Description

Rapidly Growing Supply-Side Programmatic Leader Well-Positioned For a Shifting Landscape; Potential Google Adtech Alternative

1) Thesis Description

Magnite (MGNI) is the largest, independent provider of technology solutions to automate the purchase and sale of digital advertising inventory on behalf of publishers, also known as a programmatic Supply-Side Platform (SSP). Magnite is the merger between The Rubicon Project (RUBI) and Telaria (TLRA) in ’20 and importantly, purchased competitor SpotX in early ’21. The adtech industry is undergoing a paradigm shift on multiple fronts: secular growth in digital ad spending and programmatic transactions, the rise of connected TV’s (CTV) accelerated by the pandemic, accelerating industry maturation/consolidation, increasing privacy regulations, Google’s elimination of third-party cookies, and industry demand for an alternative to the Walled Gardens (GOOGL, FB, AMZN, AAPL) for ad revenue. The net effect for independent SSPs should amount to an improved market structure with greater economic rent than has occurred in the past. As such, Magnite is a well-positioned, underappreciated beneficiary of industry transformation, and as the company executes on its’ advantages over the next few years this should become apparent in growing market share, accelerating revenue growth, and expanding margins.

The thesis is as follows:

1)      At the current price of $28.56/share, the market assumes two-thirds of the business is worthless and the CTV business (growth over 40% per annum, ~35% EBITDA margins) is valued at ~7.25x EV/Revenue or ~17x EV/EBITDA. Valuation ranges from $30/share to $505.50/share over five years, with scenarios ranging from CTV-only to the company being one of the only scaled competitors against a ‘broken-up’ Google adtech division.

2)      An opportunity is available due to several reasons:

a.       Magnite’s standing in the industry inappropriately questioned as many advertisers/publishers wish to maintain open internet viability against the Walled Gardens. Global digital ad spending has grown to ~50% of advertisers total ad budgets (~$575B) in ’21, roughly a ~$290B industry. Of that, ~60% is channeled to the Walled Gardens, leaving ~40% or ~$115B in advertising dollars for the open internet. However, consumers spend ~33% of their time inside the Walled Gardens and the remaining ~67% on the open internet. The Walled Gardens are the simplest channel to reach and target a large consumer base, which is why they command a majority share. Meanwhile the open internet adtech industry, historically, had hundreds of players with minimal scale, lack of transparency and less effective targeting means. But with this misalignment of consumer time between adtech ecosystems, advertisers do not want to direct the entirety of their digital ad budgets to the Walled Gardens and miss valuable consumer attention on the open internet. Additionally, publishers do not want to be entirely dependent on the Walled Gardens for their ad revenue either. In all, advertisers and publishers have a vested interest in supporting the open internet adtech industry despite the dominance of the Walled Gardens, reinforcing Magnite’s long-term viability as an independent entity.

b.       Privacy regulations and elimination of third-party cookies mistakenly interpreted as permanently detrimental; in actuality, long-term benefits/barriers to entry accrue to those with scale and access to first-party data. The ’18 implementation of the General Data Protection Regulation (GDPR) in the EU and the California Consumer Privacy Act (CCPA) in ’20 as well as other privacy initiatives by tech companies, introduced considerable anxiety around the end-state of the adtech industry. However, through each of these instances, near-term volatility gave way to creating long-term benefits for scaled companies as compliance for smaller adtech firms was too costly/complicated and new entrants then had to contend with greater barriers to entry as surviving entities obtained greater market share through the industry consolidation. Events should occur similarly when Apple implements Identifier for Advertiser (IDFA) opt-in/out in ’21 and Google eliminates third-party cookies in ’22. This is key as many in the open internet utilize third-party cookies/unique identifiers in an effort to scale their targeting/measurement efforts. In this case, benefits will accrue not just to the largest firms but to those with direct access to first-party data (SSPs with their publisher relationships) or those creating an independent universal identifier (TradeDesk’s UID 2.0) backed by a scaled network of advertisers/publishers/adtech companies. The open internet has been preparing for the deprecation of third-party cookies for two years, though there may be a few solutions before the industry coalesces around a standard. Channels checks show that almost no advertisers are planning to cut their allocation to programmatic/digital advertising next year. Roughly 20% of MGNI’s revenue already utilizes first-party data displacing two-thirds of the company’s ~30% revenue exposure to cookies. Importantly, ~35% of revenue is derived from CTV, which is not subject to cookie deprecation.  

