Previously, I wrote about how adding an AI chatbot to Google Search would mean that Google would be cannibalizing itself – a feat few companies in history have successfully accomplished. In this updated analysis, we'll explore successful product cannibalization examples and revisit Google's journey through a refreshed lens.
Because Google has effectively cannibalized its core search product with AI Overviews (AIOs). The catch? They haven't found a way to monetize them yet, and publishers and brands are feeling the pain.
So, who wins here? Does anyone? Only time will tell.
What Is Product Cannibalization?
Product cannibalization is the process of a company launching a new product that displaces an existing one, often measured in lost sales revenue for the original product. While frequently viewed as a negative outcome, where two products merely trade market share, cannibalization can be a necessary, and even beneficial, strategy for growth.
Familiar Examples of Product Cannibalization
Let's consider a few examples you’re likely already familiar with:
- Hardware: The lifecycle of AI training chips is often less than a year. New architectures and higher processing capabilities quickly make the previous generation obsolete. Chipmakers are forced to cannibalize their own products to stay competitive not only with rivals but also with their own previous breakthroughs.
- Retail: In fashion, new seasonal collections inherently cannibalize old ones as consumer preferences shift. This is driven by trends and seasonality rather than technological leaps.
- Tech/SaaS: New PlayStation consoles significantly replace sales from older ones, especially since they’re backward compatible with games. Another example is the growth of headless content management systems (CMS), which increasingly replace traditional coupled CMS platforms, pushing providers to innovate.
The key difference is the driver: retail cannibalization is often cyclical, while in tech, it's primarily a result of progress.
Full Vs. Partial Cannibalization
Not all cannibalization is created equal. Let's look at a few case studies to understand the difference between full and partial disruption.
Netflix: From DVDs to Ad Tiers
Netflix has made several major pivots, but two stand out: the switch from DVD rentals to streaming and the introduction of an ad-supported subscription tier.
In 2007, as internet speeds improved, Netflix successfully transitioned from mailing DVDs to streaming content directly. This was a full cannibalization driven by a technological step-change. The company's leadership made a strategic choice: rather than be overtaken by new players leveraging streaming, they would lead the charge, even if it meant disrupting their own flourishing DVD rental business by radically changing their pricing model to favor streaming. With hindsight, the decision was less risky than it appeared and proved essential for their survival and growth.
More recently, facing subscriber plateaus and increased competition, Netflix introduced a cheaper, ad-supported plan. This wasn't about replacing their core product but rather a form of partial cannibalization designed to attract a new, price-sensitive customer segment and reduce churn.
Apple: The iPhone Eats the iPod
When the iPhone launched in 2007, it had all the features of the iPod and more. This is a classic case of full product cannibalization. The iPhone was a technological leap forward, and customers were highly incentivized to adopt it. As a result, revenue from the iPod plummeted as the iPhone's sales soared.
Amazon: The Kindle's Uphill Battle
In that same pivotal year of 2007, Amazon launched the Kindle, attempting to cannibalize its core business of selling print books. However, the predicted death of paper books never came to pass. Ebooks stabilized at around 20% of the market. Why? A key reason was pricing. Due to battles with publishers, ebooks were often more expensive than their print counterparts, removing a critical incentive for consumers to switch. The technology was there, but the value proposition wasn't strong enough.
Google & Instagram: Acquire and Copy
Sometimes, cannibalization is managed by acquiring or copying competitors:
- YouTube: Google recognized early on that video was a better answer for many search queries. By acquiring YouTube, they controlled the cannibalization of their traditional web results, slowly integrating video into SERPs to enhance the user experience.
- Instagram: Seeing the threat from Snapchat, Instagram copied the "Stories" feature. While this partially cannibalized engagement with the main feed, the net result was greater overall user activity on the platform.
From these examples, we can define two conditions for successful full cannibalization:
- The new product must be built on a technological step-change.
- Customers must be clearly incentivized (through cost, convenience, or capability) to use it.
Partial cannibalization, on the other hand, is a lower-risk strategy often used to expand market reach or defend against competitors.
