Why a draft needs to drift before it settles

Almost every writer I've worked with arrives at the same moment: a draft that once felt promising now feels shapeless, and the temptation is to fix it immediately, sentence by sentence, before the larger problem has even been named. It rarely works. A draft in that state usually isn't broken — it's still deciding what it wants to be.

The drift is doing something

When a piece wanders from its opening premise, it's tempting to read that as failure. More often it's the writer discovering, mid-draft, what the story is actually about — a discovery the outline couldn't have predicted. The wandering paragraphs are not waste; they're evidence of thinking that hasn't yet found its shortest path to the page.

This is where a second reader helps more than a second pass by the same writer. It's difficult to see your own drift from inside it. An outside read can name the shape a piece is settling into before the writer has consciously noticed it themselves.

A draft that drifts honestly is closer to done than a draft that never left its outline.

What a diagnosis read actually looks for

Rather than marking sentences for correction, a useful read maps the piece paragraph by paragraph: where the argument enters, where it pauses, where a new idea quietly takes over from the one the piece opened with. Naming that shift is often enough — the writer usually knows, once it's pointed out, which version of the piece they actually want to keep.

The goal isn't to force the draft back onto its original path. Sometimes the detour is the better story, and the opening paragraphs are what need to move. A plain description of where the piece has gone gives the writer the information to make that call themselves, rather than defaulting to whichever version came first.

Give the drift its due before deciding it's a mistake. Most publishable features went through a version that looked, from the outside, like it had lost the thread.