c.        Concerns about decreasing value of ad inventory and cyclicality of ad industry disregards advertiser budget dynamics and the secular growth in digital ads/CTV/programmatic. Underpinning privacy fears is that with less data there is less targeting capability, therefore rendering ad inventory less valuable on a CPM basis. However, advertisers do not base their ad budgets on CPMs, only a fixed dollar amount. So as ad inventory cost per thousands (CPM’s) drop due to less targeting data, advertisers, who need exposure to the open internet, simply purchase more ads for the same allocated budget resulting in no/minimal change in annual revenue to adtechs. Importantly, through these privacy changes advertisers have continued to allocate more ad spending to digital channels, up from 20%-25% of total global ad budgets in ’12 to ~50% this year. Additionally, the allocation of those global digital ad budgets through programmatic pipes has increased from ~20% in ’12 to ~75% this year (U.S. ~90%). Lastly, the growth of the CTV industry is important to digital advertisers as it has not been dominated by the Walled Gardens and at ~$11.5B it has a long growth trajectory displacing the global linear TV ad spending of ~$150B in ’20. These digital trends have likely been accelerated due to the pandemic. Programmatic adtech firms with CTV exposure, like Magnite, benefit from two secular growth trends within another larger one and are resilient to CPM changes.

d.       The company’s future economic potential largely unrecognized as investors lag in assessing the company’s portfolio transformation and worry about the capital structure. In ’17, when Magnite was a turnaround (Rubicon Project), the company’s ~$155M revenue exposure comprised of ~60% desktop (5%+ per annum growth)/~40% mobile (30%+ per annum growth) and was disadvantaged in a new format of ad auctions, header bidding. Since then, the new CEO, Michael Barrett, skillfully repositioned the company, regaining market share under the header bidding auction format and flipped the revenue exposure to ~60% mobile/~40% desktop. Then in late ’19 Rubicon Project and Telaria announced a merger to become Magnite, allowing RUBI to gain considerable exposure to a nascent, fast growing CTV ad format (~20% of MGNI revenue, 40%-50% per annum growth) and giving TLRA the scale it desperately needed. Then in early ’21 Magnite acquired SpotX, a competitor with ~60% CTV revenue exposure, for ~$1.2B incurring ~$650M in debt at a ~10x EV/Revenue valuation. Now MGNI is ~45% mobile, ~35% CTV and ~25% desktop, and the largest SSP outside of the Walled Gardens, a rapid change in just a few years. Ultimately, ’22 revenue should increase well above $500M with long-term growth accelerating to 25%-30% per annum and EBITDA margins expanding over 35%, above the current market expectations, allowing MGNI to lower its net-debt-to-EBITDA from over 4x in ’21 to ~2x by ’22.

e.       Competition and pricing pressure concerns fail to give credit to the consolidation on the SSP side that has already occurred, and the acceleration expected in the future. The proliferation of adtech firms in the early ‘10s meant that advertisers and publishers interacted with hundreds of different companies amongst a lack of transparency across the tech stack and plenty of ad fraud. Industry participants indicated 5-10 SSP vendors would be sufficient to provide enough coverage to properly service ad inventory. Coinciding with this industry consolidation, take-rates stabilized, and margins expanded. Consolidation was further spurred on by privacy regulations. At present, there are roughly nine independent SSPs of which Magnite is the largest. The only SSPs bigger are within the Walled Gardens, Google Ad Manager (GAM) and Amazon Publisher Services.  Notably, in the CTV space there are five SSPs, and only two are independent. On the more matured DSP side of the market, four independent adtechs cover ~90% of the market. The latest actions by Google/Apple and Magnite’s leading position should accelerate consolidation of the remaining SSPs, further strengthening industry stability.

f.         SSPs no longer commoditized providers due to industry changes and deprecation of third-party cookies. With almost 100 SSPs in the ‘10s, differentiation was practically indistinguishable. But as the industry consolidated, SSPs were able to leverage their scale with value-added software that simplified industry complexity and increased customer retention. Since ’18, Magnite has not lost one customer. In addition, the adtech structure of the CTV market exhibits higher customer switching costs and a larger importance on value-added software development. Going forward, the deprecation of third-party cookies should accelerate SSPs standing in the industry as they are the gatekeepers of publisher’s first-party customer data. While third-party cookies were industry standard for targeting, their usefulness was questionable, in fact the use of first-party data has shown an increase in targeting efficiency and therefore CPMs as well. However, only those with large scale first-party data are expected to benefit. In Magnite’s case these changes could result in the company’s ability to increase its take rate from 12%-15% to almost 20%, closer to adtech firms that currently operate in a more consolidated market selling additional software services backed by a competitive advantage in customer data.

