Operator
Hello, and good morning, everyone. And welcome to the VerticalScope Holdings Inc.
Q3 2025 Earnings Call. My name is Emily, and I'll be coordinating your call today.
[Operator Instructions] I would now like to turn the call over to Diane Yu, Chief Legal Officer, to begin. Diane, please go ahead.
Diane Yu
Thank you, operator. Good morning, everyone.
And welcome to VerticalScope Holdings Third Quarter 2025 Earnings Call. I'm joined by Chris Goodridge, our Chief Executive Officer; and Vince Bellissimo, our Chief Financial Officer.
We'll begin with commentary on the quarter before opening the floor to questions. Before we begin, I'd like to remind everyone that today's presentation contains forward-looking information that involves known and unknown risks and uncertainties and other factors that could cause actual events to differ materially from current expectations.
These statements should not be read as assurances of future performance or results. Such statements involve known and unknown risks, uncertainties and other factors that may cause actual results, performance or achievements to be materially different from those implied by such statements.
A more complete discussion of the risks and uncertainties facing the company appears in the company's management discussion and analysis for the 3 and 9 month period ended September 30, 2025, which is available under the company's profile on SEDAR+ as well as on the company's website. You are cautioned not to place undue reliance on these forward-looking statements, which speak only as of the date of this presentation.
The company disclaims any intention or obligation, except to the extent required by law, to update and revise any forward-looking statements as a result of new information, future events or for any other reason. Our discussion today will include references to adjusted financial measures, including adjusted EBITDA, free cash flow, free cash flow conversion and MAU, which are non-IFRS measures.
All references to currency in this presentation shall refer to USD unless otherwise specified. Now I will turn the call over to Chris Goodridge, CEO of VerticalScope.
Chris?
Christopher Goodridge
Thanks, Diane, and good morning, everyone. It's a fast-moving environment.
And despite some short-term volatility in our monthly active user base, I'm really encouraged by how our teams are navigating this change and focusing our efforts on growth. The strategy that we articulated on our last call is our blueprint for driving organic growth and continuing to build our platform and cash flow.
We're building stronger direct relationships with our users, making their experience richer and more useful with AI and turning that direct engagement into diversified revenue streams. And we'll use our free cash flow and financial capacity to pursue growth opportunities that further our strategy and positioning in a more AI-centric web.
We believe that the most enduring and successful online experiences will be protected spaces with micro communities of trusted users. This has been the core of our business model for years and will continue to be the foundation of our long-term success.
Turning to our results. I'll start with some observations on our MAUs and the trends we're seeing.
We averaged 83 million monthly active users in Q3, which was down from the record levels we experienced last year as a result of a surge in search-based traffic. However, Q4 started on a much stronger note with MAUs surpassing 90 million in October and we saw gains across all major traffic sources and most importantly, with direct users, which were up nearly 60% over last year.
With the majority of our MAU now coming from non-Google sources, we think the worst is behind us and expect to see MAU growth going forward from direct and other sources. Beyond direct and search, we're also seeing opportunities to grow our paid sources of traffic and to efficiently acquire new users who will be members of one or more of our communities for years to come.
Paid channels have historically been a minor source of users for VerticalScope, but we believe this is an important channel to drive additional user growth in the years ahead as search experiences become more fragmented and people increasingly seek out authentic perspectives from real users to validate information they receive from AI. When you combine our efforts to grow direct users with paid channels, we're well positioned to grow our total user base from here.
Vince will go through our financials in detail shortly, but I wanted to offer up a few observations. First, our revenue improved modestly on a sequential basis in Q3 and came in at $14.7 million.
When we compare back to last year, revenue was 17% lower, essentially all as a results of programmatic ads tied to search traffic. Second, aside from programmatic, all major sources of revenue were either flat or up in Q3.
Direct advertising has been very resilient and was up 3% year-over-year as we saw a nice pick up with some key auto, powersports and insurance customers in the quarter, driven in part by new immersive experiences we're making available to brands. E-commerce grew by 40%, mainly as a result of the Ritual acquisition with the rest of our subscription and transaction revenue sources remaining stable.
We're still seeing plenty of opportunity to build up commerce, both in our core community experience but also with Ritual. When you combine our solid direct advertising results with e-commerce gains, ARPU pushed up 21% over last year and reaching its highest level that we've seen in 3 years, and demonstrates our ability to generate more value even with volatility in MAU.
