Should Arista Networks prioritize expanding its CloudVision platform features or focus on improving the performance of its existing Ethernet switches?

Introduction

The trade-off question at hand is whether Arista Networks should prioritize expanding its CloudVision platform features or focus on improving the performance of its existing Ethernet switches. This scenario involves balancing innovation in cloud management software with enhancing core hardware capabilities. I'll approach this analysis by examining the business context, user impact, technical considerations, and strategic implications of each option.

Step 1

Clarifying Questions (3 minutes)

• Based on recent market trends, I'm thinking cloud management platforms are becoming increasingly critical. Could you share how CloudVision's current market share compares to competitors like Cisco's DNA Center or Juniper's Mist AI?

Why it matters: Helps assess the competitive landscape and potential growth opportunities. Expected answer: CloudVision has a growing but smaller market share compared to established competitors. Impact on approach: Would influence whether to focus on feature parity or differentiation.

• Considering our revenue model, I assume both CloudVision and Ethernet switches contribute significantly. Can you provide a rough breakdown of revenue between software and hardware?

Why it matters: Helps prioritize investments based on financial impact. Expected answer: Hardware still dominates revenue, but software is growing faster. Impact on approach: Would inform resource allocation between the two areas.

• Looking at user segments, I'm curious about the overlap between CloudVision users and our switch customers. What percentage of our switch customers are currently using CloudVision?

Why it matters: Indicates potential for cross-selling and integrated solutions. Expected answer: Moderate overlap, with room for growth in CloudVision adoption. Impact on approach: Could suggest focusing on integration and migration paths.

• From a technical standpoint, I'm wondering about the current performance bottlenecks in our Ethernet switches. Are there specific areas where we're falling behind competitors?

Why it matters: Identifies critical areas for hardware improvement. Expected answer: Some lag in specific high-performance use cases or emerging standards. Impact on approach: Would guide the focus of potential switch improvements.

• Regarding our development resources, how are our engineering teams currently split between CloudVision and switch development?

Why it matters: Assesses our capacity to execute on either option. Expected answer: More resources currently allocated to switch development. Impact on approach: Would influence the feasibility and timeline of each option.