The Zhitong Finance App learned that Reddit (RDDT.US) held an FY26Q2 earnings conference call. The company's repurchases this quarter became a more substantial capital tool. A total of about US$235 million of shares and about 1.5 million shares were repurchased, with an average transaction price of about US$157 million; the current authorized amount is still about US$760 million. The three major priorities for the company's capital allocation are: business investment, opportunistic mergers and acquisitions, and share repurchases, while maintaining a high level of profitability. By the end of the quarter, cash, cash equivalents and marketable securities were US$2.8 billion, an increase of more than US$700 million over the previous year, with a slight increase over the previous year.
The company issued performance guidelines for the third quarter of 2026, with estimated revenue of 860 million to 870 million US dollars, up 47% to 49% year on year, median value of about 48%; adjusted EBITDA is 385 million to 395 million US dollars, up 63% to 67% year on year, and the median value corresponds to an adjusted EBITDA profit margin of 45%. The guidelines have been included in the current investment plan.
At the same time, the company lowered the annual equity incentive fee (SBC) ratio guideline, from the previous high ten-digit percentage to a low middle ten-digit percentage, mainly due to SBC accounting for less than 13% of revenue from the beginning of the year, and a drop of more than 1,000 basis points compared to the first half of last year. Q3 SBC and related taxes are expected to rise to US$140 million to US$155 million, mainly due to the ownership of old batches granted during periods of low share prices and re-grants at higher share prices. The full-year dilution is expected to fall at the lower end of the target range of 1% to 3%, not taking into account further dilution offsets from repurchases and potential increases in mergers and acquisitions.
In terms of key financial indicators, Q2 revenue was US$805 million, up 61% year on year, and the growth rate was over 60% for the eighth consecutive quarter; the adjusted EBITDA profit margin was 43%, and the sum of the “revenue growth+adjusted EBITDA profit margin” for the fifth consecutive quarter exceeded 100%. The gross profit margin was 91.3%, an increase of 50 basis points over the previous year. Adjusted total costs were US$462 million, up 39% year on year, with operating costs of US$70 million, up 53% year on year, and adjusted operating expenses of US$392 million, accounting for 49% of revenue (57% of the same period last year). Net profit of $253 million, basic earnings per share of $1.31, and diluted earnings of $1.25 per share ($0.48 and $0.45 in the same period last year). Operating cash flow doubled year-on-year to US$262 million, and LTM's operating cash flow surpassed US$1 billion for the first time. The fully diluted share capital was 207 million shares, up 0.3% month-on-month, and the net caliber increased 0.2% year-on-year; the net dilution was 0.3% month-on-month, excluding the impact of repurchases, and the gross spread was 1% thinner. Per capita income (LTM) surpassed $1 million for the first time, an important yardstick set at the time of the IPO.
In terms of disclosure standards, Q2 was the last quarter to separately disclose the metrics of logged-in and non-logged-in users. Starting from Q3, the company will no longer separately disclose logged/unlogged daily active users, and will continue to disclose the total number of daily active users and weekly active users in the US and international.
Q&A Q & A
Q: What are the drivers behind the strong advertising business this quarter?
A: This was another strong quarter, and the business showed strong consistency. First, our story is being widely accepted by advertisers — Reddit's influence in the buying decision journey is already clear, and advertisers are increasingly aware that their customers are giving more value to “advice from real people,” and Reddit is the location of this kind of advice, so they understand the importance of our environment. We continue to deliver results at the competitive level of the market, and our strategic potential is accumulating.
Our strategy is to make every presentation more valuable. When conversions and clicks are growing, what that means for advertisers is that they can get more results for the same or better price. Our work on signals, machine learning, and optimization continues to improve our competitiveness.
Another thing that is really starting to gain momentum is the strong adoption rate of the tool, which lowers the threshold for advertisers to get started, so you'll see the number of advertisers increase by about 70% year over year; after getting started, we deliver more efficiency, thereby retaining them and making them expand their budgets. At the same time, we're making it easier for them to stay “always open” by increasing credit limit adoption rates and reducing friction across funnels.