Why Google Must Cannibalize Itself
Applying this framework to Google Search vs. AI, we see a complex picture. AI Chatbots represent a technological step-change, and users are often incentivized by the promise of quick, direct answers.
However, my long-held hypothesis has been that Google wouldn't fully launch an "AI Mode" until the monetization model was solved. I predicted searchers would see fewer, but better, ads.
And yet, here we are. Google is actively cannibalizing its product with AIOs. But adoption and engagement tell a complicated story. Usage of the AI Mode tab in the U.S. has hovered at just over 1%, and some reports indicate that over 50% of users who try it don't return. While AIOs are seen by billions, user engagement is reportedly falling.
The problem remains that for the most valuable commercial queries—like "best plumber in Chicago" or "best toaster"—generative AI still often provides lower-quality results than traditional search. These are precisely the queries that are easiest to monetize. Furthermore, the cost of running AI at Google's scale is immense.
Despite these challenges, Google has to compete, not just with other search engines, but with AI-native products like ChatGPT. The cannibalization is happening, but the path forward is far from clear.
A Framework for Your Own Self-Disruption Journey
Did ride-sharing apps kill the taxi business? Or did the taxi industry disrupt itself through complacency? In a time when uncertainty reigns, the best defense against disruption is to go on the offense. Responsible leaders who dare to anticipate disruption and inflict it on their own organization in a smart, controlled way are best positioned for the future. This requires embarking on a structured self-disruption journey.
- Create a Sense of Urgency. Don't wait for a crisis. For self-disruption to gain traction, leaders must agree that the task is urgent and instill that mindset in the organization. You have to act as if you are at DEFCON 1—the highest level of alert—long before you face an existential threat.
- Invest in Your Iconoclasts. Every organization has them: the agitators and instigators who ask tough questions and challenge the status quo. These are the people who ask tomorrow's questions. Instead of sidelining them, empower them. Given authority, iconoclasts can connect ideas across silos and articulate a vision for the future, driving progress over perfection.
- Ground Your Efforts in a Massive Transformative Purpose (MTP). Self-disruption for its own sake is chaotic. A clear, purpose-driven mission that addresses a large-scale challenge provides the necessary anchor and vision. This MTP aligns your organization, attracts the right talent, and ensures that even your most disruptive bets are aimed at a shared, inspiring goal.
- Reframe Cannibalization as a Strategic Option. Shift your mindset from defensive to offensive. Don't see cannibalization as a failure, but as a proactive choice. Look at Netflix, which pivoted from DVDs to streaming long before it was an existential threat, securing a massive long-term advantage. Make an unconventional approach your standard operating procedure.
- Use the Full vs. Partial Lens. Is there a true tech leap and a strong customer incentive? If both are true, aim for full cannibalization. If only one exists, pursue a more controlled, partial approach. This framework helps you manage risk while still pushing boundaries.
- Protect Your Core While You Experiment. Shield your most profitable segments from early disruption. Amazon’s mantra of “disagree and commit” encourages teams to push boundaries, but it doesn't mean recklessly endangering the core business. Test your new product in lower-stakes markets to validate demand before a full-scale rollout.
- Use Cannibalization Defensively. When a rival launches a threat, decide whether to acquire them (like Google/YouTube), copy them (like Instagram/Snapchat), or differentiate if you lack a clear incentive (like Amazon/Kindle).
- Embrace a "Launch, Learn, Iterate" Approach. In today’s market, speed is critical. Waiting for perfection means you’ve launched too late. Adopt an agile, iterative mindset focused on releasing a Minimum Viable Product (MVP) to gather real-world feedback. Track engagement, revenue, and adoption rates, and be ready to adjust. A powerful example is Chegg. Obliterated by Google's AIOs, the homework-help company didn't give up. They launched Solution Scout, a tool that compares answers from AI chatbots with Chegg's human-verified archive. Instead of trying to beat AI, they're hitting it where it hurts: hallucinations and reliability.
By strategically navigating the complexities of product cannibalization and fostering a culture of relentless self-disruption, you can turn this challenge into your greatest strength. This journey is a healthy practice that should be regularly renewed to keep the dynamics created by the process, acknowledge new events, and generate sustainable value in these uncertain times.
