Most people use AI assistants the same way they use a search engine — they type in a quick question, read the answer and move on. That works, but it barely scratches the surface of what’s possible. This guide is for people who want to use AI to meaningfully change how much they can get done in a day.
The core principle is this: AI is a force multiplier on your existing knowledge and judgment. It doesn’t replace thinking — it removes the parts of your work that don’t require thinking, so you have more time and mental energy for the parts that do.
1. Get Better at Prompting
The single biggest factor separating people who get amazing results from AI and people who get mediocre results is prompting. The quality of your output is directly related to the quality of your input.
Give context, not just a request
Instead of: “Write an email about the project delay.”
Try: “I’m a project manager at a software company. We’re delaying a feature launch by two weeks because of unexpected technical complexity. Our main stakeholder is the Head of Marketing who is not technical but is under pressure from the sales team. Write an email that explains the delay clearly, without technical jargon, acknowledges the inconvenience and ends by explaining what we’re doing to prevent similar delays in future.”
Use the “Act as” framing
“Act as a senior copywriter reviewing this landing page. Focus on clarity, call-to-action strength and whether the value proposition is obvious within 5 seconds.” This kind of framing activates a different mode of response — more expert, more focused, more critical.
Iterate, don’t start over
The first output from an AI is rarely the final output. Treat it as a draft. Then say things like: “Make the opening paragraph more direct”, “Cut this by 30%”, “The third point is unclear — rewrite it in simpler language”.
2. Use AI for Email: The Biggest Time Saver
- Drafting responses — Paste in the email you received and ask the AI to draft a response.
- Summarising long threads — Paste a long email chain and ask “What is the current status, what has been agreed and what still needs a decision?”
- Difficult messages — When you need to deliver bad news, push back on a request or handle a sensitive situation, AI can help you find the right tone.
3. Document Work: Summarise, Extract, Transform
- Summarise: “Summarise this 30-page report in 5 bullet points, focusing on the key findings and any recommendations.”
- Extract specific information: “From this contract, extract all clauses related to payment terms, late fees and termination conditions.”
- Transform format: “Take these meeting notes and turn them into a structured action items list with owner, task and deadline columns.”
- Compare documents: “Here are two versions of this policy document. What changed between version 1 and version 2?”
4. Automate Repetitive Writing Tasks
Many knowledge workers produce the same types of content repeatedly — weekly status reports, performance review summaries, project post-mortems, client update emails. Instead of writing each one from scratch, build a template prompt.
5. Research: From Hours to Minutes
- “Give me a clear overview of [topic] for someone who knows nothing about it.”
- “What are the main arguments for and against [position]? Present both sides fairly.”
- “What are the most important things I should know before [decision or task]?”
- “Suggest 5 questions I should be asking about [topic] that I probably haven’t thought of.”
6. Use AI for Creative Thinking and Brainstorming
- “Give me 20 different names for this new product, ranging from literal to metaphorical to completely unexpected.”
- “What are 10 reasons our competitor might have decided to price their product the way they did?”
- “We want to increase customer retention. Give me 15 ideas, including some unconventional approaches.”
7. The Right Tools for the Right Job
- ChatGPT / Claude / Gemini — General-purpose AI for writing, analysis, research and brainstorming.
- Notion AI — If you’re already using Notion, the built-in AI is exceptionally useful.
- Otter.ai or Fireflies — AI meeting transcription and summarisation.
- Superhuman AI — Email client with AI triage and drafting features.
- Gamma — Creates presentation slides from a prompt or document.
A Realistic Expectation
None of this happens overnight. There is a learning curve with prompting, a period of finding which tools fit your specific workflow and a habit-formation phase. Most people who invest a few weeks of genuine effort into learning to work with AI well report saving at least an hour a day.
The single most important step is to start using it for real tasks, not experiments. Pick the most repetitive, time-consuming writing task you do regularly and begin there.