The LENS Index has now been running for a little over three months.
In that time it has done what we built it to do. It has beaten its benchmark. It has held every position through two separate sector-wide selloffs. And it has not triggered a single stop loss.
Here is where things stand.
Since inception on May 22, the LENS Index is up 3.63%. The S&P 500 Total Return is up 2.71% over the same window. That is 92 basis points of alpha, generated with only 48% of capital deployed. More than half the portfolio is still sitting in cash, waiting.
The index hit an all-time high of 106.64 on August 13. Portfolio beta sits at 0.70. Eleven positions, all held, zero exits.
That last number matters more than the return.
Over the past three weeks we watched four separate positions move more than 7% in a single session. One fell 10% in a day. Another rose 10% in a day. Not one of them came close to its hard stop. That is what happens when you size positions correctly and set stops at levels that reflect where a thesis actually breaks, rather than where a chart looks scary.
A new name went onto the LENS Index Watch List this week. It is the most interesting setup we have evaluated in over a month.
The Setup
A company reported quarterly results. Revenue grew 61% year over year. Earnings per share came in at $1.25 against a consensus estimate near $0.97, a beat of roughly 27%. Net income more than doubled. Operating cash flow more than doubled. Free cash flow more than doubled. Gross margin came in above 91%.
The stock fell more than 20% from its pre-earnings level in the sessions that followed.
Not because of the numbers. Because of a single phrase in the shareholder letter about traffic being choppy.
That gap, between a business compounding at 60% and a market pricing in structural decline, is the entire reason this company is now on the LENS Index Watch List.
The name, the trigger price, the thesis, and the one number that would tell us we are wrong are all below.
Reddit: The Market Is Pricing A Traffic Problem Instead Of A Business
Reddit, Inc. (NYSE: RDDT) joins the LENS Index Watch List this week. It has not entered the portfolio yet. We will explain exactly what we need to see first.
Reddit surfaced on two of our four screeners this cycle, the Estimate Revision screen and the Narrative Correction screen. That double hit is rare, and it is what pulled the company into review.
Reddit’s Growth Engine Is Still Intact. Search Traffic Is Still The Test.
Reddit entered the LENS Index Watch List after quarterly revenue grew 61% while the stock fell more than 20% from its pre-results level. The investment debate now centers on whether weakening search referrals represent structural deterioration or whether stronger direct-user retention, advertising growth and product improvements can offset that exposure.
Revenue grew 61%, advertising rose 64% and adjusted EBITDA margin reached 43%. Improving app retention could reduce Reddit’s dependence on external search traffic.
External search remains a genuine discovery dependency. Persistent referral declines or a weak data-licensing outcome could slow user growth and validate market concerns.
Q3 results on October 29 are the key checkpoint. Watch sequential direct daily active user growth and continued improvement in new-app-user retention.
Reddit’s operating performance remains substantially stronger than the post-results share-price reaction implies, but the LENS thesis depends on proving that direct engagement can offset weakening search referrals. Until that evidence arrives, the stock remains a watch-list candidate rather than an active position.
What Actually Happened In Q2
The reported numbers were not close.
Revenue of $805 million, up 61% year over year. That is the eighth consecutive quarter of revenue growth above 60%. Advertising revenue of $762 million, up 64%. Adjusted EBITDA of $343 million at a 43% margin. Net income of $253 million, up from $89 million a year ago. Diluted EPS of $1.25 against $0.45 in the prior-year quarter.
Gross margin was 91.3%, up 50 basis points, even while the company absorbed higher compute costs from expanded AI and machine learning usage.
Capital expenditure for the quarter was $1 million. That is not a typo. It is roughly one tenth of one percent of revenue. This is a business that converts revenue growth into cash with almost nothing in between.
Operating cash flow crossed $1 billion on a trailing twelve month basis for the first time. The company also crossed $1 million of revenue per employee, a target management set at IPO.
Guidance was raised, not cut. Q3 revenue is guided to $860 to $870 million, implying 47% to 49% growth, with adjusted EBITDA margin of 45% at the midpoint.
The company repurchased $235 million of stock during the quarter at an average price around $157.
Why The Stock Fell Anyway
One sentence. Management noted that search referrals were choppy during the quarter and more volatile late in the period.
