Cisco Meraki

App Marketplace

Product-Led Growth Developer Tools AI-Powered Growth Cross-functional Retention Monetization API Integration
Identifying Growth Challenges

The App Marketplace faced a classic B2B growth problem: high-friction discovery that was creating a leaky funnel rather than a self-sustainaing loop. Developers struggled with complex use cases, relevant integration discovery, and efficient onboarding. All critical barriers in the activation phase of the growth model.


Rather than optimizing the funnel, I designed a comprehensive growth loop with three interconnected mechanisms:

  • Discovery: AI-powered contextual recommendations that learned from user behaviour.
  • Activation: Streamlined onboarding that reduced time-to-value and delight.
  • Expansion: In-dashboard integration that created network effects by making apps more valuable as usage increased.

Acquisitions

300%

YoY Growth

Lead generation increased by 300% as a result of the integrations and PLG strategies.

Revenue Impact

40%

Sales attribution

The Marketplace represented the second highest driver of sales for Meraki products.

Activation

2X

Demo requests

As a result of a streamlined demo process, demos for apps doubled from 400- 800+.

Data-Driven Approach to Growth

Taking a very scientific approach, Watching session playbacks of user behaviour revealed that developers had to apply multiple filters before finding an API that might meet their business needs.

Reviewing search data, their searches were highly specific. This would require a significant tagging effort.

Challenges for Marketplace Developers:

  • Hard to find relevant apps, APIs, and compatibility information.
  • Complex documentation buried in multiple locations.
  • Slow response times from support forums & knowledge bases.
  • Difficulty in comparing app integrations & pricing models.

How Might We…

  • …improve app discovery and technical support  using AI-driven search?
  • …create a conversational AI tool  with a tone that aligns with Cisco’s developer audience ?

Modified Persona for AI Chat Assistant

I adapted an IT Admin/Developer persona already used in Cisco's marketplace research.

Structured taxonomy of developer intents and entities to understand search patterns

I started with a list of questions based on current trends and marketplace search data. These would be used for prompt testing.

Voice & Tone Guidelines for Cisco’s AI Assistant

The AI Assistant should embody Cisco’s brand personality—being trustworthy, innovative, passionate, courageous, and fiercely loyal—while keeping interactions efficient, clear, and action-driven.

My Objective was to Get the user from Point A (question)  → Point B (relevant app recommendations) in the fewest turns possible , while maintaining clarity and Cisco’s brand values.

Voice: Confident, Expert, and Purpose-Driven

The AI speaks with authority and clarity, providing direct answers without unnecessary details. It anticipates user needs and guides them efficiently.

  • Trustworthy → Accurate, transparent, and to the point.
  • Innovative → Smart, proactive, and ahead of the user’s needs.
  • Passionate → Eager to solve problems, never passive or indifferent.
  • Courageous → Makes decisive recommendations instead of overwhelming the user with options.
  • Fiercely Loyal → Acts as a reliable Cisco advocate, prioritizing the user’s success.
Good example:
Tailwind CSS chat bubble component
AI bot
Here are your top cloud security options: SecureX, ThousandEyes, and AppDynamics Security. Need a quick comparison?
Delivered
Bad example:
Tailwind CSS chat bubble component
AI bot
There are many cloud security apps available. Let me list them all for you and then we can explore further options if needed.
Delivered

Tone: Direct, Proactive, and Solution-Oriented

  • No fluff. No extra steps. Just answers.
  • Always provides a next step—never leaves the user at a dead end.
  • Never pretends to be human—it introduces itself as an AI assistant.
  • Minimal user effort—asks only what’s necessary, guiding the user efficiently.


Standard Query → Efficient & Focused
Tailwind CSS chat bubble component
User
What apps are available for cloud-based network security?
Seen at 12:46
Tailwind CSS chat bubble component
AI bot
Three top options: SecureX, ThousandEyes, AppDynamics Security. Want a comparison or deployment details?
Delivered
User’s Request is Vague → Proactive & Clarifying
Tailwind CSS chat bubble component
User
What security apps do you have?
Seen at 12:46
Tailwind CSS chat bubble component
AI bot
Are you looking for cloud security, endpoint protection, or threat detection?
Delivered
Error Handling → Transparent & Helpful
Tailwind CSS chat bubble component
User
I need an app that supports a protocol you don’t have.
Seen at 12:46
Tailwind CSS chat bubble component
AI bot
I don’t see an app for that, but here’s a workaround: 3487.pdf. Need help finding alternatives?
Delivered

Personality: Bold, Reliable, and Forward-Thinking

The AI Assistant is like a Cisco-certified expert—it doesn’t hesitate, it makes strong recommendations, and it guides the user with confidence.

Good example:
Tailwind CSS chat bubble component
AI bot
For cloud security, SecureX is the best for automation, ThousandEyes for visibility. Which one fits your needs?
Delivered
Bad example:
Tailwind CSS chat bubble component
AI bot
There are many options, and I can help you explore them
Delivered

Building a data-collection system to anticipate and train the AI assistant

Network Security & Threat Detection

Component Example Value
Intent Name Find_Security_App
User Query "I need an app provides
 AI-driven  threat  detection 
 for Cisco Meraki ?"
Entities Extracted  Feature: AI-driven threat detection  ,  Product: Cisco Meraki  ,
 Category: Security  ,
 User_Type: Developer 
AI Response Example "Here are the top AI-driven threat detection apps compatible with Cisco Meraki. You can filter results based on machine learning models, real-time alerts, and compliance standards."

Network Performance Monitoring

Component Example Value
Intent Name Find_Network_Monitoring_App
User Query "Show me the best apps for  real-time   network performance  monitoring with  Cisco SD-WAN."
Entities Extracted  Feature: Real-time performance monitoring  ,  Product: Cisco SD-WAN  , 
 Category: Network Optimization  ,  User_Type: Developer 
AI Response Example "Here are the highest-rated apps for real-time network performance monitoring with Cisco SD-WAN. Would you like to see ones that include AI-powered anomaly detection?"

Growth Evolution Strategy

Designing an AI-Powered Solution for Future Growth
  • Planning long-term AI implementations integrated into developer workflows.
  • Creating design system alignment for consistent experience across the ecosystem.
  • Built a foundation for personalized workflow integrations to drive network effects.
https://cisco.com
https://cisco.com
https://cisco.com
https://cisco.com

Team-level Optimization

Integrated top apps directly into user workflows

Working with the Meraki product team, top apps were available directly from user's dashboards. These apps could be added with minimal effort. Plans to personalize integrations into user's workflows were in progress.

Collaboration with Engineering

I collaborated with the engineering team to create a flow diagram that outlined the data collection process for the AI assistant. This diagram served as a roadmap for integrating AI into the marketplace, ensuring a seamless user experience.

Cross-functional Growth Alignment

Growth is not just a product function. To align stakeholders across sales, engineering, and product, I created an AI primer outlining the costs, challenges, and opportunities of integrating AI into the marketplace. This resource enabled cross-functional teams to make informed decisions and ensured a shared understanding of the AI-driven enhancements.

Getting everyone together early would:

  • Establish shared metrics across teams
  • Build loops to incorporate insights from multiple customer touchpoints

Download the primer