Effective prompting

Getting the most from Kowalah’s AI Chat depends on how effectively you communicate your situation and what you need. This guide covers the key elements of a good prompt and common patterns that work well.

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:

I need to develop evaluation criteria to compare three 
AI approaches for our customer service use case.

2. Specific instructions

Tell the AI exactly what you want it to do:

Help me create a structured comparison document that 
I can share with our leadership team. Include a scoring 
matrix and a recommendation section.

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

We're a financial services company with 2,000 employees 
handling sensitive customer data. Our team has limited AI 
expertise, and we need to present a recommendation to our 
executive team in three weeks.

4. Examples and format

Provide examples of what good looks like or specify the format you need:

We previously created a vendor evaluation for our CRM 
that used a 1-5 scoring system with clear criteria 
definitions. I'd like a similar format.

5. Constraints and exceptions

Mention special cases or limitations:

Data residency is critical — we operate in the EU and 
must comply with GDPR. Our security team has flagged 
concerns about data handling.

Putting it together

Here’s an example that combines all the elements:

Objective: I need a proof of concept plan for integrating 
AI into our customer support workflow.

Instructions: Create a structured plan covering the testing 
approach, success criteria, and timeline. Format it as a 
one-page brief suitable for our operations director.

Context: I'm the IT Director at a healthcare company with 
1,500 employees. We currently use ServiceNow for ticketing 
and have 20 support agents. Our budget is approximately 
$50,000 for the pilot phase.

Constraints: We need to consider HIPAA compliance 
throughout, and we cannot use actual patient data without 
additional security measures.

Working with project context

When you open the AI Chat from within a project or deliverable, 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 a deliverable: “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?”

If you find typing long prompts tedious, try a voice-to-text tool like WisprFlow to speak your prompts instead. It makes it much easier to provide detailed context.

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

Help me define an AI use case for [department]. We're a 
[industry] company with [size] employees. The main pain 
point is [specific problem]. I need a structured problem 
statement that quantifies the impact and identifies who 
would benefit.

Building a business case

I need to build a business case for [AI initiative]. 
Our organization is [context]. The key stakeholders 
include [roles]. Help me create a document that covers 
the problem, proposed solution, costs, expected benefits, 
and timeline.

Preparing for a meeting

I have a meeting with [who] about [topic] tomorrow. 
Help me prepare an agenda, key talking points, and 
questions I should ask. The meeting should result in 
[desired outcome].

AI Chat

Start using the AI Chat

Templates

Browse structured templates

Overview

Prompts by function

Guides

Why detail matters

The anatomy of a good prompt

1. Clear objective

2. Specific instructions

3. Contextual background

4. Examples and format

5. Constraints and exceptions

Putting it together

Working with project context

Refining your prompts

Common mistakes to avoid

Prompt templates

Defining a use case

Building a business case

Preparing for a meeting

AI Chat

Templates

​Why detail matters

​The anatomy of a good prompt

​1. Clear objective

​2. Specific instructions

​3. Contextual background

​4. Examples and format

​5. Constraints and exceptions

​Putting it together

​Working with project context

​Refining your prompts

​Common mistakes to avoid

​Prompt templates

​Defining a use case

​Building a business case

​Preparing for a meeting

AI Chat

Templates

Why detail matters

The anatomy of a good prompt

1. Clear objective

2. Specific instructions

3. Contextual background

4. Examples and format

5. Constraints and exceptions

Putting it together

Working with project context

Refining your prompts

Common mistakes to avoid

Prompt templates

Defining a use case

Building a business case

Preparing for a meeting