g.        Hidden asset within the recent SpotX acquisition, SpringServe, not only unpriced equity value but indicative of Magnite’s future industry aspirations well beyond current market narrative. The transformative effects from Magnite’s acquisition of SpotX are only starting to be absorbed by the market but SpotX’s investment in SpringServe point to the creation of something more valuable that is not well understood by market participants. SpringServe is a private ad server focused on CTV and all forms of video, it is estimated that SpotX, and now Magnite, owns ~50% with a right to purchase the remaining interest. Industry experts believe that SpringServe has superior technology and is growing rapidly, now roughly one-fifth the size of Comcast’s Freewheel, a similar CTV/video ad platform (predominantly an ad server). If MGNI channels its CTV/video inventory through SpringServe, the ad servers’ size could expand to ~50% of Freewheels’. With Freewheel valued over $4B, MGNI’s investment in SpringServe could amount to $450M-$1,100M or $3.50-$8.50/share. More importantly, with SpringServe, Magnite should have a full tech stack to offer publishers, something highly desirable to industry participants, which other independent adtechs have been unable to create. This should allow MGNI to further expand its leadership position in the open internet, displace Freewheel in CTV/video, while setting the stage to provide a legitimate alternative to Google, who command almost 40% of the digital advertising market.

h.       Antitrust actions against the Walled Gardens increasingly probable and embeds substantial value option in Magnite. For the last few years, politicians of all types have increasingly come out against the tech giant’s anti-competitive behavior, coinciding with the public’s growing negativity towards the platforms. The current and previous administrations have shown the will to bring minor legal action against the tech giants, while also laying the groundwork for possible large scale anti-trust action. This possibility is a likely reason the Walled Gardens have not gone into the CTV market or completely displaced the open internet adtech industry. While it would be inadvisable to base a thesis on politicians doing the right thing, a large swath of the digital ad market could rapidly aggregate to the largest independent adtech firms on the off chance the government separates Google. This low probability event coupled with Magnite’s industry positioning as a Google alternative amounts to a substantial value option currently priced at zero.

3)      The largest micro risks in the name are development issues with CTV expansion, M&A integration issues and capital structure concerns. The largest macro risks are ad budget allocations to digital/programmatic channels, Walled Garden share shift and changing regulations.

2) Business Analysis

Before detailing Magnite’s relevant history, an explanation of the programmatic adtech environment is necessary to eliminate misconceptions as the industry is complex. This is not a comprehensive description

Programmatic Advertising Structure

Source: Moloco.

Demand-Side Platforms (DSP) help maximize an advertiser’s budget by effectively directing their ads to consumers who are likely to purchase or interact with said ad/product. A DSP has access to the advertiser’s ad campaign through a Data Management Platform (DMP). Supply-Side Platforms (SSP) help maximize a publisher’s revenue stream through the sale of ads. A SSP utilizes its own DMP for user information such as audience segment, social data, purchasing records and other data relevant for targeting. Both advertisers and publishers interact with multiple DSPs and SSPs to maximize impression views and monetize 100% of available ad inventory.

Ad purchases and sales are conducted through an Ad Network or an Ad Exchange. Ad Networks are intermediaries that incorporate software into an app or website code that deploy ads on request, these networks are used to a lesser degree as they are not programmatic. Ad Exchanges are online marketplaces where supply and demand partners can buy or sell programmatically within a fraction of a second, Real Time Bidding (RTB). These Ad Exchanges can also manage ad inventory directly. Daily transactions are in the billions and conducted on-premise, in the cloud (or both). These can be done through RTB ad auctions (second price), private marketplaces or directly. An Ad Server (not shown above) both stores ads/user data and delivers the ad to the consumer once the winning bid comes from the Ad Exchange/Network.

Revenue for adtechs is generated when an ad bid is won through their platform or a flat fee is charged as a % of total ad spending through the system (take rate = fees generated/advertiser spending). DSPs/SSPs are essentially the inventory/payment processors of the digital advertising industry.

Advertiser/publisher evolution in digital advertising started with direct deals, then managed services until they were conditioned to conduct full self-service programmatic through DSPs/SSPs as pipelines/rails.

Switching costs for DSPs/SSPs are essentially the same, though the DSP side of the market consolidated quicker to four players vs. the SSP side, which is just now at roughly nine players, giving a false signal the DSP side is superior as customers had fewer places to go if they switched. In terms of customers and data load, DSPs contend with hundred’s ad buyers, while SSPs manage hundreds of thousands of publishers.

