TNS
VOXPOP
Do You Resent AI?
If you’re a developer, do you resent generative AI’s ability to write code?
Yes, because I spent a lot of time learning how to code.
0%
Yes, because I fear that employers will replace me and/or my peers with it.
0%
Yes, because too much investment is going to AI at the expense of other needs.
0%
No, because it makes too many programming mistakes.
0%
No, because it can’t replace what I do.
0%
No, because it is a tool that will help me be more productive.
0%
No, I am a highly evolved being and resent nothing.
0%
I don’t think much about AI.
0%
DevOps / Operations / Platform Engineering

Platform Teams: Start Small to Win Big

With minimum viable and thinnest viable platforms, platform teams can optimally allocate resources, deliver features quickly and maximize business value.
May 16th, 2024 7:39am by
Featued image for: Platform Teams: Start Small to Win Big

It’s every platform engineer’s worst nightmare: Your developers hate the internal developer platform your team has spent months (or years!) building. Platform adoption is low, and key stakeholders are ready to bail on platform engineering. What went wrong? And how can your platform team avoid this unfortunate fate?

Practitioners quickly learn that in platform engineering, bigger isn’t always better. Instead, success comes to platform teams that start small and stay lean. As ”Team Topologies” co-author Manuel Pais puts it, “A good platform is just big enough, but not bigger than that.”

This doesn’t mean skimping on quality but rather using two key concepts: minimum viable product (MVP) and thinnest viable platform (TVP). Using these approaches, platform teams can optimally allocate internal resources, ensuring they deliver the right features quickly and maximize the platform’s unique business value.

Start Small: Using Minimum Viable Product

“The Lean Startup” author Eric Ries defines a minimum viable product (MVP) as “that version of a new product which allows a team to collect the maximum amount of validated learning about customers with the least effort.”

In other words, an MVP is the simplest product version needed to get user feedback. By starting small, platform teams can validate core assumptions and avoid wasting resources on features developers neither want nor need.

Following an MVP approach, platform teams start by talking to their organization’s developers. They conduct user research to identify and understand developers’ most significant challenges. Chances are, developers have already found and implemented different solutions for persistent problems. Platform teams should identify the limitations of existing solutions and design a compelling alternative. The MVP is supposed to be a learning tool, not a finished product, so it should not attempt to support all use cases. Platform teams should build the minimum features needed to address the pain point.

Mia-Platform’s delivery manager Francesca Carta recommends deploying the MVP to a small “pioneer team” for testing. Platform teams can then use the pioneer team’s feedback to iterate upon the solution before releasing it to the broader user base.

Developing an MVP typically takes one to three months. This abbreviated timeline enables platform teams to learn what developers want quickly. Platform teams can adjust based on early feedback and create an effective and compelling platform with minimal investment.

In Carta’s experience, platform teams can also use an MVP approach to implement third-party tooling successfully:

“Don’t start by automating everything; start by solving your team’s main issue. For example, if deployments are a huge pain point, you should focus on making them a self-service capability. Then, in the second phase, you can add the plug-in for the infrastructure configuration.”

Platform teams can thus iterate on the success of more limited implementations of new tools. Fast feedback is still important because migrating developers to third-party solutions can create just as much friction as migrating developers to internal tooling.

Building an MVP isn’t just about getting developers on board. Too many platform initiatives fail because platform teams can’t prove their value to stakeholders fast enough. Carta says this problem most often arises when platform teams try to design for all use cases at the beginning instead of focusing on the MVP. She recommends teams lock in stakeholder investment by starting with a quick but meaningful win, then iterating.

Stay Small: Thinnest Viable Platform

While an MVP approach helps platform teams develop new features, a thinnest viable platform (TVP) approach helps teams optimally allocate internal resources over the full lifespan of the platform. Pais and his coauthor Matthew Skelton define a TVP as:

“the smallest set of APIs, documentation, and tools needed to accelerate the teams developing modern software services and systems.”

An organization’s TVP changes over time. As third-party tools become more competitive, platform teams must choose between improving custom components or buying from vendors. When defining a TVP, Carta says, “You have to choose as a company: What are you going to build inside your company? What are you going to maintain?” With a TVP approach, platform teams allocate internal resources to only what provides unique business value.

Syntasso’s Abby Bangser shared how MOO’s custom observability tooling was retired in favor of vendor-provided options despite concerns about wasting previous investments and fears around vendor lock-in. The benefits were apparent within months after the switch:

“Since MOO was now paying for another company to innovate in tracing, we as internal platform engineers could float higher up the stack and focus on providing more MOO-specific benefits instead of maintaining the now-outdated internal service.”

In his PlatformCon 2022 talk, “The Magic of Platforms,” author Gregor Hohpe discussed platform thinness in terms of floating versus sinking platforms. Imagine all platforms resting on a base platform like Kubernetes. Over time, this base platform gains new features and functionalities. Hohpe likens this to a rising water level. Platforms that don’t adapt to the rising tide will sink; they’ll continue duplicating functionalities offered by the base platform. With a floating platform, the platform team strategically outsources or retires redundant features. This frees up internal resources to focus on innovation, building functionalities beyond what the base platform offers.

According to Hohpe, neither option is inherently better than the other. The key is to proactively decide whether your platform will float or sink, then communicate the plan to stakeholders. When building a floating platform (i.e., maintaining a TVP), platform teams should be transparent about the potential for outsourcing or retiring internally developed features. Otherwise, stakeholders might become upset when months to years of work are discarded in favor of a third-party solution.

Similarly, platform teams should be mindful of how maintaining a TVP impacts developers. In her QCon talk, Kognic’s Jessica Andersson explained that platform teams earn “trust credits” when they ease pain points and follow a product mindset. Conversely, migrations (even if unavoidable) spend trust credits. Platform teams should be intentional about balancing earnings with expenditures.

In short, maintaining a TVP enables platform teams to focus on the tools and features specific to their organization. However, setting accurate expectations is crucial to sustaining stakeholder buy-in and keeping developers happy.

Building Smart With MVPs and TVPs

Today’s platform teams are pioneers in platform engineering. Since there is no one-size-fits-all internal developer platform, teams must figure out how to deliver a solution that fits their organization. Great platforms can be built or bought, but most organizations use a combination of built, bought and open source components. They use a platform-as-a-product approach to understand developers’ needs, get stakeholder buy-in and drive platform adoption. MVP and TVP are product management concepts that ensure platform teams deliver unique business value quickly and at scale.

Platform teams can maximize efficiency and impact by embracing the “start small, stay lean” philosophy. MVPs ensure that development efforts are directed toward features developers truly want and need, avoiding wasted resources on unwanted features. This rapid feedback cycle fosters continuous improvement and user-centric innovation.

TVP is a complementary concept that encourages platform teams to allocate internal resources strategically throughout the platform’s lifetime. By focusing on building what delivers unique business value and leveraging third-party solutions for the rest, platform teams can maintain a TVP that stays ahead of the curve.

Mia-Platform Console is an internal developer platform that allows platform teams to manage and monitor the life cycle of their cloud native software in one place. Platform teams use Mia-Platform to create self-service capabilities quickly and at scale. Learn how Mia-Platform can boost your platform.

Group Created with Sketch.
TNS owner Insight Partners is an investor in: Kubernetes.
TNS DAILY NEWSLETTER Receive a free roundup of the most recent TNS articles in your inbox each day.