I think most impressively in the quarter, though, was adjusted EBITDA improving by 45% over Q2, margins exceeding 40% and our free cash flow conversion growing to 94%. This result clearly demonstrates the strength of our business model and the adaptability of our team.
Turning to product initiatives. AI continues to open up completely new experiences for our users and is increasingly providing us with operational gains.
Machine translations are introducing our communities to new audiences speaking German, Spanish, Portuguese, Dutch, Polish and French and AI summaries are making it easier for new users to find what they're looking for in threads. We're most excited about the potential of our AI community assistant, Fora Frank.
We introduced Fora Frank last quarter, and we've continued to roll it out across our communities. Its main purpose initially was to encourage posting activity from our users and help them ask better questions to elicit higher quality responses.
We're taking a lot of inspiration from how other platforms have successfully used conversational AI to drive engagement, including the X platform and how it's deployed Rock. Our early results suggest that thread engagement increases by over 3 times when Fora Frank is mentioned.
So we're looking for more ways to grow Fora Frank's visibility and distribution within the communities while continuing to improve its capabilities. Our goal is to have AI enhance the human experience within our communities, not to replace it.
There are so many interesting applications we see for the technology. For example, with our SMB subscription product, Fora Frank can help our customers post and drive more engagement with their content, adding value and improving retention.
We've also built an AI prospecting tool for our sales team that identifies new potential customers and incorporates relevant community content into our outreach messaging. We see this as a game-changing capability that we believe will boost the efficiency and close rates of our sales team.
Beyond the work to improve our community experience and grow direct sales with AI. We're starting to envision completely new AI-driven services powered by our rich data sets and that can leverage our distribution to millions of users.
It's very early but we're excited about these opportunities, and we'll have a lot more to say in the quarters ahead. Turning to our data license efforts.
I'm keenly aware that we've been discussing this opportunity for several quarters. This space is evolving very quickly and I believe our patient approach will pay off.
We're at the very beginning of a major shift with how content on the web is both created and consumed and how consumer decisions will be made. We know that users like the experience of engaging with an AI chatbot and we expect that they will also start to transact more within those experiences.
But for that to happen to be a great user experience, the AI must have access to the best data available. Over the past several months, we're starting to see a shift in market conditions as the lack of a real value exchange between LLMs and content owners becomes an increasing source of friction.
For example, major lawsuits have been filed against the LM companies, including Reddit's recent lawsuits against Anthropic, Perplexity and a host of other data scrapers for the alleged unauthorized use of Reddit user content. In parallel, industry efforts are underway to establish standards for data licensing and compensation through the RSL Collective, a nonprofit organization that is using the music industry as a model and which is supported by a number of significant platforms, including Reddit, Internet Brands and VerticalScope.
And finally, we're seeing key players powering the infrastructure of the web, taking steps to close off access by AI companies. Cloudflare, which is approximately 20% of total Internet traffic flowing through its network, has been particularly active on this front and is now blocking AI scraping by default.
Base model training is obviously still relevant for LLMs but we see bigger potential opportunities with AI companies requiring access to the most up-to-date information through retrieval augmented generation. We expect this need to accelerate as more and more AI services are launched and scaled.
As this happens, we believe that access to accurate, up-to-date information will be critical. Data quality will be the currency.
Our communities contain the authentic, high-intent human context that AI agents need to understand what people actually want to do, and we believe this positions VerticalScope as part of the foundational data layer for this new agentic web. So where are we at with all of this?
We've taken steps over the past several quarters to block all known AI scrapers at the CDN level. We have very clear terms of use that prohibit scraping outside of a licensed relationship, and we're regularly updating our robots.txt to disallow AI scraping.
Another big step forward on this front is our recent partnership with TollBit. TollBit's technology integrates with our CDN to intercept AI scraper traffic and redirect it to a paywall.
From there, the AI company can either pay our set rate to start scraping or contact us to license access to a richer structured data feed through our API that would come at a premium. We've been working through the process of onboarding TollBit across our network of communities over the past few weeks and are close to completion.
TollBit's analytics suite gives us a detailed view of AI bot activity we can detect across our communities. And early data highlights just how aggressively AI systems are trying to access our content.
To illustrate the point, TollBit data from the past week detected scrape attempts, which we're actively blocking, outnumbering real users in some cases by up to 13 times. While this reflects only the activity we can observe, it underscores the scale of AI demand for our data.
We're actively monitoring this changing landscape and continue to evaluate how regulatory, legal and commercial developments may affect our strategy. At the same time, our capabilities are improving, our sense of demand and value is growing.
