This isn't a rant, just an observation.

It's taken me over a decade of building software (not selling it) to truly understand that there are fundamentally two different types of "platform" companies. The difference between them is critical for any developer, customer, or partner to understand.

1. The "True Ecosystem Builder"

This first type of platform company operates from a core belief that they cannot and should not build everything. Their employees and leadership genuinely want to extend the platform's capabilities by fostering a healthy, successful ecosystem of partners.

This requires a difficult but profound cultural shift from a "feature-first" to an "API-first" mentality. Instead of asking, "What new feature can we ship next quarter?" they ask, "What powerful, stable, and well-documented API can we build so our partners can ship 100 new features?"

Prime examples are Salesforce and Atlassian.

Salesforce focuses on being the best core CRM platform, but for crucial adjacent functions, it relies on its AppExchange partners. Look at DocuSign. It built a massive, public company by solving the "last mile" of the sales process: getting the contract signed. Salesforce could have tried to build a "good enough" e-signature tool, but it instead fostered deep integration, making DocuSign the de facto standard for its customers.

Similarly, Atlassian focuses on its core development and collaboration tools (Jira, Confluence). For deep, enterprise-grade functionality like time tracking or resource planning, it relies on its Marketplace. A company like Tempo Software has built an entire, multi-million dollar business by creating a best-in-class time management suite on top of Jira.

In both cases, this creates a powerful "win-win-win" scenario:

• The Platform Wins: It builds an incredibly strong defensive "moat". A customer won't leave Salesforce or Atlassian because it would mean ripping out not just the core platform, but also dozens of other critical, best-of-breed apps (like DocuSign or Tempo) they rely on.
• The Customer Wins: They get the best of all worlds: a stable core platform plus a marketplace of specialized, top-tier solutions that solve their specific problems far better than a one-size-fits-all feature ever could.
• The App Partner Wins: They can focus on building a real, sustainable business and a best-of-breed solution for a niche they deeply understand. They are treated as true partners, and in many cases, become prime acquisition targets for the platform company itself.

These platform companies are rare because this long-term, API-first discipline is difficult to maintain, especially when Wall Street demands new, marketable features every quarter.

I fed Gemini my content and it came up with this. Seems about right!

2. The "Free R&D Lab"

The second, and unfortunately more common in recent years, type of platform company uses its app partners as a free, external research and development department.

These companies operate on a "feature-first" mentality. Their product managers are incentivized to "fill gaps," "own the whole experience," and "complete the product." 

They use their app store to see what ideas get traction (i.e., market demand) and let partners do the hard work of building a solution and testing the nuances with real users.

The process is predictable and painful:

1. An innovative app partner identifies a real customer need the platform has ignored.
2. The partner lives and breathes this problem (typically for years), building a best-of-breed capability.
3. The app becomes successful, proving market demand at zero risk to the platform.
4. The platform company's product team sees this success, marks it as a "gap," and adds it to their own roadmap.
5. The platform releases its own "native" version of the feature. It's often bundled for free, but it's frequently a half-baked solution, built to "check a box" in their sales process.

This process is famously known as being "Sherlocked." The term originated when Apple's "Spotlight" search tool directly copied all the key features of a popular third-party app named "Watson," making it obsolete overnight. We see this constantly with Apple integrating features from password managers and clipboard utilities into iOS, and with Amazon monitoring successful sellers and launching identical, cheaper "Amazon Basics" versions.

In this model, the platform succeeds by outsourcing its R&D risk, and the app partner is left with nothing. The customer's "win" is also often short-lived. They get excited to reduce their spend by dropping the third-party app, only to find the new native feature lacks critical capabilities and receives little to no engineering investment moving forward.

When you decide to build your business on a platform, you are making a multi-year, high-stakes bet. The fundamental question you must ask is: Does this platform view me as a true partner, or as a temporary feature on its roadmap?

Choose wisely. 

Unfortunately, after 10 years, I am beginning to think we didn't.