MeshMesh

Prompting Tips

Master AI communication with tips, tricks, and best practices

Master AI Communication

Learn how to communicate effectively with MeshMesh AI to get the best results. These techniques help you unlock the full potential of natural language interaction.

MeshMesh understands context and learns from conversation. Clarify, ask follow-ups, or request modifications to get exactly what you need.

Core Principles

Be Specific

Include details like date ranges, segment criteria, and data fields

❌ "Show me email performance"

✅ "Show me open rates for emails sent to VIP customers in the last 30 days"

Use Context

Reference previous conversations and build on earlier requests

❌ "Create another report"

✅ "Create a similar report but for Premium segment instead of VIP"

Describe Goals

Explain what you're trying to achieve

❌ "Show me subscriber data"

✅ "Identify customers at risk of churning - subscribers who were active but haven't engaged recently"

Prompting by Use Case

Data Queries

Start with the outcome

Find customers who might be interested in our new product based on past purchases

Specify time ranges

Show email engagement metrics for last quarter, broken down by month

Include comparisons

Compare newsletter vs promotional email click-through rates

Content Creation

Describe the audience

Create a welcome email for new B2B subscribers who signed up for product demo

Specify tone and style

Write a casual, friendly email to celebrate customer anniversaries

Include personalization

Generate email with product recommendations based on browsing history and loyalty tier

Analysis & Insights

Ask for patterns

What patterns do you see in customer behavior before purchase?

Request actionable insights

Analyze email performance and suggest 3 specific improvements

Compare segments

How do mobile users engage differently with emails compared to desktop users?

Prompting Workflow

Advanced Techniques

Break complex requests into logical steps

First, identify customers who haven't purchased in 90 days.
Then, look at their email engagement during that period.
Finally, create a re-engagement campaign targeting the most promising ones.

Benefits: More accurate results, better understanding, easier to modify

Start broad, then narrow down based on results

Show me customer segments →
Focus on high-value segment →
Show their email preferences →
Create targeted content for this group

Benefits: Discover unexpected insights, build understanding gradually

Provide background about your business goals

We're launching a premium tier next month.
Help me identify existing customers who would be good candidates
based on engagement and purchase history.

Benefits: More relevant suggestions, strategic insights, goal-aligned outputs

Prompt Templates

Performance Analysis

Analyze [metric] for [audience] during [time period]
and compare to [baseline]. What insights can you share?

Example:

Analyze open rates for VIP customers during Black Friday week
and compare to their average performance. What insights can you share?

Content Generation

Create a [content type] for [audience] that [main goal]
with a [tone] tone, including [specific elements].

Example:

Create a welcome email for new subscribers that encourages first purchase
with a friendly tone, including personalized product recommendations.

Data Exploration

I want to understand [business question].
Show me [relevant data] and help me identify [what to look for].

Example:

I want to understand why email engagement dropped last month.
Show me performance metrics by segment and help me identify potential causes.

Tips and Tricks

DO:

  • ✅ Be specific with dates and criteria
  • ✅ Explain your business goal
  • ✅ Build on previous conversations
  • ✅ Ask for clarification when needed

DON'T:

  • ❌ Use overly technical language
  • ❌ Make requests too broad or vague
  • ❌ Assume the AI knows your context
  • ❌ Mix multiple unrelated requests

PRO TIPS:

  • Start simple, then add complexity
  • Use follow-up questions
  • Ask for examples or samples
  • Request step-by-step guidance

Ready to Practice?

Head to the chat interface and try these techniques. MeshMesh learns from context, so refine your requests and ask follow-ups.