Finally, I'd like to say that our business is being decentralized on multiple dimensions, which makes it very resilient: 11 of the 15 vertical industries grew by more than 50% year over year, the number of active advertisers increased by more than 70%, scaled channel revenue for mid-sized customers and SMBs doubled, and other regions grew by more than 80%. Our landscape has become more diverse, and our business has become more resilient and resilient as a result.
Q: What do you want investors to understand about the user growth trends in early Q2 and Q3, and the ability to attract high-quality users?
A: As we said last quarter, our target is 1 billion day jobs globally and 100 million day jobs in the US, and that hasn't changed. Reddit is one destination. We have nearly 200 million weekly visitors in the US, and we are growing mainly through product work to drive daily direct users, and we are making progress in this area. We added Nikkatsu through our products this quarter, but this was offset by a decline in search traffic.
On the product side, we've improved user retention for new apps — one of our most important metrics and priorities, with a 50% year-over-year increase in relative caliber. Our reach is expanding, with weekly activity surpassing 500 million, and weekly activity both in the US and international growing month-on-month. On the content contribution side, we make it easier for users to find and post in the right community. Currently, there are 26 billion posts and comments on Reddit, adding 1 billion more since Q1, and the active community has grown significantly. We are also more effective at turning web users into higher-value app users and direct visitors. Video and review functions have also expanded the upper funnel.
We did see headwinds in search diversion, especially in the latter part of the quarter. The visibility of channeling traffic is still very low, but we are focusing on what we can control, that is, direct product-driven users. We believe this work will be transformative and that it will take time, but this is how we will reach our goal of 1 billion in the world and 100 million in the US.
Q: Since Zhouhuo's growth is stronger than Nikkatsu, which products can complete the transformation from Zhouhuo to Nikkatsu? What's the focus on feeds in particular?
A: The increase in daily activity is essentially an increase in frequency, which is mainly achieved by making Reddit more sticky. You mentioned feeds — short, medium, and long-term feed improvements will increase the frequency, and there is a lot of room. Frankly speaking, our feed technology lags behind the most advanced in the industry, which in turn gives us a lot of room to catch up. I'm particularly encouraged by the talent brought in over the past quarter who will help us build the next generation feed.
But all the minor improvements are just as effective. We carried out hundreds of experiments last quarter, and many of them were cutting through gaps in retention and quality little by little on the way to full volume, which in turn led to growth. These tasks are spread across new user guidance, posting, and overall app performance, and we are really starting to feel the potential.
There's also conversion — turning our huge web upper funnel into mobile app users. We're also stepping up our efforts in this area to drive more downloads. Many of these users are already Reddit users themselves; they have good retention and perform better in the app. So progress is being made on all lines, and there is still plenty of room.
Q: There have been many reports about data licensing cooperation recently. As the renewal period at the beginning of next year approaches, what are the strategic updates?
A: The strategy has not changed; we are working hard to advance it. Reddit's content is in demand, and it's not popular content—people actively search for it by name, and many people go directly to Reddit. Our partnership is multi-level: Reddit is used for training, for post-training to teach the model how to “talk”, for grounding (factual support), and as a search index. So this matter has many levels and dimensions, and the value of Reddit data is far more than the original training data.
Nor are these deals binary. They have to be meaningful to Reddit, but I think there are many ways value flows back to Reddit, and our goal is to maximize the value that flows back to Reddit.
Q: Stock prices have dropped sharply, and investors generally believe that the company has a “user problem” — especially in the US. Despite huge investment over the past year, Nikkatsu has declined slightly; and as search moves to AI, unlogged traffic will continue to be pressured. What makes management so confident that they can reach the goal of 100 million registered users in the US and 1 billion globally?
A: Let's start with the conclusion: We are confident because our product is working. The essence of Reddit is community and conversation, and the community is universal. We think in America we have content for everyone; the question is just how to reveal it, and we are making progress in this direction. We're proud to significantly increase new user retention within the app within a quarter, and this improvement in retention drives growth and is a job we can control.
