Hello friends, I hope you’re having a fantastic weekend!
In one of my very first posts, in Nov 22, I wrote how the tech industry crunch moment would lead to a lot of focus back on the “basics” of businesses, and how I thought that the following months there would have been a lot less focus on growth and much more on building profitable and sustainable businesses. The tagline of the post was:
“If the last 10 years was about beta, the next 10 years is about Alpha.”
I predicted that growth (Beta) companies would have left the center stage of the industry, and Alpha (profits and financial rigor) would go much more at the center of the winning strategies.
And now, over 2 years from that prediction, I would like to reflect on the best (in my humble opinion) pivot from Beta to Alpha that we have seen: Meta.
Let’s start from the end: a few weeks ago Meta (the holding company of Facebook, Instagram, Whatsapp..) posted its Q4’23 earnings and the stock jumped back to pre-2022 levels. Meta had indeed been heavily punished by the market as the CEO pivoted the company in search of growth in emerging categories (e.g. the heavy investment in metaverse) allegedly losing its focus on the core business.
As I have already said several times I love turnaround stories, and I think it’s now time to give Zuck due recognition as he drove a very intentional strategy turnaround that delivered incredible results.
Meta’s price was slightly below $100 when I wrote the post in Nov’22, and it is today almost 5x that… and while the market has surely recovered as a whole from Q4’22, Meta’s results are driven by the internal performance of the company, that managed to revert the Free Cash Flow trend in the last 5 quarters:
Meta currently has, literaly, half the planet (over 3 Billion people!!!) on its “family of apps” daily, with users’ growth showing no sign of slowdown:
I am sure there’s something we can praise and learn from the story, but let me start from the beginning…
The start of the crisis
The beginning of Meta's crisis can be traced back to a combination of factors, but one of the biggest ones in my opinion is a very technical detail: the introduction of Apple’s AppTrackingTransparency feature in 2022, which allowed users to opt-out of being tracked by apps. This change led to an estimated 62% of Apple's iOS users choosing privacy over targeted advertising. This obviously posed a substantial challenge to Meta's established advertising model because Apple basically broke the “targeting machine” Meta had built, which was the key to delivering effective (and hence lucrative) ads to its users by profiling people.
The situation was further compounded by broader market conditions, including a global slowdown in ad spend and increasing competition from platforms like TikTok. These factors, combined with internal challenges such as leadership changes and negative public perception, contributed to Meta's first-ever revenue drop and ongoing struggles.
In parallel a lot of people started questioning Zuckerberg’s “Metaverse” vision, as the company spent an immense amount of money on its VR ambitions. And Zuckerberg had to go through several hearings in the US Congress that added pressure to the situation.
The year of efficiency
To address the myriad challenges, particularly the significant impact of Apple's privacy changes on its advertising model, Mark Zuckerberg led Meta in a strategic redirection dubbed "The Year of Efficiency" in 2023. This set of actions aimed to streamline the company's operations and reinforce its technological focus. Key elements of this strategy included:
Flatter Organizational Structure: The company flattened its management hierarchy to reduce latency and risk aversion in decision-making and information flow. This involved transitioning many managers to individual contributor roles and optimizing the manager-to-employee ratio for more effective supervision and faster communication.
Leaner Organization: Meta's approach to becoming leaner involved reducing its workforce. This was an acknowledgment of the indirect costs of maintaining lower priority projects and a bloated organizational structure, hindering efficiency.
Prioritizing Technology: Meta reaffirmed its identity as a technology company, aiming to refocus on its technological roots. The strategy involved rebalancing teams to give precedence to technical roles, ensuring that all departments streamlined their operations to support this technological focus. Zuckerberg reminded the company its product-oriented mission: “we don’t build services to make money; we make money to build better services”.
Investing in Efficiency-Enhancing Tools: The company invested in developing AI tools to aid engineers in more efficient coding, automating workloads, and phasing out obsolete processes. This focus on tool development was part of a broader vision for long-term operational effectiveness.
The leaner version of Meta shows very clearly in the “cost leverage” data shown below: the company had 80% of revenue eaten by operating expenses in Q3’22 (when the stock started collapsing), which decreased to 59% in a little more than a year… the diet worked!
Apple’s rules changes and Meta’s response
The advent of Apple's privacy policy changes, particularly the implementation of the AppTrackingTransparency framework, prompted Meta to innovate and adapt its advertising strategies. Faced with the new reality where a significant portion of users opted out of being tracked, Meta recognized the need to evolve its approach to advertising to maintain its relevance and effectiveness in the digital ad space.
And Meta’s response was a testament of the quality of its technical team and it was shaped around its investment in AI solutions for advertising. This move was aimed at overcoming the challenges posed by the loss of access to user tracking data. By leveraging AI, Meta sought to develop new methods of ad targeting and measurement that did not rely on traditional forms of user tracking. This shift not only addressed the immediate concerns raised by Apple’s privacy changes but also positioned Meta at the forefront of a more privacy-conscious advertising model.
Meta's AI-driven approach included the development of sophisticated algorithms capable of delivering personalized advertising experiences while respecting user privacy. This was a delicate balance to achieve, considering the growing global emphasis on data privacy and user consent. The company's investment in AI also extended to improving its analytics capabilities, allowing for more accurate measurement of ad performance in a landscape where direct tracking was increasingly limited.
This move was not just a response to immediate needs but also a strategic step towards future-proofing the company's advertising business against further privacy-centric changes in the industry.
From a WSJ article on Meta’s turnaround:
Meta may have spooked investors with plans to spend billions of dollars on what would be a “metaverse first, not Facebook first” future. But, most of the company’s AI targeting efforts involve optimizing its traditional social-media platforms, especially Facebook. This fits with Meta’s December blog post that the company is, in fact, still devoting 80% of its total investment dollars to improving its legacy business.