That is it. That is the sell-off.
The fear underneath it is real enough to take seriously. Reddit has historically received a large volume of logged-out visitors arriving from Google search results. As Google shifts weight from ten blue links toward AI-generated overview answers, the concern is that those referrals shrink permanently. Fewer arriving users, fewer eventual daily users, slower growth, broken model.
Layered on top is an unresolved data licensing negotiation with Google, which added a second reason for investors to price in a bad outcome.
The market took a company growing 61% and repriced it as a company whose traffic engine is about to die.
Why We Think That Is The Wrong Read
Three things stand out from the earnings call.
First, the business does not live where the fear lives. CEO Steve Huffman was direct about this when pushed. Direct app users are worth multiples more than search referral traffic. Reddit’s revenue comes from engaged, repeat, logged-in users, not from drive-by visitors who bounce. The proof is in the numbers already reported. Search referrals declined during the quarter. Revenue still grew 61%. Advertising still grew 64%. Active advertisers still grew more than 70%. If the ad business depended on the traffic the market is worried about, that would not be possible.
Second, the metric that matters is moving in the right direction. New app user retention improved 50% year over year on a relative basis in the quarter. Management called this their single best measure of product quality. It is also the metric that determines whether marketing spend generates a return, and CFO Drew Vollero was explicit that better retention is the unlock for spending more on user acquisition. The company added daily active users through product work in Q2. That gain was offset by the search decline. The underlying engine improved even as the external channel weakened.
Third, the product headroom is unusually large and unusually specific. Huffman described Reddit’s feed models as significantly smaller than state of the art, updating in days rather than hours or minutes, and incorporating roughly 10% of available user activity as signal. The candidate set for recommendations is limited to content from the past week, on a platform holding 26 billion posts and comments, much of it timeless. Parenting advice. Book reviews. Product comparisons. None of it currently eligible for the feed.
That is not a company that has run out of levers. That is a company that has barely pulled any.
The AI Angle Wall Street Is Underweighting
Here is the part that makes this a LENS Index thesis rather than a cheap-stock thesis.
The market has decided AI is the threat to Reddit. We think AI is closer to the opposite.
As the open web fills with generated content, the scarce asset becomes verified human opinion. Huffman framed it as being the antidote to an automated web. AI compresses information into summaries. People increasingly want the thing the summary was made from.
That has a commercial expression already showing up in the numbers. Advertisers are paying for intent, and conversion volume grew over 100% year over year while click volume grew over 30%. Third-party research from TransUnion identified Reddit as the most efficient paid social channel in EMEA for retail advertisers, at roughly 7x higher return on ad spend.
And the data licensing question is not the binary the market is treating it as. Reddit’s content is used for training, for post-training, for grounding, and as a search index. Those are separate commercial relationships with separate economics, and the buyer set is expanding beyond the two names everyone focuses on.
The narrative is that AI takes Reddit’s traffic. The reality may be that AI makes Reddit’s inventory more valuable.
Why It Is On The Watch List & Not In The Portfolio
We are going to be direct about the risk here.
Reddit’s reliance on external search for top-of-funnel discovery is a genuine structural exposure, not a misunderstanding. Management’s own visibility into search is low, and they said so repeatedly. The licensing negotiation could land badly.
The question is not whether the risk exists. It is whether a 20% de-rating already more than compensates for it while the core business compounds at 60%.
We think it probably does. We want one more quarter of evidence before we commit capital.
Watch List entry: RDDT | Tier if added: SPEC 2% | Current price near $156
Trigger to add: Confirmation at Q3 earnings on October 29 that direct daily active users grew sequentially, with new app user retention continuing to improve. A pullback toward the low $140s before that date would also trigger entry on valuation, given management was buying at $157.
Thesis break condition: Revenue growth decelerating below 35% year over year, or two consecutive quarters of declining direct daily active users in the United States. Either would mean the product work is not offsetting the search decline, and the bear case is correct.
Search Risk Is Real; Direct Engagement Now Determines The Thesis.
The LENS Index holds eleven positions with 48% of capital deployed and 52% in reserve.
We are not in a hurry. The cash is there for a reason.
Every position in the LENS Index got there because the market was pricing a story that the numbers did not support. This time the story may be about the last place on the internet where people still talk to each other.