DSPs and SSPs cannot disintermediate each other as both function as a broker for their respective party and each party has different goals (buyers maximize ad budget vs. sellers maximize ad revenue). On the open internet, this means that TradeDesk (TTD) and Magnite are tied at the hip, though merging the two entities would be ill-advised. The Walled Gardens have DSPs and SSPs, which is an inherent conflict of interest, and advertisers/publishers use them begrudgingly because of their data and scale advantages. Other adtechs with DSPs/SSPs under one roof have had issues with customer adoption for similar reasons.

The value of an ad is based on how effectively it can be directed towards a consumer, priced on a cost per thousand basis (CPM), and its format (CTV, video, audio and display), which is also subject to supply/demand dynamics. Take rates also vary between formats.

A Brief History – Turnaround Sub-Scale SSP Rapidly Transformed into an Industry Leader

The Rubicon Project, founded in ’07 by Frank Addante, Duc Chau, Craig Roah, and Julie Mattern, was a provider of automated ad exchanges and a supply-side platform. Originally, the fee structure consisted of both buyer and seller fees. However due to header bidding competition and demand for increased transparency in ’17, the firm now only generates revenue from seller’s fees and software sales of its Demand Manager (<5% of revenue).

In ’17, the company’s ~$155M revenue exposure comprised of ~60% desktop/~40% mobile and was disadvantaged in a new format of ad auctions, header bidding. Prior to header bidding, waterfall bidding was the industry standard which highly benefited Google’s AdX (now under GAM). Rubicon was second in line in the waterfall with the rest of the SSPs further down. The shift to header bidding was an open source initiative by publishers and advertisers to reduce dependency on Google. New management was hired in ’17 and repositioned the company closer to its original standing during the waterfall era.

In late ’19, Rubicon Project and Telaria announced a merger to become Magnite. Telaria was originally Tremor Video, a video focused adtech firm, which renamed itself following the sale of its DSP in ’17. The merger allowed RUBI to gain considerable exposure to the nascent, fast growing CTV ad format, while giving TLRA the scale it desperately needed. Notably, the company closed the merger just as the COVID pandemic erupted.

Then in early ’21 Magnite acquired SpotX, a competitor with ~60%/~40% CTV/video revenue exposure for ~$1.2B at a ~10x EV/Revenue valuation. SpotX was a U.S. subsidiary of the European-based RTL Group and was essentially a stranded asset within the parent company. Included in the transaction with SpotX was an estimated ~50% investment in a CTV/video ad server, SpringServe.

Following the transaction, MGNI’s revenue exposure is approximately 45% mobile, ~35% CTV and ~25% desktop, and is now the largest SSP outside of the Walled Gardens. Roughly 30% of revenue is from international sources.

Magnite has three segments: CTV, Mobile and Desktop and it provides programmatic solutions for almost every digital content format (CTV, video, audio, display).

Connected TV (CTV)

The company’s CTV segment provides software solutions and programmatic services to streaming services that partly fund their operation through ads. This may include digital-only firms (Pluto, Roku, Tubi, etc.), traditional media creators moving into the digital space (Disney+, Discovery+, Peacock, etc.) and even TV OEMs (Visio, Samsung, etc.).

Mobile

Magnite’s Mobile segment offers solutions for advertising on mobile web browsers, and mobile apps of all kinds that monetize operations/content with ad revenue.  This can be in several formats, display, video and audio.

Desktop

The company’s Desktop segment provides programmatic solutions that go through desktops, predominantly web browsers or applications accessed through the browser.

Management History –Top-Tier Executives and Industry Talent, Modest Insider Ownership (~5.5%)

In early ’17, Rubicon Project hired Michael Barrett, who was known for flipping ad tech companies. Barrett sold Millennial Media to AOL in ’15 for ~$240M and sold AdMeld to Google in ’11 for ~$400M. He was also an investor in several ad tech companies including HookLogic, which was acquired by Criteo for ~$250M.

Barrett was brought into Millennial Media (mobile ad network and DSP) in ’14 to help right the ship. Interestingly, his pay package was entirely structured around a sale of the company (at any sale price), worth ~$10M. The company’s mobile ad business was being beat out by Facebook/Google, as programmatic displaced direct ad sales. Almost all of the key senior management and founders were gone by ’14 (Company IPO in ’12). Upon becoming CEO, Barrett reorganized the company and improved operating performance considerably. However, the company was sold for a valuation of ~$240M, below the ~$600M market cap when Barrett took the helm. This is likely a win, considering the firm wasn’t well run and was probably going to zero without outside intervention.

