Building a personal AI workflow
Most people use AI reactively. That’s leaving value on the table.
The typical AI user opens a chat when they’re stuck, gets help with one specific thing, and closes it. This is useful but it’s a fraction of what’s possible.
A workflow approach is different. You build AI into your regular work patterns so it helps you systematically — not just occasionally.
The three-step framework: before, during, after
The simplest way to build an AI workflow is to identify tasks you do regularly and ask: where could AI help before this task, during it, or after it?
Before: Research, preparation, structuring your thinking. Before a difficult conversation, ask AI to help you prepare. Before writing, ask it to help you outline. Before a meeting, ask it to summarise the background.
During: Getting unstuck, first drafts, checking your work. When you’re writing and get stuck, ask AI to suggest how to continue. When you’ve written something, paste it in and ask for feedback.
After: Summarising, communicating outputs, creating follow-up tasks. After a meeting, turn your notes into an action list. After completing a project, ask AI to help you write a summary for stakeholders.
A practical daily workflow example
Here’s what a realistic AI workflow looks like for a knowledge worker:
Morning (10 minutes): paste your task list for the day, ask AI to suggest an order based on priority and energy. Paste any long document you need to read, get a 5-point summary.
Writing tasks (ongoing): start every piece of writing with an AI outline. Write your draft. Paste it back to get editing feedback. Final check before sending.
End of day (5 minutes): paste your notes and ask AI to extract action items and decisions. Use this to set up tomorrow’s task list.
You're not replacing your thinking with AI. You're using AI to handle the parts of work that consume time without requiring your judgment — structuring, formatting, first drafts, summarising. Your thinking goes into the parts that actually need it.
Building habits, not just skills
The bottleneck isn’t knowing how to use AI. It’s remembering to use it when it would help. Build a habit by attaching AI use to an existing trigger.
Pick one recurring task — the same type of task you do every day or every week. Commit to using AI for that specific task for two weeks. Once it’s automatic for that task, add another.
Common workflow mistakes
Trying to use AI for everything at once — you’ll get overwhelmed and revert. Using AI after you’ve already finished thinking — you get little value from asking AI to review a piece you’ve already decided is done. Not saving what works — when you find a prompt that produces great output for a recurring task, save it immediately.
List the five tasks you spend the most time on in a typical week. For each one, ask: could AI help before, during, or after this task? Pick just one of the five — the one where time savings would matter most. Build that single AI habit first. Come back to the others once the first is automatic.