We get it. The CEO of the world’s largest company sleeping with Chinese spies is hilarious. But after a week of NVDA reality TV, we feel like there are no Jensen Huang memes that haven’t already been made.
Our interest goes towards how NVDA will perform under a CEO who does not aggressively push product to China (allegedly), and how the web of cross-financing and round-tripping of revenue will unfold under a CEO who might be less inclined to finance NVDA’s own customers.
Squeeze first, ask questions last
Most of our companies have reported earnings by now, and to say that it’s been a volatile earnings season would be an understatement.
This is nothing new. Last quarter we wrote an article called ‘The Big Picture’, where we spoke about investors being too fixated on one data point, or a quarterly hit-or-miss, and missing the fact that a company was rapidly growing its profitability, while trading at a low multiple.
Remember how last quarter, our trade school theme got taken to the slaughterhouse after both UTI and LINC reported a slowdown in student start growth? Ouch..

Both companies faced different problems. LINC cited difficulties with Google Search but promised to offset this by acquiring more students through the high school channel, which goes through sales reps.
As a result, UTI spoke about a more competitive high school channel.
LINC positively surprised markets by returning to 15% student start growth. Juicy! Last quarter’s 0% was an outlier.

Which gets us to today’s theme: why is the market increasingly acting like your bipolar ex-girlfriend, fluctuating from extreme euphoria to extreme pessimism based on a single data point?
Maybe selling off a stock by 60% on 1 weakish quarter is a bit of an overreaction?
To put it in Notorious B.I.G. terms, the market chooses to squeeze first, ask questions last.

We’re all for being long-term, fundamental investors, but we also like to live a stress-free life and to keep a full head of hair.
Do you trade the volatility, or do you take a 3-5 year view and ignore the short-term noise?

Mo’ ads, mo’ problems
Let’s circle back to the trade schools. Last quarter, Phoenix University was the first company to drop the ball on their Google Search problems. LINC followed, and even though student recruitment seemed under control last quarter, their cost-per-sales-lead is growing by about 20% per year.
This quarter, the problems are spreading. Brookdale is complaining that lead-gen websites are no longer getting the Google traffic they used to.
Reddit is the posterboard child this quarter: Google AI Search is reducing click-through traffic to Reddit.com, and their DAU are expected to decline double digits in Q4.
So Google’s AI Search is diverting traffic from normal search.
Or is it? One company’s loss should be another company’s gain. Where are the winners from AI Search?
What if the consumer just keeps weakening and there are fewer customers to divvy up? What if we’re all just blaming Google Search for a weakening consumer?
The other explanation that doesn’t receive sufficient attention. Google’s traditional search is dying. It’s overmonetized. Google was cool because it was a no-frills search engine that sent you to the best website. Today, the first page of results includes AI search, sponsored results, one or two organic results, then more sponsored results.

Figure: Google Search overloading you with sponsored results
Imagine you’re a high school student and you just listened to LINC’s sales rep telling you about their cool welding certificate. You go home, search specifically for LINC, and somehow you get bombarded with ads from all of LINC’s competitors. This isn’t just a lost lead. It is a very high-quality lead with strong historical conversion that’s been lost to the highest bidder.
We’ve heard more executives complain about this issue. But we also encourage you to read the following X post with more anecdotal evidence on why the Google Search experience is deteriorating, including for Google’s paying customers.

Reddit: What’s beef?
What would you do if your Google ads budget no longer gets the right ROI? Well, you divert your budget to other platforms, of course, which brings us back to Reddit, given that they are one of the many beneficiaries of companies diversifying their ad budget.
We never needed every American to spend their toilet break on r/gardening. We needed the ones already there to see an ad for a leaf blower instead of a life insurance policy that had no interest in.
But a more diversified online ad budget is a nice additional tailwind for Reddit.
If you talk to LINC, they will tell you how they are doing exactly that. Spending more on platforms such as Reddit. But beyond ad spend, they are also telling their sales reps to engage with potential customers on Reddit.
Reddit has been sold off on newly disclosed beef with Google. Their deteriorating leads from Google are hurting DAU. Even though we all know that logged-out Google traffic doesn’t stick around. They don’t scroll. They don’t post. They don’t click ads.
They’re almost irrelevant for revenues. More than 80% of revenue comes from users who are either logged in or using the Reddit app.

Of course, losing Google search traffic means losing your top-of-funnel customer acquisition. But there are many ways to attract new users… just think of every time someone sends you a link to an X post (as we just did… did you click?).
And Google traffic helps to re-engage existing customers. But push notifications can do this too.
So, we are now in a situation where Google is sending less traffic to Reddit, and Reddit is responding by moving to logged-in-only, which makes it harder for Google to scrape their data.
Call us skeptics, but given that both companies are currently negotiating a new AI data deal, we think Reddit and Google are stuck in a game of chicken.
We think Reddit has options. What if they renew their data deal with OpenAI? Does any LLM want to risk being left behind over a $100m annual data payment?
Social media assets have been impossible to kill. Reddit looks like it was built 20 years ago and hasn’t been updated since. But it’s got a core user base that is very unlikely to move to a different platform.
Not to mention the strategic value of social media assets. We all made fun of Elon paying $44b for Twitter/X while all advertisers were pulling their ads. And only a few years later, every X visitor has been taught that Europe has been overrun by immigrants and AfD is the only solution. Is the highest value-in-use of social media as platforms that influence elections?
The beef with Google adds a significant amount of uncertainty to Reddit, and the shares should be down – somewhat. But concluding that Reddit has lost and should trade at a low-teens P/E, despite probably growing 30%+ next year, seems premature.
That’s exactly what this market does: squeeze first, ask questions last.
