Why detail matters
When you provide clear context and objectives, the AI can:- Understand your specific situation and constraints
- Draw on relevant expertise for your industry
- Give more actionable, targeted guidance
- Reduce the need for follow-up questions
- Generate more useful documents and recommendations
The anatomy of a good prompt
For best results, include as many of these elements as relevant.1. Clear objective
Start with what you’re trying to accomplish:2. Specific instructions
Tell the AI exactly what you want it to do:3. Contextual background
Share relevant details about your organization and situation:- Organization context — industry, size, regulatory requirements
- Technical environment — existing systems, integration needs
- Team context — stakeholder roles, expertise levels
- Timeline and constraints — budget, deadlines, blockers
4. Examples and format
Provide examples of what good looks like or specify the format you need:5. Constraints and exceptions
Mention special cases or limitations:Putting it together
Here’s an example that combines all the elements:Working with project context
When you open the AI Chat from within a project or initiative, the AI automatically has context about that entity. Use this to your advantage:- From a project: “Draft a status update for this project’s stakeholders”
- From an initiative: “What tasks are still outstanding?”
- General chat: “Compare the progress across all our active projects”
Refining your prompts
You don’t need to write the perfect prompt on the first try. Build on the conversation:- Clarify: “That’s helpful, but I’m specifically looking for…”
- Add context: “Let me provide some additional background…”
- Adjust format: “Can you restructure this as a table?”
- Change audience: “Now adapt it for a non-technical audience”
- Explore alternatives: “What other approaches should we consider?”
Common mistakes to avoid
| Mistake | Better approach |
|---|---|
| ”Help me with AI strategy" | "Help me define three priority AI use cases for our finance department” |
| Omitting industry context | Mention your industry, size, and regulatory requirements |
| Asking one huge question | Break complex questions into steps and build on each response |
| Not specifying the audience | Say who the output is for (“our board”, “my team”, “our IT director”) |
| Ignoring format | Ask for a specific format (“table”, “one-page brief”, “bullet points”) |
Prompt templates
Starting points for common scenarios:Defining a use case
Building a business case
Preparing for a meeting
AI Chat
Start using the AI Chat
Templates
Browse structured templates