Yes, external searches fluctuate, especially on unlogged web pages, but that's not where we do business. We do our best to get as much value as possible from this traffic, which is why we're always driving the conversion from web to app, and the efficiency of this matter is also increasing. So in the long run, we are as confident as ever that we will reach the milestones we set. However, this work takes time, we still have a lot of work to do on the product, and we have indeed seen some progress in driving direct user growth within our control.
Q: Is it possible that Google and OpenAI will not license data next year?
A: The results span a wide range, and we have to look at every aspect of this matter to ensure we maximize the value of Reddit. As far as Google is concerned, the relationship is more than licensing — our relationship actually predates formal data licensing agreements, including its long history in the “Ten Blue Chains” era, and Reddit's current position in AI Overviews; licensing is only one part of it. We've also seen the way Reddit content is being used is expanding, as is Google and others.
So I don't think this is a binary result. Whether it's training, AI overviews, or the new forms that come next, these different touchpoints are independent decisions, and we'll ensure that each one maximizes the value of Reddit. And for us, the focus isn't just about traffic or clicks — we want to bring users into the Reddit ecosystem. Reddit is a product of community and conversation, so we want to bring users into this ecosystem and Reddit's flywheel, which is what we want to achieve in all of these kinds of collaborations.
Q: Are there more data buyers besides Google and OpenAI?
A: Of course. One phenomenon we've seen over the past year is that more and more actors are interested in Reddit's data for different use cases and at different scales. So we're looking at this as an expanding market with lots of different opportunities.
Q: How do you understand search drift headwinds in the latter half of the quarter—regular fluctuations that occur from time to time, or are they larger/lower than normal? What are the causes? (The response to the search guide section of the transcript is missing. The recorded answers are feed technology, video presentation within the feed, and search monetization)
A: Regarding feeds, the focus is on improving the technology that supports feeds because it takes on a lot of the heavy lifting — scheduling the various subreddits that users subscribe to, and is one of the main drivers of subreddit discoveries, and I expect it to continue to do most of the heavy lifting. You've also touched on one of the things we're doing right now: how to present videos in the home feed.
As there is more video content on Reddit, we've been thinking about this for a while. It has two levels: the first is how to establish the video as a whole in the feed, because it's a mixed-media feed — with video, text, and images, so the flow from feed to theater mode, to review pages, and back to the feed is critical. There are great opportunities for improvement, and we're doing something that can really make the process smoother. The second is to create an “audible Reddit” experience — so that users can not only watch videos on Reddit, but also listen to posts and listen to posts in the background. We've seen people outside the site already doing this. An emerging type of content is emerging on the internet, essentially a podcast where people read Reddit content aloud. So I think this kind of “listening version” or “oral version” of Reddit would also be very appealing; it's almost a completely different form. These are all things we're working on right now and are looking forward to testing later this year.
We're making steady progress on Answers and Search. The search box is now globally available within the app, and the number of users and searches grew this quarter. Reddit is now the best platform for searching Reddit content, at least in terms of many of my own queries — this wasn't always the case in the past, and we're continuing to close this gap. Any query you want to know about something, or want to see multiple opinions - “What should I watch?” What do people think of this?” What should I buy?” ——Reddit provides the best answers on the internet. I'm very encouraged by the progress in this area, and we're also starting to focus on advertising on this interface.
From the advertising side, search is highly tied to the shopping experience. Here are a few different perspectives. First, we think the search page can be enhanced by richer rich media modules — when users are searching for and discussing a specific product, they can access visual material from our existing catalog. We've started doing this, starting with early testing in the electronics and consumer electronics categories, and now expanding into more categories. This will enrich the core search experience, hopefully increase engagement and allow users to get more out of it. I also agree that the Agentic feature on Reddit Search is now the best way to search Reddit.
The second is acceptance when participation matches at the product level. The shopping listing ads I mentioned earlier can accommodate multiple different retailers, product types, and brands in one module. This is perfect for the search page scenario, and I think it's the most appropriate way to present ads in search. This is a space we're watching. We clearly have the ability to implement it. We'll continue to track the page's development and user adoption. We do see advertising opportunities here. The good news is that the infrastructure and most of the capabilities are already in place.
