changes the meaning of the other, so the cutting between them is doing something beyond alternation.

In practice, convergence at the end is conventional and not compulsory, and strands that never meet can still resolve if their thematic relation is clear. The editor usually determines the final rhythm, which is why parallel structures often change substantially between script and finished film.

Why the technique goes in and out of favour

Fragmented structure spread widely enough that it became a recognisable style, at which point audiences began reading it as a signal of ambition rather than a choice. That familiarity reduced its power, since a device only disorients while it is unexpected and eventually becomes a convention like any other. Streaming viewing habits complicate it further, because a viewer watching across several sittings loses the continuity that complex chronology assumes.

By the second act, some writers have responded by simplifying structure and locating complexity in behaviour instead, which is a defensible correction. The underlying principle survives the fashion: rearrange only when the rearrangement is the point.

The takeaway

Move the order only when the assembly work you hand the audience pays them back.

Watch the transitions. That is where the argument of a film usually is.