fbpx

AdTheorent Stock for Programmatic Digital Advertising

In a sign of the times: Alphabet, Google’s parent company, reported an 8% decline in quarterly profits for Q1-2022. Don’t worry: We don’t have to start a GoFundMe campaign or anything. The titan of the tech giants still had net profits of more than $16 billion on about $68 billion for the most recent quarter. Somewhere around 80% of the company’s revenues are advertising dollars. Our own experience as an online publisher turned us off to the whole digital advertising ecosystem years ago, which is why we scrape by on the subscriptions of our most loyal readers. 

At the end of the day, Google is a digital advertising company that dabbles in a variety of other (mostly money-losing) ventures like digital healthcare. Retail investors interested in this theme would probably be best served by owning shares in Alphabet and calling it a day. In fact, we assumed there wasn’t much competition outside of Facebook. Certainly, no pure-play digital advertising companies for investors. That turns out not to be true, partly thanks to the boom-and-bust cycle of the SPAC era. Mergers between private tech firms and special purpose acquisition companies (SPACs) thrust many under-the-radar startups into the public spotlight. 

Most have wilted in the glare. A few are creating significant revenues, with a much smaller percentage actually beating pre-SPAC merger projections. Turning a profit? Nearly unheard of since SPAC hype peaked in the middle of 2021. So imagine our surprise when we found a deSPAC company that hit the trifecta – significant revenues, realistic projections, and profitability. Meet New Yawk-based AdTheorent (ADTH), which uses artificial intelligence to make programmatic digital advertising more effective. 

What is Programmatic Digital Advertising?

Yeah, we had to look it up ourselves: Simply put, the term refers to the automated buying and selling of digital ad space by using algorithms or some sort of sophisticated software to figure out how best to convert eyeballs to dollar signs for advertisers. There’s a pretty good breakdown of how it works here, or we can just cheat and look at this crappy low-res graphic courtesy of AdTheorent:

Programmatic digital advertising
Credit: AdTheorent

Digital ads are sold in online exchanges where advertisers bid against one another for marketing space that algorithms have predicted will buy them the most bang for the buck. The whole real-time bidding process happens in milliseconds – millions and millions of times every hour – so you want the best algorithms fighting for you. This is where AdTheorent comes in: As a display-side platform (DSP), it does a bunch of machine learning magic (more on that later) behind the scenes to ensure its clients convert customized advertising into new customers.

About AdTheorent Stock

Click for company website

AdTheorent had been around for about a decade before it opted to go public through a merger with a blank check company. The deal crossed the finish line in December 2021 and the stock started trading on the Nasdaq before the end of the month. It’s not entirely clear how much money from the merger actually made it to AdTheorent, which waived its minimum cash requirements before the merger could be completed. There was $315 million in the SPAC trust, and we assume institutional investors redeemed much of that, based on the waiver agreement. Presumably, the additional $121.5 million in private equity attached to the deal did go through. One significant name attached to that particular pot of money is Palantir Technologies (PLTR), a $20 billion enterprise AI company that churns out insights for its clients using algorithms and big data. 

Last year, AdTheorent reported revenue of more than $165 million, an increase of 36.6% compared to the previous year. Perhaps more importantly, the year-end total is more than 4% higher than the company projected in its shiny 2021 investor deck. Its guidance of between $202 million and $206 million in revenue for 2022 is also in line with its previous promise to deliver $201 million this year. While the stock has experienced its share of volatility – dropping as much as 55% at one point – it’s currently trading close to its $10 baseline with a market cap of $715 million. 

Machine Learning in Advertising

AdTheorent goes into great detail about its technology platform in its first-ever 10K annual report to the Securities Exchange Commission. That gives us some confidence that the company isn’t just throwing around buzzy AI terms. Here’s the short version of how it works, parsed and paraphrased for your enjoyment:

AdTheorent develops custom machine-learning models for its clients based on historic and real-time data to predict where digital ads will perform best and then bids on those placements. It scores more than one million potential ad impressions per second, which works out to something like 75 billion to 90 billion digital ad impressions per day. It determines each predictive score by correlating non-personal data attributes associated with the particular impression against data corresponding to previously purchased impressions that yielded consumer conversion or engagement activity. These attributes include everything from the publisher and content to device make and operating system to geographic data, weather, and demographics.