At the time Mr. Barrett came onboard Rubicon, his pay package comprised of ~90% stock compensation, ~10% salary. The stock compensation was contingent on improvement in Revenue and EBITDA, clearly incentivizing his efforts on the turnaround. Notably, he purchased on a considerable sum of shares in ’18 in the ~$3.50/share range.

Since then Mr. Barrett’s shrewd capital allocation has placed the company, now Magnite, in a tier of its own. At present, Mr. Barrett’s compensation has shifted to ~75% equity based (~$2.55M) and ~25% salary/cash bonus (~$0.95M).

The CFO, David Day, was formerly the CAO of Rubicon Project from ’13 until his promotion to CFO in ’16. His compensation is ~60% equity (~$1.1M) and ~40% salary/cash bonus (~$0.7M).

The CTO, Tom Kershaw, was a Director of Project Management at Google until ’16, when he joined Rubicon Project. His compensation is ~60% equity (~$1.1M) and ~40% salary/cash bonus (~$0.8M).

Importantly, the executive teams’ equity compensation is based on Adjusted EBITDA less Capex (~50%), Revenue (~35%) and CTV Revenue (~15%) performance metrics.

Additionally, the engineering teams at the company are considered top rate and MGNI has the largest concentration of CTV engineers in the industry (over 300 people).

The executive team and insiders own ~5.5% of MGNI.

Customer Dynamics – Publishers Continue to Monetize Content with Ads, Desire Simplicity and Scale, Rising Demand for Walled Garden Alternative

The company has thousands of media companies and over 100K websites as clients, which it can match to almost a million brands, agencies, and DSPs.  The earlier years of the adtech industry were rife with ‘bad actors’, mis-payments, fraud, etc.; generating a certain level of mistrust. Customers demanded greater transparency and accountability, which Magnite provided, thus garnering a high level of trust with high quality publishers.

The primary form of funding for websites, apps and other media creators is to sell ad space. This method of monetization has not changed despite ad blocking, privacy regulations and the rise of subscription models.

Advertisers budget on an annual basis, and allocate fixed dollar amounts to certain ad channels to reach their customers as effectively as possible. This results in adtech revenue seasonality, smaller spend levels in 1/2Q with ad buyers then spending the majority of their budget in 3/4Q. Budgets are not based on CPMs, when CPMs drop advertisers simply buy more ad inventory until the dollar amount allocated is fully spent.

Publishers and advertisers are aware consumers spend ~67% of their time on the open internet and ~33% inside the Walled Gardens. Advertisers do not want to direct the entirety of their digital ad budgets to the Walled Gardens and miss valuable consumer attention on the open internet and publishers do not want to be entirely dependent on the Walled Gardens for their ad revenue. This is most pronounced with CTV companies, who don’t want to sell ad inventory through the Walled Gardens and pseudo Walled Gardens (Comcast, used to include AT&T/Verizon) since they have services that are their primary form of competition. Further, the black box, non-transparent nature of the Walled Gardens create hesitancy with advertisers/publishers, who are seeking the opposite.

Despite the dominance of the Walled Gardens, advertisers and publishers have a vested interest in supporting the open internet and direct a considerable amount, though not the majority, of their ad spending to the independent adtech industry. The lack of scale for independent adtechs means that targeting is less effective, and revenue generated is less than if the transactions occurred through the Walled Gardens, a dynamic advertisers and publishers understand/accept. An independent SSP competitor to Google would be in high demand as they could provide advertiser’s open internet ad access at scale, while maximizing a publisher’s revenue transparently. Additionally, the Walled Gardens continue to increase their prices on advertisers/publishers every year, increasing demand for an open internet alternative to relieve cost pressures.

To date, there has not been an independent full stack adtech firm with enough scale to compete with Google, though Magnite could, should they tuck-in the remaining interest in SpringServe.

Supplier Dynamics – Unique Software Assets Minimize Data/Input Costs

Magnite’s suppliers are data server/analytic firms (cloud, on-premise providers), as the company has no true inputs other the data associated with its customers ad inventory/first-party data.  Given that the company handles trillions of ad impressions a month, data costs can be exorbitant if not handled properly.

In mid-’17, Rubicon acquired nToggle, an AI network traffic shaping tool, for ~$39M to address inventory management issues associated with header bidding. It analyzes auction traffic data to shape real-time bidding requests, optimize traffic, and reduce the number of duplicate and irrelevant bid requests DSPs must process.