Q: What is the specific reason for the 50% year-over-year increase in user retention for new apps?
A: The 50% increase in user retention for new apps is a difficult number to drive, and one of our most important metrics. I think it's our best measure of product quality — users are willing or unwilling to come back next week after opening Reddit, which is something we've been promoting.
There are several factors behind it. The biggest one is conversion: we turn core users or high-frequency users on the web into app users, and they become better users in the app, and we have many such users. The second is the app itself: new user guidance has been greatly improved to more effectively connect users to their “home” on Reddit. There are also “battlefield” jobs, such as logging in, recovering accounts, and keeping the login state after reinstallation, etc., which added up to push up this figure.
I think there's still a lot of room for improvement in the absolute level of new user retention. This is something we've been talking about for years: once users “understand” Reddit, they really love it, but the product is really a bit difficult to understand, so every time we make it easier to understand, we grow. This is the healthy structural shift we've been looking for; I can already see potential energy; this is where our confidence comes from.
Q: What are the specific areas of focus for improving the feed model?
A: Broadly speaking, the models we use today are much smaller and significantly smaller than the most advanced in the industry, so we can scale up the models. Our models update more slowly and are measured in “days,” while the most advanced level is “hours” or even “minutes.” The user behavior we actually include in the model is only about 10% of the total user behavior, and this ratio can be significantly increased.
There's also a shortlist — that is, posts that are eligible to be recommended, and Reddit is currently limited to the most recent week. So our feed is basically like a live feed, and we actually have an extremely large corpus of content, much of which is not affected by time, such as parenting advice, book reviews, or film reviews. These contents have been valuable for a long time, and we haven't displayed them on the feed at all.
Therefore, in almost every dimension, such as candidate set selection, model size, model speed, and included user signals, we can make significant improvements, and some even have orders of magnitude improvement. We'll be doing this work over the next year, and I expect every improvement we make will work because our starting point is just too low.
Q: Can the progress of the SMB business be broken down (Shopify access, Max performance), and how should this customer base be expected in the second half of the year?
A: SMB is a very growing segment, but there are still thousands of SMB's around the world that haven't joined our platform. There are a few things: One is that we continue to get customers on this channel, and Shopify's partnership clearly allows us to do that. Now that we have the ability to automatically sync and complete the integration of product catalogs from Shopify merchants, we will do some joint marketing to speed up customer acquisition, and really increase the strength of this kind of joint marketing.
Second, today's SMB experience on Max is not the best. We're still doing the groundwork — what's the right entry point for SMBs that removes more friction, fits their needs, and lets them get started and enjoy Max faster. This is a work ahead, but we think it's important for reaching more customers, making customer acquisition smoother, speeding up customer acquisition, and enabling customers to have a good experience faster.
That's what we do for SMB reserves. Of course, SMB also benefits from all of our horizontal work, which is to generally improve the efficiency of all delivery goals, and everyone benefits. So I feel good about the potential of the SMB business, but I think there's still some more customized product work to be done for this customer base.
Q: Other platforms that have reached the level of 1 billion users have creator ecosystems. Does Reddit need to make different arrangements at the content level to reach this scale, or is the existing content sufficient?
A: There has been a lot of progress in the creator ecosystem this quarter: the number of posts per person and number of comments per capita (based on login date) has all risen globally and in the US, and the active community has also grown. There's a lot of work behind it: we've added a new poster and a new commenter on both mobile and web to make posting easier; we're better at guiding users to subreddits where they are more likely to post successfully, and community recommendations are more accurate. We're still doing what I think has the greatest chance — improving the posting success rate, making it easier for new users to post, and the results are more consistent.
One of the current challenges is that new users' posts are easily deleted because their accounts aren't “old” enough. This type of strategy, which has been around for a long time, is out of date. With LLM-assisted reviews, we can bring in potential high-quality users while continuing to better keep spammers and other bad behavior out. Video and comments are a new feature that has contributed more than 10% of video posts, and engagement performance is good.
