Deep: New API products and strategies explored – substack.com

New products, strategic decisions and the impact of AI on the future of APIs as a product

🔒DoP Deep goes deeper into the concepts and ideas that are covered in the Weekly Briefing to help you learn lessons from the experiences of top tech companies. If you’d like to upgrade to receive these in-depth pieces of analysis you can upgrade below. New reports added every month.

Hi everyone 👋,

Spotify recently decided to cut off developer access from two of its APIs and engineers weren’t particularly happy about it, as you might expect. The APIs in question included access to the “Audio Analysis” API which gives engineers access to a song’s structure and musical content, fearing it could be used to train competitor AI models.

Strava made similar announcements a few weeks back, Yelp caused a mini revolt a few months earlier when it monetized its API and after the major backlash against Reddit in 2023, it later took things a step further and signed lucrative API agreements with companies like OpenAI and Google to give their AI models access to their content.

And we can probably expect to see more of this type of re-evaluation of API offerings in 2025.

Product teams no longer need to only figure out if their APIs are valuable, now they have to evaluate API offerings through an AI strategy lens, too.

In this deep dive, we’re going to explore how product teams are adapting their API strategies. We’ll focus on two dimensions in particular: first, what new API products have been released recently which might be useful for product teams and second, what strategic decisions have been made regarding API products.

Coming up:

• New API products and strategies explored from top tier companies including: Spotify, Palantir, Figma, Stripe, ElevenLabs, Telegram, Uber, Android, YouTube – and what they tell us about the future of APIs

• Why Spotify’s controversial API shutdown reveals a major shift in how tech companies protect their AI moats

• How one of Figma’s new APIs is being used as a subtle but powerful retention driver

• How companies are combining APIs, hardware, and AI to create entirely new product experiences

• The story behind Pinterest’s generative AI API (and the 55% engagement boost it delivered for one company)

• All of the 20+ companies featured in more detail

How this analysis is structured

This analysis looks at over 20 different API products and / or strategic decisions made by top tier companies. Here’s a sample of some of the companies featured:

Each API featured is broken down into the following:

• Company – the company shortlisted for this deep dive. This includes a broad spectrum of B2B SaaS and B2C companies including Spotify, Palantir, Figma, Stripe, Uber, X / Grok, YouTube and more.

• APIs – the API/s featured. This includes recently released, new API products which are noteworthy for product teams as well as more established APIs that have been the subject of recent strategic decisions or changes.

• GraphQL vs REST – does the API featured use GraphQL or REST?

• More details – this is broken down into two subcategories: new API launches includes recently released APIs that product teams might find interesting and strategic decisions which help us to understand how the API product landscape is changing. For example, a new API launch from Stripe which allows agents to take payments is noteworthy for product teams since it may impact the types of payment journeys they build. And new strategic decisions include Strava’s decision to block third party access to APIs or X’s decision to reduce its prices due to efficiency gains which makes it cheaper to serve.

• Links to more info – each API and strategic decision includes further details for more information.

The full list of companies featured is available for paid subscribers at the end of this post.

A deeper look at the companies featured

Now that we’ve set the context, let’s dig a little deeper into some of the companies featured. We’ll explore API strategy and new API releases separately and we’ll start with an exploration of how API product strategy is evolving.

Strategic decisions

Spotify’s recent API changes are a testament to the disruption that AI is currently unleashing across tech; not only is AI impacting how we interact with typical products through UIs, it’s also influencing and shaping API strategy, too.

The APIs that were closed off include the Audio Analysis API and the Audio Features APIs. It’s worth noting that these APIs are still available for existing integrations but no new integrations can be built upon them. That’s a strategic lesson in itself; if you’ve got existing customers but want to deprecate access to an API, you have to carefully manage existing users through that process.

The APIs themselves are actually pretty fascinating – and they’re not APIs that I’d previously spent much time exploring.

The Audio Features API lets users request a representation of a song based on its unique characteristics. In this example, you can see the characteristics of a track that Spotify allows users to generate. This includes things like tempo, danceability, liveness, loudness and other characteristics.

Third party developers were using this API for all kinds of things including one developer who says1 he was using it for measuring “the therapeutic impact of music on patients undergoing life saving treatments”.

In its announcement about the changes, Spotify says the APIs were deprecated due to security reasons but while this is the official line, AI is likely to play a far greater role in influencing this decision.

As one engineer on Spotify’s community forum put it, this shift is likely because of the risk of third parties using the data to train AI models. If Spotify was to continue to make these APIs available to the broader public, it risks losing its competitive moat by allowing third parties to build recommendation models of their own.

Implications for product teams

A strong API product strategy answers 3 key questions:

1. Is this API valuable?

2. How will we monetize it?

3. How will we support the engineers who use it?

But in an AI-first world, there’s a fourth question to consider: what are the unintentional consequences of making this API available to third parties who might use this to compete against us?

But while the story of Spotify’s API changes are important for product teams to understand, they don’t tell us the whole picture.

Other companies are starting to embed AI directly into their API products – and in one instance this led to a significant 55% increase in user engagement. In some cases, AI itself is even being used by product teams to identify and eliminate technical debt in API products which create cost savings that can be passed on to users in the form of cheaper API access which could ultimately lead to an increase in user adoption.

Let’s take a look at those now.