Google: Shaping the Future With History

By: Aaru Chandra & Tarik Irshad

The Ivey Business Review is a student publication conceived, designed and managed by Honors Business Administration students at the Ivey Business School.


The GenAI Landscape and Google’s Theatrical Attempts

As a market leader in the global technology sector, Google’s meteoric rise to the top traces back to its pioneering work as the internet’s premier search engine. With a marked diversification strategy that has seen its revenue streams branch into the then-emerging fields of cloud (Google Cloud Platform) and streaming video (YouTube), Google has always sought to leave its mark as a trailblazer in growing technology fields. In 2024, it made its foray into generative artificial intelligence (GenAI) – but not before key competitor OpenAI beat it to the market with a stronger product.

OpenAI’s release of its GenAI chatbot, ChatGPT, in November 2022, has taken the world by storm. ChatGPT reached 100 million users in only two months, far quicker than any other platform in history. Netflix, for instance, required 18 years to accomplish the same feat. Powered by large language models (LLMs), ChatGPT demonstrated the capability to understand and analyze data while autonomously generating new content such as art. Moreover, its ability to simulate human-like conversations with users while directly answering all their queries proved exceedingly useful for students, professionals, and enterprises alike. In addition, the reveal of Sora, OpenAI’s new text-to-video GenAI model, has only furthered OpenAI’s – and Microsoft’s, a key competitor to Google, by proxy – dominance in the space. 

Vying for a share of the highly attractive GenAI market that is estimated to be worth $67 billion and projected to grow at a 42 percent compounded annual growth rate (CAGR) over the next ten years, many tech giants including X and Meta have released LLM-powered GenAI models of their own in an attempt to emulate OpenAI’s enviable success. However, newly released models have struggled to gain traction and overcome OpenAI’s network effects, stemming from its first-mover advantage. 

Google, not wanting to miss out on the opportunity, released its AI chatbot Bard in February 2023. Rushing the product’s release to catch up to OpenAI created a disappointing product that made several factual errors in its first demo, causing Google’s share price to fall 9 percent in a single day and further cementing OpenAI’s position as the unreachable market leader. After rebranding to Gemini and releasing image generation capabilities in February 2024, Google was forced to pause this feature less than three weeks later after controversies surrounding the creation of historically inaccurate pictures. Google’s failures have largely been attributed to the slow and tedious decision-making processes of a blue-chip incumbent. Nevertheless, as Google has incrementally improved Gemini, the chatbot has gained 147 million MAUs, compared to ChatGPT’s industry-leading 181 million MAUs. 

Despite explosive growth, the GenAI industry is facing several issues. As industry players persistently strive to enhance their LLMs by exploring new avenues for data acquisition to train these models, they sometimes encroach upon copyright regulations by training without proper permissions from publishers. Recently, The New York Times escalated these concerns by initiating a lawsuit against Microsoft and OpenAI, among others, alleging the unauthorized use of “millions” of articles from The New York Times. The lawsuit contends that this unauthorized usage has contributed to the augmentation of ChatGPT’s capabilities, positioning it as a competitor to the newspaper as a trusted source of information. 

Looking forward, several experts have suggested that the performance of LLMs will converge as open-source models continue to accelerate the commoditization of their underlying algorithms. As a result, similar to how commoditized streaming platforms are currently competing on proprietary content production, the true, long-term moat for GenAI platforms is the ability to train using proprietary data. Therefore, if Google aims to create a competitive product for OpenAI, it must look for sources of advantage in the data it uses. One potential area to collect proprietary data is Reddit, which hosts huge amounts of text data and information. Presently, Google has established a $60 million partnership with Reddit to use its data. However, Google should treat the partnership as a trial run utilizing Reddit’s data and should look to fully acquire Reddit and permanently secure the data that Reddit has for the long term.

The Front Page of the Internet

Founded in 2005, Reddit has grown to become one of the most influential social networking sites globally, boasting more than 100,000 communities and 268 million weekly active unique (WAUqs). The forum-style platform enables users, or Redditors, to “dive into anything” through a wide variety of user-created communities, or subreddits, where users can share, comment on, and vote on content. The company primarily generates revenue from two streams: advertising and premium membership subscriptions. Reddit continues to rapidly grow, as it grew revenues by 21 percent to $804 million in 2023, amidst stark competition from social media sites such as TikTok. 

Throughout its history, Reddit has raised a total of $1.3 billion through several rounds of venture capital funding, led by numerous high-profile investors including Marc Andreeson and Peter Thiel. Most recently, in August 2021, Reddit raised a $700 million Series F funding round led by Fidelity Investments, valuing the company at $10 billion ahead of confidentially filing for an IPO in December 2021, seeking a $15 billion valuation. 

However, these high valuations are partially a byproduct of the U.S. government’s low-interest rate policy during the COVID pandemic to stimulate the economy. In the following years, interest rates have risen rapidly causing the cost of capital to increase significantly. Thus, investors have shifted from favouring growth-at-all-costs to cash-flow-positive companies with a proven value proposition. 

Accordingly, Reddit has made strides to achieve profitability. In July 2023, the platform began charging developers for access to its previously free application programming interface (API), finally monetizing its invaluable database. While this move followed X’s (formerly known as Twitter) decision to charge developers for the use of its own API, the rate of ~$0.24 per 1,000 API calls meant several third-party applications such as Apollo for Reddit would be forced to shut down. Hence, these efforts were met with heavy protests from the company’s user base as thousands of subreddits went private in solidarity with these third-party applications. Nonetheless, Reddit remains unprofitable, recording a loss of $69 million in adjusted EBITDA in 2023. 