We intend to service that value in a way that best serves our communities and shareholders for the long-term. Finally, before passing it over to Vince, I'll offer a few comments on capital allocation and M&A.
We've made 4 small acquisitions so far this year and we expect to complete a couple more small ones before the year-end. We continue to see higher inbound activity, which we think is tied to how smaller companies are navigating the broader industry change.
Again, we think our shareholders will be best served by a patient approach and so we expect to continue to accumulate cash and build capacity to pursue larger opportunities that will accelerate our long-term growth strategy. And with that, I'll turn it over to Vince to walk through the numbers.
Vincenzo Bellissimo
Great. Thanks, Chris, and good morning, everyone.
I'll walk you through our third quarter results and share how we are executing against our financial and strategic priorities. Overall, this was a solid quarter for our business, underpinned by our disciplined and hyper-focused approach to execution.
We delivered improved performance sequentially, including expanded adjusted EBITDA margins and stronger free cash flow conversion while continuing to strengthen our balance sheet. Our growing cash position provides flexibility to reinvest in growth, both organically and through strategic M&A when the opportunity is right.
Our focus on audience quality, ARPU growth, operational efficiency and liquidity continues to position us well in a rapidly changing environment of AI content discovery. Now turning to our results.
Total revenue in the quarter increased 1% sequentially to $14.7 million, reflecting stability of our core audience but declined 17% year-over-year, primarily on a 32% decline in MAU compared to all-time highs in the prior year, led by lower value search traffic. This led to a correlated decline in digital advertising revenue, which finished the quarter at $11.7 million, down 25% year-over-year.
The decrease was driven by a $4.1 million decline in programmatic advertising revenue, which contributed just over $7 million in the quarter on lower display impression volumes compared to the prior year. Despite the declines in programmatic, we saw a return to growth in both our higher-value direct and video advertising channels, supported by strong demand for custom content campaigns and ongoing optimizations with our video ad unit.
Direct advertising grew 3% to $4.6 million, representing 40% of overall digital advertising revenue compared to a 29% share in the prior year. The growing share of direct revenue highlights the strength of our relationships with advertisers and brands across our core categories and a rising demand for customized high-intent campaigns that reach our enthusiast audiences.
In an era of AI-driven content discovery, this audience is becoming increasingly scarce and valuable, a combination that will drive growth opportunities in the periods ahead. Turning to e-commerce.
Revenue grew 40% year-over-year to $3 million, primarily from contributions from our April 2025 acquisition of Ritual, a local food pick up and ordering app that connects users with restaurants in Canada, the U.S. and Australia and stable performance across our other e-commerce offerings.
Excluding Ritual, approximately 60% of our e-commerce activity continues to come from subscriptions, underscoring the stability and loyalty of our user base. We continue to view e-commerce as an important long-term growth driver, supported by ongoing product innovation and the use of AI to enhance discovery and personalization across our communities.
Overall, ARPU increased 21% year-over-year, including a 10% increase in digital advertising ARPU and a 106% increase in e-commerce ARPU. ARPU remains a key metric for us and a reflection of our ability to grow and monetize a high-value direct user base, capitalize on premium advertising and commerce opportunities and unlock new monetization opportunities that are powered by data and AI.
Turning to our profitability and free cash flow generation, both key highlights of the quarter. Adjusted EBITDA for the quarter increased 45% sequentially to $6.2 million, driven by cost efficiencies but declined 16% year-over-year due to lower revenue, partially offset by cost savings realized in the period, including the benefits of headcount and operational changes made in the first half of the year.
The quarter also included a benefit from -- the quarter also benefited from $600,000 in tax incentives under the Canada Revenue Agency’s SRED (sic) [ SR&ED ] program, which we apply for annually as part of our ongoing investment in technology on the Fora platform, and are recognized as a reduction in wages on the P&L. Historically, these incentives have been approved and recognized in the fourth quarter but timing can vary year-to-year.
Together, these factors contributed to adjusted EBITDA margin expanding 12 percentage points sequentially to 42% compared to 30% in Q2 and consistent with 42% margins in the prior year. This improvement highlights the impact of operating with smaller, more focused teams that are leveraging automation and AI tools to deliver on key growth strategies.
Our profitability this quarter also demonstrates the resilience of our business model even as MAU levels remain well below last year's record highs. This reflects a more diversified revenue base and the growth of a higher-value direct audience that increasingly insulates our results from search volatility.