I do think our creator ecosystem will be different from other platforms. My usual description is that social media platforms work for that 1% of creators, while Reddit works for everyone else. Our opinion is that everyone has something to say, even if you're not a professional content creator—everyone has questions, needs, suggestions, or something interesting to share, which is the point of Reddit. So there are tons of professional creators on social media, and Reddit belongs to everyone else. Making Reddit easier to post and improving everyone's posting success rate is the key here, and we've made great progress this quarter.
Q: Considering the current level of ARPU, the ROI for marketing campaigns seems quite attractive, so why not invest more? What is the user retention situation brought about by these channels, and what can be done to improve it?
A: Marketing is a very important part of achieving our goals. This is critical. Of course, we want to continue to do user acquisition and marketing. Our model is pretty straightforward, and I think it's very similar to other companies in the industry.
I think the key unlocking point is that we need better user retention. As mentioned earlier, this indicator increased by 50% this quarter, which is beginning to confirm the effectiveness of product work, but there is still more work to be done — relatively speaking, we are still below our peers in this field, so we will continue to attack this matter.
On the marketing side, we also launched the “People Are the Best” brand campaign this quarter, which really stirred things up, and we received a lot of positive coverage. I think the more we can balance user acquisition and brand targeting, the better the end results will be. So overall, we are committed to marketing investment. We believe the brand will benefit from it, and we are willing to continue to invest throughout the cycle. Internally, discussions are also being made about whether to invest more in the brand. But overall, what really unlocks more UA launches is better retention; it's that simple. Our gross margin model is excellent, so an incremental user here is worth a lot. You also pointed out that the ARPU is very high, both of which are great and would really help us. We just need to go one step further in terms of retention to really speed up this matter.
Q: As the company continues to expand, where are the biggest operating leverage opportunities in the next few years?
A: Our incremental profit margin has been high -50% s to low -60% s for most of the past few years, and the incremental profit margin here is very strong. But we don't manage the business with incremental profit margins; we look at the individual ROI of every investment. I think we've always been very lucky with incremental transfers — when we get incremental revenue, especially on the gross margin line, the turnover rate can reach 95% to 99%. This is how I think about incremental things, and then it's just a question of “what investments do you want to make”.
Now what we're doing is investing we want to make. We don't withhold our investments because “we need to maintain incremental profit margins as high as -50% s”; this is not how the company works. We really look at a single investment; where it makes sense, we put money in. I think our investment has led to a lot of success: we've invested a lot in the ad tech stack, and you can see results in monetization; we've invested a lot in the sales team, and you can see more than 60% growth for eight consecutive quarters — these are probably our most notable investments. As mentioned before, we are also investing heavily on the talent side, introducing machine learning talents, and substantially upgrading our growth team.
We are making these investments because they are the right ones. I do think profit margins will continue to accumulate. The long-term target is 50%, which is the direction we are currently focusing on. I think most of this will come from the leveraging effect of incremental revenue. Currently, there isn't a single large amount of cost we are staring at and “this piece can be cut off.”
Q: Does the company emphasize the importance of high-quality (app-side, login) users. Does that mean that even if impacted by diversion traffic, the value of a high-quality user is several times that of diversion users, so it can use the upward shift in the user structure over time to offset the impact of about 300,000 month-on-month decline in daily activity in the US? What relevant metrics will be disclosed in the future?
A: Your understanding is absolutely correct. Direct users, app users are several times more valuable than search or viral traffic. We're not building Reddit for that kind of “pass-by traffic,” we're building it for direct use; this is where our business is. You can see this in the performance of the ad business — despite fluctuations on the search side (we've seen this kind of fluctuation many times in the past), our revenue growth is still very strong and steady.
The reason is that our business is based on those direct repeat users and the growth of this base. Through both conversion and retention efforts, I think we've made good progress this quarter, and this is something we can control no matter how the broader search ecosystem changes.