This tool lowers the capital spending/operating cost of executing server-side auctions and streamlines the customer experience, creating substantive capex/opex scale advantages as Magnite grows.

Competitor Dynamics – Privacy Regulations Accelerating Naturally Occurring Consolidation, Converging on Mature Industry Structure; Pseudo Walled Gardens Being Disintermediated, Walled Gardens Under Increasing Pressure

In the early ‘10s, advertisers and publishers interacted with hundreds of different open internet adtechs. Industry participants back in ’17/’18 indicated 5-10 SSP vendors would be sufficient to provide enough coverage to properly service ad inventory, and as such started channeling ad inventory to the largest players. Consolidation was further spurred on by privacy regulations, raising costs for compliance, pushing out smaller players and keeping new entrants out due to larger incumbents. Coinciding with this industry consolidation, take-rates have stabilized, and margins have expanded.

At present, there are roughly nine independent SSPs (Magnite, OpenX, Index Exchange, TripleLift, Xandr, PubMatic, MoPub, Sovrn, and Freewheel) of which Magnite is the largest. Consolidation should accelerate following the actions of Apple/Google, as well as Magnite utilizing its scale to dictate actions onto the market.

With further consolidation, Magnite may be in a position to eventually increase its’ take rate from 12%-15% to almost 20% as the elimination of third-party cookies makes publishers first-party data highly coveted, and there should be no other independent SSP large enough to provide scaled access as well as unique software offerings tailored to said data. In the current third-party cookie era, Trade Desk has a take rate closer to 20% as they up-sell third-party data aggregation marketplaces and develop software to manage the data. Further, Googles’ SSP take rate is ~20%.

In the CTV space there are five SSPs (Magnite, Freewheel, Google, Xandr, and PubMatic) and only Magnite and PubMatic are independent. Magnite has access to ~90% of ad-supported CTV’s in the U.S. PubMatic has a small, inferior offering targeting smaller publishers in display, video, audio (who may be highly disadvantaged by the ’22 third-party cookie deprecation) and is attempting to apply header bidding technology to CTV, unsuccessfully. Additionally, PUBM is competing on price, and Magnite is not countering their actions. Google’s CTV offering is merely a placeholder, Xandr continues to have issues under AT&T management and industry participants uncomfortably use Freewheel (Comcast) simply because of their scale. Notably, Freewheel has 50%-60% market share in the CTV space.

The success of the Walled Gardens has brought fast followers (Comcast, AT&T, and Verizon) attempting to create similar walled gardens with matching tech stacks. To date, AT&T is now breaking up its pseudo walled garden by separating its media assets and merging them with Discovery (a MGNI CTV client), while keeping its Xandr DSP and SSP. Verizon sold its Yahoo/AOL assets (including its DSP/SSP) to Apollo. Across these companies’ numerous data points show that the firms had issues filling the majority of their digital ad inventory, either due to inferior tech or customer hesitancy stemming from conflicts of interest (both a DSP/SSP, CTV competition). These entities, under new management, are expected to expand their adtech relationships beyond their own internal tech stacks, like other smaller media entities, employing multiple independent DSPs and SSPs.

Approximately 60% of global digital advertising dollars are directed to the Walled Gardens (Google, Facebook, Amazon, and Apple) and competition against them from the open internet players has been non-existent, to date. With the completion of the SpotX transaction, Magnite is now the largest independent omnichannel SSP and best positioned to offer a legitimate alternative to Google.

Lastly, industry participants are observing moderating aggressiveness in the marketplace from the Walled Gardens believed to be due to the increasing political pressure/scrutiny. Interestingly, Google’s deprecation of third-party cookies, ostensibly for consumer privacy, increases the difficulty for open internet adtech players to target/measure audiences, potentially exacerbating anti-trust pressures as a second order effect.

Market Trends – Secular Growth in Digital Advertising/Programmatic, CTV Displacing Linear TV Ad Spend

Advertising Budgets

Total global ad budgets should grow ~6.5% this year to ~$575B, according to dentsu. Total global ad spending is expected to grow ~7% in ’22 to ~$620B driven by travel (+14.5%), media/entertainment (+13.5%) followed by technology (+6%) and retail (+5.5%) advertising.

According to eMarketer, advertisers have increased ad spending through digital channels, up from 20%-25% of total global ad budgets in ’12 to ~50% this year, ~$290B (~8% growth y/y). The open internet should amount to ~40% digital ad spending or ~$115B. Digital ad spending is expected to increase to 65%-70% of total ad budgets in the next few years as consumer attention increases on digital channels, per MAGNA. The formats growing the fastest within the digital ad space, outside of CTV, are mobile video/audio/display (+20%), video (+15%) and audio (~8%).

