Discover more from Memos from Hawk Hill Ventures
🏗 The Power of Product Platforms
Good startups own the rails they run on. Great startups lease them to others.
Every developer tools startup wants to build a platform. I’d like to propose that they should build two platforms.
Infrastructure Platform: Delivers value to the customer via economy of scale.
Product Platform: Delivers value to the customer via software abstraction.
Products that live within the Infrastructure platform are the fundamental primitives. They’re generally expensive enough to run such that usage-based billing has better economics than DIY. They don’t have to be isolated, and often times work best together.
Products within the Product Platform, on the other hand, might not be too expensive to operate but have a high implementation cost. This might mean cobbling together multiple vendors for a supported workflow, or doing deep customization. These products are often much closer to the value creation moment for the customer.
In both platform models, the stickiest products are also technically complex! Complex enough that the average developer wouldn’t want to maintain it on their own. This is valuable to the customer! I’ve seen it first-hand with Stream Club: “I wanted to add live streaming to my product but I gave up.”
Startups gain a competitive advantage when they make the incremental cost to buy a new product cheaper than getting it from someone else. ie: If you’re already using Twilio for SMS and voice, why not use it for your email too?
I wish the cost/benefit analysis was that simple. Often times, competitors will specialize and build one feature really really well. And reducing friction to adoption won’t be enough.
The key to countering a fragmented market with many specialized competitors is moving up the stack (“going vertical”) and offering integrated solutions that combine several products to solve a specific workflow.
The title of this section is called 1+1=3 from the creative technique used to find associations. There are permutations of your products that are being used today that could be bundled together to solve common workflows.
Stream Club wants to build a better live streaming software for the browser. They buy GPUs from Amazon, WebRTC from Daily, and video transcoding from Mux. In this scenario, each of these vendors are a small part of the product, and potentially swappable for another provider (a lack of stickiness).
But, Stream Club isn’t happy with their current solution. They’re paying three different bills and dedicating a whole cofounder to running this zombie system.
Mux can create customer value here by offering integrated Mux products that solve the common workflow. By giving Stream Club one bill instead of three, and a purpose-built product for this common video workflow, they’ve won a customer who is now deeply integrated into the Mux stack.
And thanks to the power of software, they can now offer this integrated workflow to all of their current and future customers.
It’s a common pattern for technology companies with a programmable offering to offer products that simplify the common workflows of their customers (see appendix below). This is often called verticalization (or cannibalization).
With the Infrastructure Platform/Product Platform models, it can be easier to think about where your product fits into that picture, and how you can combine it with other products to create valuable combinations.
If you’re wondering what the first offering of the Product Platform should be, it may look like this:
Combines several Infrastructure Platform products
Is complex to build on it’s own
Solves a desirable workflow with no strong market competitors
Appendix - Simplified Examples
Product: Cloudflare Pages (Hosting)
Infrastructure: File Storage
Product: Dropbox Paper (File collaboration)
Product: Contract Lifecycle Management