Those investments now seem to be bearing fruit. Facebook and Instagram are starting to see a path to recovery in terms of engagement, with a 20% gain in time spent in Reels consumption, in part driven by AI improvements.
Meta's Role in Generative AI and LLMs: The Development of LLama
Another very encouraging signal of Meta’s recovery is the pivotal role it is playing in the GenAI race alongside OpenAI, Google, Amazon, Microsoft...
Meta's Large Language Model Meta AI (LLaMA) is a foundational large language model with several unique features and achievements. It was introduced to support researchers in advancing AI subfields. LLaMA stands out due to its smaller, more efficient design compared to other large models, requiring less computing power for testing new approaches and exploring use cases. It is available in various sizes, ranging from 7 billion to 65 billion parameters, making it versatile for different applications.
A notable aspect of LLaMA is its training on a broad set of unlabeled data across multiple languages, emphasizing inclusivity and broad application scope.
Meta's release of LLaMA under a noncommercial license for research use is also a significant step towards democratizing AI research. This approach facilitates wider access and collaboration in the AI community, especially among those with limited resources.
The Metaverse
Meta's investment in Augmented Reality (AR) and Virtual Reality (VR) technologies has been probably the most debated component of its strategy in the past years.
Historically, Meta made a landmark investment in VR with the acquisition of Oculus VR in 2014 for approximately $2.3 billion. This acquisition marked the beginning of its serious foray into the VR space. Over the years, Meta continued to invest heavily in AR/VR development, spending over $10B annually on R&D in the past 3 years.
Key achievements in Meta's AR/VR journey include the launch of various Oculus VR headsets, which have been well-received in the market. Additionally, in collaboration with Ray-Ban, Meta released the Ray-Ban Stories smart glasses which integrate AR into everyday wearable technology, allowing users to capture photos and videos.
And actually, as the VR race properly started when Apple launched the Vision Pro, Zuckerberg came out with a video where he reviewed the device and had a lot of bullish comments about Oculus superiority vs Apple. He did this by recording a very straight-forward video on a sofa where he had some very interesting takes:
- Quest is 7x less expensive yet way more comfortable
- Quest is better for gaming, Xbox, YouTube
- "I'm surprised how much better Quest is better for majority of things"
Now… dunking on Apple has not been a good idea in the past (remember Ballmer laughing at the iPhone??) but I think Zuck could be right that:
This will eventually be a very large market, with likely a “premium” product (Apple) like the Macbook or the iPhone were in the pc/smartphone category;
Meta could have some advantage over particular forms of VR. He says something very interesting in the video: he thinks that the “smartphone-alternative” will likely be something like sunglasses (i.e. something you check frequently during the day, that you carry with you) whereas the visors will be more of a “laptop-alternative”.
In his own words: “In every generation there’s an open and a closed model, in smartphones closed models won but for instance in PCs the Microsoft open model won… this is still the beginning of this platform”.
The Significance of Corporate Governance: Meta vs. Snapchat
In one of my posts a few months ago I wrote how we often overlook talking about corporate governance, claiming it is a boring topic for lawyers. But I actually advocated that the impact of the governance of corporations is in reality very material. And I think this is true here as well, Meta has a very balanced incentive framework that allows Zuck to take fast decisions but also provides shareholders the opportunity to influence the strategy. And that is for instance in stark contrast vs the largest public competitor of Meta: Snapchat.
Meta's governance structure, while centralized around Zuckerberg who holds a significant portion of voting power, still operates with a degree of shareholder influence.
In contrast, Snapchat's governance is even more centralized. CEO Evan Spiegel and CTO Bobby Murphy collectively control over 95% of the voting shares, granting them almost absolute control over the company's direction. This concentration of power is often seen as a double-edged sword. On one hand, it allows for quick decision-making and a clear, uncompromised vision for the company's future. On the other hand, it can lead to potential risks associated with a lack of diverse perspectives and accountability, which might be mitigated in a more balanced governance structure.
Unfortunately for Snap this governance did not play in their favor, and probably the push toward change that Zuckerberg received from shareholders was a positive contributor to the outcome we see today.
What’s next?
I was thinking about how to close the post today while running, and randomly I pushed play on this very interesting podcast episode of Odd Lots (a Bloomberg podcast I often listen to). The hosts had a very interesting take on the key trends we are going to see in 2024 in consumer goods: they argue that while 2023 was the year where producers increased prices as much as possible (surfing the inflation narrative), 2024 is probably the year when that’s not going to be possible (because consumers are reacting reducing spending) and thus most companies will look for volumes, moving the Gross Margin “fat” they got from price increase to marketing spend to acquire more customers.
And Meta is going to benefit from this increase in ads spend which will (using hosts’ words) “allow tech companies to fund GenAI R&D for free”!
In other words it looks like the restructuring that happened in the last 18 months put Meta in a very good place to benefit from topline growth that will likely come from all their key businesses (ads, VR, chips, AI, etc).
My final point is that Zuckerberg did this master strategy change, and showed his leadership skills, when his “babysitter” (as many labelled her) Sheryl Sandberg was out of the company for a long time so all the credit goes to him and his team. Always impressive and humbling when you think that Zuck is 40 years old!
Now, I know that a lot of people have a lot of criticism toward Social Media and Meta’s apps in particular. And I read a lot about it, and share the critics on many fronts. But I wanted this post to focus not so much on the company or the product, but rather reflect on the strategic decisions that drove such an admirable outcome!
Have a fantastic weekend!
p.s. please do not take any of this as investment advice, I am an horrible investor and stopped doing single stock investments years ago after losing almost every time I tried!