As the incumbent in Cloud, Amazon seems unbeatable.
For all their bluster about disrupting decades old industries, the industry that tech disrupts the most is itself. In the last 3 decades it has become typical to watch companies rise, fall, and some times even rise again long after marked for dead.
Once a company holds an incumbent position long enough in tech, as Amazon has held their position in Cloud for nearly a decade, there’s an expectation, even sometimes an assumption, that their days are numbered. It’s only a matter of time before an upstart disrupts their position.
In many ways, Amazon exhibits traditional incumbent behavior. They often have the most complicated APIs and offerings among their competitors. Their dashboard looks like a Java Applet from 2000 re-written for “Ajax” in 2006. Their pricing model for storage requires a dual engineering and accounting degree to estimate. At first glance they seem ripe for disruption, resting on their laurels as the market leader.
And yet, their competitors make only meager gains with only Azure showing the kind of growth that might challenge their position.
I believe this is due to two factors unique to Amazon and the current software developer ecosystem.
The Open Source Ecosystem is too evolved.
The differences in ease of use between Amazon and its competitors are mostly invisible to Amazon’s customers.
Product developers live in stacks above the layer that Amazon and its competitors offer. Developers live in languages and frameworks with highly evolved ecosystems that do most of the work for them.
The difference in API design between Amazon and GCP for a given service is barely noticeable once the open source ecosystem has developed plugins and libraries that integrate into developer tools and frameworks. Even that awful dashboard is almost never viewed if you are running one of the many open source orchestration suites to manage your services.
This is often overlooked, but the open source ecosystem owns the majority of the developer experience and cloud services are just a necessary commodity.
Amazon won’t be beat on price.
Classic incumbent disruption goes something like this.
- An early to market company gains a leading position which they maintain as the market grows.
- This market leader starts specializing offerings for the most profitable market segments. This allows them to increase their profit at a greater rate than the rest of the market is even growing.
- When an upstart company starts undercutting their pricing they choose not to react with price cuts themselves because the losses from their up-market customers are greater than down-market customers are worth.
- Down-market technologies, companies, and pricing models eventually consume the entire market and the incumbent is too late to adjust.
This presents a unique challenge when facing Amazon, because they simply will not be beat on price.
Amazon is a retailer, price sensitivity is in their DNA. In the past, when new companies like Digital Ocean have gotten large enough to show up on their radar they’ve reacted with deep price cuts.
This sounds obvious but it’s actually quite exceptional. Amazon has chosen, one multiple occasions, to lose millions of dollars in revenue from existing customers in order to maintain competitive pricing. They’ve chosen position over profit, and that is quite rare.
Amazon’s latest offerings around Lambda are another example of their willingness to eat their own lunch.
Nobody has more to lose than Amazon in adjusting compute pricing to milliseconds instead of hours. The base cost of compute for a startup has gone from hundreds to sub-dollar recurring monthly billing. This model wasn’t just adopted by Amazon, they were first to market!
Adding this all up, Amazon is terrifying to compete with.