"What if Google Builds It?" is No Longer a Bullshit Question
For as long as I can remember, VCs have been faithfully asking founders variations of the question, “What if <incumbent> builds it?”
And for as long as I can remember, it’s been mocked by founders and investors alike as one of the laziest questions a VC could ask.
In some cases — typically when trying to figure out “is this a feature or a product?” — asking a founder about potential competition from incumbents can lead to an enlightening discussion. But more often than not, there are far better ways to explore the competitive landscape. Deep down, asking a founder “what if Google (or Facebook, or Amazon, or Oracle, or…) builds it?” has always been a lazy question.
Until now.
The advent of LLMs, alongside feature-rich offerings from OpenAI and others, has led to a rush of development around AI. Countless products are being built by small teams uncovering use cases that can be solved using AI.
Historically, the development of new, vertical-specific applications would take many months (or years) and teams of 10 or more. This simple fact is why the question “what if <incumbent> builds it?” was generally a lazy one: it presumed that the incumbent would allocate a considerable amount of time and resources to an experiment.
(The fact that one of the most impactful business books of all time focuses on why incumbents can’t do this should have been enough to eliminate this question from the average investor’s lexicon, but I digress…)
Here’s the thing: the calculus for AI is different.
If a team of 2-3 founders can create a full-fledged application in a matter of weeks, an incumbent can absolutely do the same. The cost is lower, the risk is lower, and incumbents around the world are actively doing it.
As a case-in-point, over the past year, countless startups raised funding and came to market with services that used generative AI to create backgrounds and images for e-commerce sites. The idea made sense: instead of spending a huge amount of time and money to stage photographs of every product you sell, take photos of the items in a showroom and let AI do the rest.
In one fell swoop, this entire category of startups was rendered moot. And this isn’t an outlier.
In his Q2 earnings call, Amazon CEO Andy Jassy stated that,
“Every single one” of Amazon’s businesses has “multiple generative AI initiatives going right now,”
While Amazon itself is an outlier when it comes to e-commerce businesses, it isn’t when it comes to tech incumbents. Every incumbent around the world is actively exploring how and where they can incorporate generative AI into their businesses.
Founders need to understand that and appreciate that it is no longer lazy for investors to ask about incumbents in this way. In fact, it’s their duty to. Given the sheer breadth of advancements in generative AI, it is reasonable and rationale to presume that every incumbent in every industry has people exploring what is possible with these new technologies. Some of what they develop will be exceptional. Some of it will have the stain of an old guy desperate to re-live his glory days. A lot of it will be “good enough” for their customers.
Don’t ever forget that this man once scored four touchdowns in a single game
As a founder, you can no longer scoff when an investor asks, “what if <incumbent> builds it?”
More than ever, you need to provide a compelling answer as to why, even if the incumbent does try to build something similar, your version will win. You need to lay out the case as to why customers will pay to use your product over the maybe-not-as-good-but-free offerings that incumbents will inevitably roll out.
Will inevitably roll out.
“What if Google builds it?” is no longer a bullshit question. It's a reflection of the changing competitive landscape. If you're building something compelling, then it shouldn't scare you. But if you can’t (or won’t) credibly answer it, then you may very well be building a feature.