In February 2024, Reddit officially filed its IPO prospectus with the Securities & Exchange Commission (SEC), seeking to go public by the end of March 2024 at a valuation of approximately $6.5 billion ($31-$34/share), representing a nearly 66 percent devaluation from its initial IPO plans in 2021. Soon after, Reddit announced it struck a $60 million per year training deal with Google, providing the company with access to Reddit’s data API. While this marks a step in the right direction for both Reddit and Google, the non-exclusive nature of the deal paired with the potential for the partnership to be amended in the future means Google is risking losing access to Reddit’s data in the long-term.

In March 2024, the company completed its IPO under the ticker RDDT at $34 per share. One month later, RDDT has traded up 24 percent at around $42 per share, representing a ~$6.9 billion market capitalization.

Google and Reddit: Two Challenges, One Acquisition

Google should position itself for a strategic acquisition of Reddit. As part of the IPO process Reddit completed in March 2023, the company prepared several forms such as the S-1 that contain company-specific information such as a business prospectus and consolidated financial statements. This will enable a relatively swift due diligence process that will enable both Reddit and Google to capitalize on the myriad of potential synergies immediately in this rapidly evolving environment of GenAI.

Google will be able to significantly improve Reddit with its expertise and resources. At present, Reddit constantly receives user complaints regarding the social networking site’s poor user interface (UI) and ineffectual search engine. Given Google’s leading expertise in search engine optimization and vast experience designing UIs, the company will undoubtedly be able to improve Reddit’s overall user experience. 

Moreover, Reddit has historically been unable to adequately monetize its existing user base due to its anonymity. Compared to Facebook’s $68.44 average revenue per user (ARPU) in the U.S. and Canada, Reddit’s ARPU of $5.51 in the U.S. is abysmal. This effect can be attributed to Facebook’s ability to identify users and provide targeted advertisements. Being acquired by Google will enable the platform to profile users through Google’s abundance of data and AI, providing the necessary capabilities to significantly boost ARPU in line with that of competitors. 

In addition, since a key component of Google’s acquisition rationale is Reddit’s rich sets of user-generated data, the acquisition will essentially allow Reddit to finally monetize its historically underutilized, expansive database. Nonetheless, Google’s sheer scale and strong cash flow position will reduce pressures on Reddit to achieve profitability. Instead, Reddit will be able to steadily improve its profitability metrics through gradual enhancements in the platform’s business model that will more effectively align with user demands. Although an acquisition by Google would require a sizeable premium for Reddit’s valuable assets, the cost would be manageable thanks to Google’s $111 billion cash reserves as of December 2023.

For Google, on the other hand, the benefits of acquiring Reddit are evident. Reddit’s years of operating as a forum-style channel for users to ask and answer questions on millions of topics has produced an optimal database for LLM chatbot training given the similarity of Reddit’s unstructured, freeform content with content expected to be generated by Gemini. Training Gemini on Reddit’s rich, proprietary database that no other LLM platform will be able to take advantage of will provide a sustainable competitive advantage for the platform, manifesting in the form of more accurate responses. Subsequently, a superior product will attract new users and enable Google to establish dominance in the highly attractive, rapidly expanding, $67 billion GenAI market.

Furthermore, the transition from search to GenAI threatens the fundamental existence of Google’s business model. The company currently holds 92 percent of the $167 billion industry, which generates nearly 57 percent of the company’s revenues. Failing to capture the GenAI market and losing its monopoly position, or even a couple of percentage points, in such a valuable industry would cause investors to reevaluate the merit of Google’s historically untouchable position to reflect a more pessimistic view of the business and sizeable multiple compression.

Why it Would Work?

Google’s $1.65 Billion Acquisition of YouTube

Google’s 2006 acquisition of YouTube for what executives claimed to be an approximately 65 percent premium is hailed as one of the most successful acquisitions of all time. At the time, despite boasting 20 million MAUs, YouTube was facing an existential crisis stemming from several lawsuits over copyrighted content. Today, the social media platform boasts 2.7 billion MAUs. YouTube’s success story demonstrates Google’s willingness to pay sizable premiums for businesses in which they have conviction, willingness to incur short-term expenses such as litigation, and ability to utilize its vast resources to considerably compound the user base and value of a platform over the long-term.

Facebook’s (now Meta) $1 Billion Acquisition of Instagram

In 2012, Facebook acquired Instagram for around a 100 percent premium to its valuation. Since the acquisition, Instagram has rapidly expanded its user base from 20 million MAUs to 2.4 billion MAUs, further demonstrating the potential of a social networking platform with vast capabilities. Moreover, Meta has recently used Instagram’s data to train its GenAI personal assistant, Meta AI, to add features such as the “Imagine with Meta AI” image generator that was trained on 1.1 billion Instagram and Facebook photos. In Reddit’s case, Google can take advantage of its resources to capitalize on Reddit’s data to add features that improve the social networking platform’s historically underwhelming user interface.

Generating a Path Forward

Given the synergies for both parties, Google should seek to acquire Reddit at a sizeable premium to its present valuation of $6.9 billion to cultivate a mutually beneficial relationship. With Reddit’s rich database, Google will enable itself to flourish into the ultimate leader in the GenAI industry by establishing an unmatched data advantage that will be the source of long-term differentiation and market share dominance over competitors such as OpenAI. Simultaneously, with the help of Google’s vast resources, Reddit will be able to improve its profitability and user experience without compromising its existing user base. Finally, acquiring Reddit will enable Google to satiate its long-standing desire as a profit-seeking, data-driven company to own and operate a social media company that has existed ever since Google+ launched in 2011 but ceased operations in 2019.

Previous
Previous

Hershey: Healthy Growth

Next
Next

Six Flags: A Thrilling Expansion