Additionally, programmatic market share of global digital ad spending has increased from ~20% in ’12 to ~75% this year. In the U.S., approximately 90% of digital ad spending is transacted programmatically, and the rest of the world is expected to exhibit a similar composition long-term.  

Lastly, CTV ad spending should increase ~40% this year to ~$11.5B of which ~60% is conducted programmatically, according to eMarketer.  Global linear TV ad spending decreased ~12% y/y to ~$150B in ’20 and is expected to continue to decrease as advertisers shift to CTV format, per eMarketer.  Last year, agencies/advertisers earmarked ~5% of their budgets for CTV as a hedge against linear. In ’21, the same agencies/advertisers plan to allocate 30%-40% of their budget to CTV.

Over 30 million U.S. households have ‘cut-the-cord’ (canceled a pay TV service and continued without it) as of YE’20, which should increase to ~50% of all U.S. households (~65M) by the end of ’24, according to eMarketer. Additionally, the largest streaming applications have historically monetized via subscription with several potentially offering cheaper subscriptions partly funded by ads, and nascent CTV sellers with ad-supported models have only just begun to implement programmatic transactions given the technological complexity.

3) Why now?

Magnite was a former turnaround that has quickly developed into a supply-side programmatic market leader with exposure and competitive advantages in the fastest growing sections of the digital ad industry. The name reached an all-time high in early ’21 at ~$65/share, trading at ~18x EV/’22 Revenue, close to the normalized valuation multiple of DSP leader TradeDesk, and broke down on concerns over Google’s privacy actions as well as inflation (capital-light business, revenue grows along with nominal GDP).  At present, the name trades at ~8x EV/’22 Revenue and ~28.25x EV/’22 EBITDA based on consensus revenue and EBITDA of ~$510M and ~$145M (~28.5% margin) in ’22 with growth decelerating to ~16% per annum by ’23, below the company’s revenue guidance of ‘well above $500M’ in ’22 and long-term guidance of ~25% per annum revenue growth and 30%-35% EBITDA margins. If the company hits our expected ’22 revenue and EBITDA numbers, MGNI trades at ~7.5x EV/’22 Revenue and ~21x EV/’22 EBITDA, below the low end of the valuation range for a 25%-30% per annum revenue grower with EBITDA and cash flow margins inflecting upward. This is also well below Magnite’s sister company’s current valuation of ~18.75x EV/’22 Revenue and ~47.25x EV/’22 EBITDA, TradeDesk, which exhibits similar characteristics. We advocate entering into a position as the company’s growth and competitive advantages accelerate coming out of the industry changes occurring in ’21/’22, which may induce near-term volatility, and the market lags in catching up to the paradigm shift.  

A few key points below illustrate the company’s value potential at this point in time:

1)      Substantive Revenue Acceleration Beyond Digital Ad Budget Growth Driven by CTV/Video and Industry Consolidation: While overall digital ad spending should grow ~8% this year, mobile and video ads should grow ~20% and ~15% this year, respectively.  Additionally, CTV is expected to grow over 40% this year. In ’19 as the company shifted to faster growing formats and channels away from display and desktop, management increased their long-term annual revenue growth guidance to ~20%. With the recent acquisition of SpotX, almost two-thirds of Magnite’s revenue is derived from CTV (~35%) and video (~30%) formats, coupled with mobile channel exposure (video, display audio formats) of ~45% of revenue. As such, long-term annual revenue growth guidance was increased to ~25% in May ’21.  However, as CTV expands its revenue share, long-term revenue growth guidance is expected to be increased to ~30%, more closely in-line to TradeDesk. In fact, management, known for being conservative, have hinted at increasing their annual growth guidance to ~30% in as early as a few months from now when they host a SpotX Analyst Day. Additionally, these growth figures do not assume market share gains from the ongoing industry consolidation, known as Supply Path Optimization (SPO) or possible take rate increases. As detailed above, privacy changes/cookie deprecation may cause near-term volatility but channels checks show that almost no advertisers are planning to cut their allocation to programmatic/digital advertising next year. Going forward, we anticipate Magnite grows organically ~28% per annum through ’26, below the expected ~30% guidance, and roughly in-line with TradeDesk’s expectations.