Net loss for the quarter was $400,000 compared to net income of $1.2 million in the prior year, primarily reflecting lower revenue and partially offset by lower operating and income tax expense recognized in the period. The net loss for the period included noncash depreciation and amortization expense, primarily related to acquired intangibles of $4.8 million compared to $4.4 million in the prior year.
Turning to cash flow and liquidity. We once again delivered strong cash generation in the quarter.
Operating cash flow was $4.7 million and free cash flow increased 56% sequentially to $5.9 million, representing a 94% conversion from adjusted EBITDA, up from 86% in the prior year. We ended the period with $68.4 million in total liquidity, including $12.4 million in unrestricted cash and $56 million in undrawn revolver capacity.
Our balance sheet remains a key strength with net leverage of 1.24x as defined by our credit agreement, providing ample flexibility to invest in growth initiatives and pursue opportunistic M&A, particularly in areas that accelerate our progress in AI and direct traffic initiatives. Maintaining a strong liquidity position remains our priority.
From a capital allocation standpoint, we continue to believe that reinvesting in growth, expanding our audience, data capabilities and AI-driven monetization will create greater long-term value for our shareholders. In closing, there is no change to the full year guidance that was published in April of this year.
Our third quarter results demonstrate the impact of the actions we've taken in a short period of time to streamline operations and focus on strategic priorities that drive direct audience growth and higher ARPU across our platform. The world of AI content discovery depends on high-quality human-generated content that's exactly what the Fora platform provides.
Our users and the authentic content they create are our most valuable assets and we will continue to do everything possible to protect these assets and unlock opportunities that are sustainable. With a strong balance sheet, differentiated data and a disciplined approach to execution, we are well positioned to create long-term value for our shareholders and employees in this new phase of the agentic web.
And with that, I'll pass it back to Chris to close things off.
Christopher Goodridge
Thanks, Vince. With that, we'll open the floor to questions.
Operator
[Operator Instructions] Our first question today comes from Gabriel Leung with Beacon Securities.
Gabriel Leung
Just a couple of things. First, just on the cost side of the equation, Chris or Vince, how are you feeling about the current structure right now of your roughly 170 full-time equivalents?
Do you feel that's the current -- the right cost structure to drive the growth you're planning on implementing over the next 12 to 24 months?
Christopher Goodridge
Yes. Thanks, Gabe.
Thanks for the question. So yes, for the most part, we feel pretty good about where the headcount is.
There'll be -- we'll be adding selective roles here and there. What's really changed for us and something we're really trying to push within the organization is using AI tools to become more efficient, right?
So we're rolling that out really across all of our teams, not as a means of reducing headcount further, but really to drive more productivity out of the team that we've got. But we do think there's opportunities to add certain skilled positions to the business over time.
It won't materially change the headcount over the next, call it, year or so. But we expect headcount certainly not to go down from here.
Gabriel Leung
And then just secondly, on the data licensing. I know you're, I guess, in the tail end of deploying TollBit and their tech across your community.
So I'm curious beyond the initial observations, do you or TollBit, have you thought about how the revenue side of it might play out over the next 12 months or so? Or have you had early discussions with any of the -- some of the AI companies in terms of either licensing or paying a fee on the scraping?
Christopher Goodridge
Yes. Thanks, Gabe.
There's absolutely discussions that are ongoing with the major players. TollBit's monetization, that part of the platform is relatively nascent to be fair.
And the marketplace opportunity that they see, they think is quite big, and it really doesn't apply to just the major players. It's meant to be really anyone who's built an AI service and requires access to kind of fresh data to power it.
So they see that as many, many players in the space over the long-term. So I think that is more of a long-term play for us from a monetization side.
Like I said, though, what it does is it really empowers us with a lot of great data that helps us articulate the value proposition of our underlying data asset. And so I think the bigger players in the space that are building models that are building kind of chat experiences where they require RAG to offer up fresh information to what the core model offers, that's where I think it's more likely that you'll see direct deals over time.
With respect to financial impact, we're not at a stage where we're going to provide a forecast with respect to how that's going to play out over the next period of time. But as that unfolds, you guys will have a lot more information.
Operator
At this time, we do not have any further questions registered. [Operator Instructions] We have not received any further questions, and so I will turn the call back over to Chris for closing comments.
Christopher Goodridge
Well, thanks, everyone, again, for joining us today. I hope the rest of your year goes quite well, and we look forward to getting back together with everyone again in March to review our year.
Thanks again, and take care.
Operator
Thank you, everyone, for joining us today. This concludes our call, and you may now disconnect your lines.