Hey there! Laura Taco here. 🙋‍♀️ As CTO at DX, I’ve had front-row seats to evolving trends in platform engineering and organizational design. Recently, I had the pleasure of hosting a lively discussion with Manuel Pais, co-author of the widely acclaimed “Team Topologies” book. What follows is a comprehensive summary of our talk, designed to bring clarity to some of the hottest (and often most confusing) topics in tech today—everything from DevOps vs. Platform Engineering to tips on starting a killer platform team.

So, whether you're a curious engineer or an organizational leader, grab your coffee ☕ and dive into this detailed guide to mastering platform engineering, team interactions, and beyond.

1. What Are Team Topologies?

The Team Topologies framework, created by Manuel Pais and Matthew Skelton, provides a practical blueprint for building effective teams in modern technology organizations. The framework identifies four key team types, each with specific goals and interactions:

• Stream-Aligned Teams: Deliver customer value by focusing on a single product or flow of work.
• Platform Teams: Build and maintain shared platforms to help streamline-aligned teams perform better.
• Complicated Subsystem Teams: Handle highly specialized aspects of the system (e.g., machine learning algorithms, or other high-complexity tech).
• Enabling Teams: Act as consultants, helping stream-aligned teams adopt new skills or practices.

Each team type works together, like pieces of a puzzle, to reduce cognitive load and minimize friction across development cycles. It's no wonder companies like Amazon and Spotify use variations of this model to thrive.

🔑 Takeaway: By clearly separating responsibilities across team types, Team Topologies reduces messy handoffs, foggy team goals, and unnecessary context-switching.

2. How Is Platform Engineering Different from DevOps?

Question from Byron: “How is Platform Engineering not DevOps?”

This is THE question that sparks endless debates on LinkedIn. So, let’s break it down simply:

• DevOps: At its heart, DevOps is a mindset and culture shift that encourages collaboration between development and operations to reduce friction. Techniques like CI/CD pipelines, infrastructure as code, and automation fall under this umbrella.
• Platform Engineering: This takes the principles of DevOps and manifests them into tangible products and services. Platform teams act as enablers, building internal tooling or ‘paved roads’ (think CI/CD, self-service environments, monitoring tools) that development teams consume to speed up delivery.

In Manuel’s words: “Platform engineering, done well, creates optional, product-like experiences for teams. It reduces friction and focuses on meeting the needs of internal customers (developers).”

🤔 Why the Confusion?

DevOps practices often revealed the need for better tooling, which led to the rise of platform engineering. However, the shift from “culture” to “product” mentality is the key differentiator.

3. What Is “Platform as a Product” Mindset?

Think of platform engineering as internal product development. Instead of solving external customer needs, you’re solving for your most important stakeholder: developers. Here’s what it entails:

• Understand the Customer (Developers): What pains do they face? Is every deployment taking hours instead of minutes? Are test environments unreliable?
• Iterate and Validate: Use feedback loops and metrics to prioritize highly impactful improvements. ⭐
• Optional Adoption: Make platform services optional. If they play hard, users (internal developers) will opt in naturally, signaling your platform’s value.

Manuel emphasizes: “If you mandate platform usage, there’s no incentive to make it great. Make it optional and delightful to use.”

4. How to Measure the Impact of a Platform Team?

This is a huge challenge. Unlike external products, internal platforms don’t directly generate revenue, leading to potential budget cuts. So, how do you prove platform value?

Metrics that Matter:

1. Time Saved (e.g., quicker deployments due to standardized CI/CD pipelines).
2. Onboarding Speed (time taken for new developers to become productive).
3. Developer Satisfaction: Run surveys or track Net Promoter Score (NPS).
4. Retention & Reduced Attrition Costs: Happy devs stick around. Replacing them costs $$$.

Manuel added: “Use success stories! Showcase how an internal team delivered new features faster because of your platform.”

5. How Big Should Your Platform Team Be?

Audience Question: “For 250 developers, what’s a good ratio for platform engineers?”

The safe rule of thumb, according to industry benchmarks, is 15–20% of your engineering org. For ~250 developers, this translates to roughly 40–50 platform engineers. However, size isn’t everything.

The goal is superlinear impact with sublinear growth—aka maximizing results with fewer people.

Manuel shared this gem: “If you’ve got a 5-person platform team supporting 250 devs with 20+ services, either those 5 people are superheroes, or they’re absolutely overwhelmed.”

6. What’s the Best Way to Start a Platform Team?

Manuel suggests: Start small, but start smart.

Here’s the four-part starter kit Laura recommended:

1. Sponsor: You need an executive ally to advocate for funding and support.
2. Champion: A leader who embodies the team’s mission and keeps momentum going.
3. Charter: A clear document outlining goals, scope, and boundaries. What exactly will your platform team do?
4. Allies: Find like-minded colleagues who will collaborate and help make the case.

You’ll also need to manage expectations: “Internal customers (like dev teams) require time to adjust to being treated like external customers,” says Manuel. “It might feel weird when someone asks for their feedback, but it’s key to success.”

💡 Pro Tip: Treat the platform team as an experiment. Show measurable wins early to gain more trust and expand.

7. Where Should Platform Teams Sit in an Organization?

There’s no one-size-fits-all approach, but here are common configurations:

• Centralized Model: The platform team acts as its own entity, often reporting to a CTO or VP of Engineering.
• Domain-Specific: Large organizations create platform teams within business domains, increasing specialization.
• Hybrid Models: Combine a core platform group (for central services like CI/CD) with smaller domain-focused platform teams.

Trade-offs to Consider:

• Centralized teams foster consistency but risk becoming bottlenecks.
• Decentralized teams offer agility but risk duplication of effort.

The MyBookDigest Advantage: Learn Like Never Before  

Strapped for time, but curious about Team Topologies, DevOps, or platform engineering? I recently started using MyBookDigest, and let me tell you—it’s a game changer. Their 15-minute audio summaries condense gems from top books like "Accelerate," "The Manager's Path," and, yes, "Team Topologies!"

With over 500 book summaries optimized for busy professionals (and growing weekly), you’ll never feel out of the loop again. Here’s why I’m hooked:

• 15-Minute Key Ideas: Great for learning during a coffee break. ☕
• Offline Listening: Perfect for commutes or pre-bedtime knowledge binges.
• Massive Library: From leadership to software engineering, they’ve got you covered.

Start exploring MyBookDigest’s library today. Who knows? One short summary might unlock the next big idea for your organization or career.

Final Thoughts

Platform engineering is here to stay, and its role within modern software organizations will only evolve. Whether you’re just getting started or fine-tuning existing processes, remember to put people first—your platform team, internal customers, and executive stakeholders. With tools like Team Topologies to guide you and frameworks like Platform as a Product to inspire you, success is just a matter of experimentation and iteration.

Thanks for reading! Got questions? Drop them below or connect with me on LinkedIn! 🙂

‍