Claude may help you start work, not finish it
A 30-day work log found that with Claude took 45% of total work time but produced 38% of finished . without Claude took 55% of total work time and produced 62% of finished . Work with Claude felt faster because it created more drafts, more options, and more starting points.
The problem was the repeated loop of drafting, changing the prompt, reading the answer, editing, again, and reading again. Each loop was small, but together they carried a high attention cost. Without Claude, the work felt slower, but more happened before writing, so the first version was closer to finished.
The practical conclusion was that Claude is strong for starting work and weaker for completing it. The adjusted uses Claude for the first 30% of a task, such as research, outlines, and drafts, then finishes the remaining 70% without AI.
Key points
- Claude used 45% of working time but produced 38% of completed .
- Non-Claude used 55% of working time and produced 62% of completed .
- Claude helped create more drafts, options, and starting points.
- Repeated and editing created an attention cost.
- A better used Claude for the first 30% of a task and finished the last 70% without AI.