The end result, according to the company, is a better return on ad spend for its customers because the algorithms don’t waste their time on suboptimal impressions, including those predicted to be at a greater risk for fraudulent traffic or even potentially detrimental to the brand. For instance, AdTheorent built custom machine learning models for a Fortune 500 pharmaceutical company to identify people suffering from a specific health condition who would be most likely interested in the company’s drug product. Some of the results:

Case study from AdTheorent
Case study. Credit: AdTheorent

One particular thing to note about AdTheorent is that its platform does not rely on individual user profiles or cookies for its algorithms to build predictive scores. Its targeting approach is statistical, not individualized, so there are no privacy concerns to worry about. That’s especially important given the swing toward more user control over privacy settings. For instance, Facebook admitted that the new privacy settings initiated by Apple would take a $10 billion chunk out of its advertising revenue this year. The ability to customize digital ads without cookies potentially gives AdTheorent an edge against the competition, as more users opt out of being tracked.

Customers and Competition

Privacy is especially important to some of AdTheorent’s customers in financial services and healthcare, two of six industries the company serves. Last year, those two verticals accounted for nearly half of all revenues in Q2-2021:

Verticals served by AdTheorent
Credit: AdTheorent

Unfortunately, AdTheorent doesn’t provide the same breakdown in revenues in its latest SEC filing. It does note that the number of active customers grew more than 14% in 2021 to 309. An active customer is defined as one that spent more than $5,000 during the previous 12 months. Customer concentration is a slight concern, with five customers representing about 25% of revenue. That’s an improvement from the year before when one customer accounted for about 10% of revenue and five others combined for another 32%. Still, that’s a lot of eggs in too few baskets, especially if any of them are in more vulnerable markets like travel, dining, and hospitality – industries that got rocked during the hardest days of the pandemic.

Outside of Google and Facebook, the most notable publicly traded competitors in programmatic digital advertising are The Trade Desk (TTD) and Viant (DSP). The former is a $22 billion company that earned more than $137 million on $1.2 billion in revenue last year. The latter is a small-cap company worth barely $300 million, despite revenue of more than $224 million in 2021. However, both companies saw earnings drop significantly between 2020 and 2021, while AdTheorent more than tripled profits. Generally, we’re more interested around revenue growth than profit, but the fact that AdTheorent posted profits while the others lost ground is an interesting development.

Should You Buy AdTheorent Stock?

One of our basic rules for investing is to pick the leading company in its market. That’s what we’ve done in emerging tech industries like geospatial intelligence and telehealth. While AdTheorent is obviously building momentum, The Trade Desk appears to be the leading pure-play company in programmatic digital advertising, with a market cap currently more than 30 times bigger than AdTheorent. In fact, we never invest in companies with a market cap below $1 billion, so AdTheorent is off the table until it consistently remains above that threshold. 

The bigger question is whether we are even interested in pursuing a pure-play digital ad company in the first place. The total addressable market (TAM) is certainly huge:

Digital advertising revenue.
Credit: eMarketer

However, our mission is to educate and inform investors about disruptive technologies. The only thing disruptive about digital ads is how intrusive they are to our day-to-day browsing. While this theme might appeal to some investors, we’re probably going to close the book on this one unless our readers tell us otherwise.

Conclusion

AdTheorent is the rare post-SPAC stock that appears to offer good value to retail investors interested in the digital ad theme. Our simple valuation ratio shows that AdTheorent stock is quite fairly valued at a score of about 4, with anything 40 or higher considered overpriced. (In comparison, The Trade Desk scores about 18.) Be warned, though, that AdTheorent has some customer concentration risk, with a quarter of all revenues coming from just five customers. It is very U.S.-centric, as well, though there are plans to expand overseas at some point. In addition, size does matter, and AdTheorent remains a small fish among some very big predators.

Share

Leave a Reply

Your email address will not be published.