Planning Writing Time Realistically (Not Optimistically)
Most missed writing deadlines are not caused by procrastination alone; they are caused by a planning error that happens before a single word gets written — estimating how long a document will take to draft using your typing speed instead of your writing speed. Those are very different numbers, and confusing them is the single most common reason a schedule that looked comfortable on paper turns into an all-nighter. Getting this estimate right is mostly a matter of using an honest pace and doing the arithmetic properly.
Typing speed is not writing speed
You can probably type sixty, eighty, maybe over a hundred words a minute when you are transcribing something you already know how to say — an email, a text, a sentence you have already composed in your head. Sustained drafting of new, original prose is a completely different activity. Every sentence requires you to decide what claim to make, how to support it, how it connects to the sentence before it, and whether it is actually true. That thinking happens in the pauses, not in the typing, and the pauses dominate the total time. A realistic sustained drafting pace for real analytical or academic writing is commonly a few hundred words an hour — nowhere close to typing speed, and the gap is exactly why "I type fast" does not predict "I draft fast."
The arithmetic, worked at one pace
Take a 5,000-word term paper as a working example, and assume a sustained pace of 300 words an hour — a reasonable middle estimate for considered, non-trivial writing. Dividing 5,000 by 300 gives a total drafting time of about 16.67 hours. Spread across a realistic 2 hours of focused writing time per day, that comes out to 9 days of work, using the plan-forward rule that any leftover partial day still counts as a full day of the schedule. That is the whole calculation: total hours needed, divided by hours you can actually give it each day, rounded up so a leftover half-day still counts as a full day on the calendar.
Why the pace assumption matters this much
The honest answer to "how long will this take" depends enormously on which pace number you plug in, and it is worth seeing the swing directly. At a more conservative 200 words an hour — realistic for unfamiliar material, heavy revision-as-you-go, or a subject you are still learning as you write — the same 5,000-word paper takes 25 hours, or 13 days at 2 hours a day. At a faster 400 words an hour, achievable when you already know exactly what you want to say and are mostly executing rather than discovering the argument as you go, the same paper drops to 12.5 hours, or 7 days. That is nearly a two-to-one spread in the total schedule from a single input assumption, and it is why guessing your own pace instead of tracking it is such an expensive habit — a wrong guess in either direction throws off the whole plan by days, not minutes.
Hours available per day matters just as much as pace. Keep the pace at 300 words an hour but raise your daily writing budget from 2 hours to 4, and the same 16.67 total hours compresses from 9 days down to 5. Doubling your daily hours does not simply halve the day count because of the ceiling rounding at the edges, but it gets close, and it demonstrates the second lever you actually control: not just how fast you write per hour, but how many hours a day you protect for writing at all.
Finding your own real number
The most reliable way to get an accurate pace is to stop guessing and measure it. Time an actual, typical drafting session — not editing, not outlining, genuine first-draft composition — and divide the words you produced by the hours it took. Do this two or three times across different kinds of writing, because your pace on familiar material with a clear outline will be meaningfully faster than your pace on a section you are still figuring out as you go. Once you have a real number, feed it back into the same arithmetic above in place of a generic assumption, and your schedule stops being optimism and starts being a plan you can actually hold yourself to.
Where AI genuinely changes the number — and where it does not
Used honestly, AI assistance can raise your effective drafting pace in specific, limited ways: a solid outline going in means less time spent staring at a blank section deciding what belongs in it, and a quick explanation of a concept you are stuck on can turn twenty minutes of confused searching into two minutes of clarity. Both of those are real time savings, and both happen before you start composing your own sentences. What does not honestly raise your drafting pace is having a tool generate the paragraphs themselves, because at that point you have not drafted anything faster — you have replaced drafting with something else entirely, and the time you saved is time you will often pay back later reworking prose that does not sound like you, verifying claims you did not originate, or simply rewriting it once you realize you cannot explain what it says. If you use AI for structure and for getting unstuck, it is honest to estimate a somewhat faster words-per-hour pace for yourself. If you are asking it to produce the sentences that end up in your document, you are no longer measuring your own writing speed at all, and the schedule you should actually be planning for is a very different, much riskier one, with the added risk that any invented facts or fabricated citations in the generated text are now yours to catch. If your institution requires disclosure of how AI was used in a piece of writing, this is also the distinction that disclosure statement needs to capture clearly: whether the tool helped you write faster, or whether it wrote some of the words that ended up in the document, because those are different claims about the work and your reader deserves to know which one is true.
The planning fallacy is not a personal failing
If you have consistently underestimated how long writing takes you, that is not evidence of poor discipline — it is a well-documented, near-universal bias in how people estimate their own future work. When asked to predict how long a task will take, people systematically imagine the best-case path: no interruptions, no getting stuck, no rewriting a section that turned out not to work. Actual writing time nearly always includes some of those things, which is exactly why a number pulled from imagination tends to run short and a number pulled from an actual timed session tends to run closer to the truth. Building your estimate from measurement rather than from memory of how you hope a session will go is the single biggest correction available, and it costs nothing but a few minutes with a timer.
Scaling the same arithmetic to a much longer document
The formula does not change as the document gets longer, only the numbers get bigger enough that the planning stakes rise with them. A dissertation chapter running to 8,000 words at the same 300-words-an-hour pace and 2 hours a day comes out to roughly 26.67 total hours and 14 days — not an unreasonable stretch on its own, but now imagine stacking five or six chapters of similar length back to back without adjusting the plan for revision and research time in between, and it becomes obvious why long-form projects slip by months rather than days when the per-chapter estimate was optimistic from the start. This is precisely why it matters to get the per-thousand-word pace right early, on a small, low-stakes piece of writing, before trusting it to scale up to a much larger commitment where a small error in the hourly rate compounds into weeks of miscalculated schedule.
Building the buffer in
A drafting-time estimate, however carefully calculated, is not the whole project timeline. It does not include the reading and research that has to happen before you can write with confidence, the revision passes that happen after a first draft exists, or the verification work — checking every fact, number, and citation — that any honest writing process requires regardless of what tools were involved. Treat the drafting-hours number as one clearly-labeled segment of a longer schedule, add explicit time for research, revision, and verification around it, and build in a buffer for the day that simply does not go as planned, because in any project long enough to need a schedule, that day always arrives at least once.
For a sense of how the drafting-time estimate sits next to reading time and page count across a range of typical document lengths, the document planning reference lays out a short response, a standard essay, a term paper, a thesis chapter, and a full dissertation side by side, all computed the same way as the worked examples above.