2)      Considerable Margin Expansion from Impressive Operational Leverage; Possibility to Increase Take Rates: Magnite’s unique business model with a high percentage of fixed operating costs allows it to generate significant incremental EBITDA margins. In its growth phase during the waterfall bidding era, the firm was much more lax on operational efficiency and yet still generated ~65% incremental EBITDA margins, reaching ~26.5% EBITDA margins in ’16. After Michael Barrett and team streamlined operations and acquired nToggle in ’17, incremental EBITDA margins increased to 75%-80% by ’19. Exiting its turnaround, EBITDA margin guidance increased from 20%-25% to ~30% in ’20. At present, the company is conservatively guiding to 30%-35% EBITDA margins. With the industry consolidating and shifting to first-party data, MGNI could be in a position to raise take rates outright or maintain them and add software sales to amplify revenue per customer. Magnite should continue to expand margins from ~30% in ’21 to 40%+ by ’25 assuming a subdued ~50% incremental EBITDA margin (similar to TradeDesk, though well below MGNI’s capability) and assuming no increase in take rates/incremental sales from software.

3)      Inflection Point in Free Cash Flow Generation Driven by Capex Scale Underpinned by Unique Software Assets; Deleveraging of Capital Structure: Following the integration of SpotX, Magnite should be in a position to expand EBITDA margins beyond the current 30%-35%. The company’s large working capital balances are in actuality the payments between the DSPs and publishers, and over time net to zero cash flow impact as MGNI is essentially a pass-through entity. Capital spending has decreased from ~26% of revenue in ’17 to ~8% this year partly due to scale and MGNI’s nToggle asset. Should these growth/scale relationships hold, the company’s capex as a % of revenue should decrease from ~8% this year to less than 4% by ’25. With these influences, free cash flow margins should rapidly increase from breakeven in ’19 to 35%+ by ’25. Importantly, unlike other small, fast growing tech companies, roughly 50% of Magnite’s executive equity compensation is contingent on hitting Adjusted EBITDA less Capex targets. Lastly, the ’21 acquisition of SpotX was partly financed with ~$650M in debt, putting net-debt-to-EBITDA above 4x, and the substantive cash flow generation over the next year should reduce net-debt-to-EBITDA closer to 2x by ’22.

4)      Value Creating Management and Concentrated Industry Talent to Compound Advantages: At the time Mr. Barrett came to Rubicon, the company was valued at less than cash value (sub-$100M market cap), with revenue of ~$155M and exposure to weaker sections of the market. Focus on execution and conservative guidance resulted in a solid turnaround as revenue and margins inflected upward (RUBI ~$500M market cap). With CTV rapidly becoming a dominant ad channel, free of Walled Garden and privacy pressures, the management team merged with Telaria in ’20, and expanded their CTV exposure with the acquisition of SpotX (MGNI ~$3,500M market cap) in ’21. Further, as the largest independent SSP, Magnite is now able to compound its advantage against competitors given the largest concentration of leading talent in the adtech industry, outside of Google. As such, the company is well-positioned due to strategic consolidation of the best independent SSPs, and should conservatively generate ~30% revenue growth/30%-35% EBITDA margins going forward, with upside potential due to the capital allocation skills of management.

5)      Further Value Creation with the Possible Acquisition of Remaining Interest in SpringServe; Compete with Freewheel and Google: We anticipate Magnite purchases the remaining interest in SpringServe, the hidden CTV/video ad server asset within the SpotX transaction. With Freewheel valued over $4B, MGNI’s investment in SpringServe should be valued today at ~$450M (~$3.50/share). If MGNI purchases the entirety of the business then channels its CTV/video inventory through the asset, the ad servers’ size could expand to ~50% of Freewheels’, valuing SpringServe at ~$2,000M (~$17/share). More importantly, with SpringServe, Magnite should have a superior full tech stack to offer publishers, something highly desirable to industry participants, which other independent adtechs have been unable to create. In effect, Magnite would then be able to displace Comcast’s Freewheel and provide the market a legitimate alternative to Google on the SSP side.

4) Concerns/Thesis Pressure Points

Advertising Budget Risk

The growth in ad budgets is contingent on GDP growth. Given Magnite’s operating leverage, changes in the level of ad spend going through digital channels can have an outsized impact on margins and cash flows. However, digital ad budgets are less cyclical as consumer attention continues to move from the physical to the digital world. This dynamic has likely been accelerated by the pandemic, particularly for CTV.

Walled Garden Anti-Competitive Behavior Against Open Internet

With ~60% share of the digital advertising market, the Walled Gardens are the first choice for digital advertising and their actions reverberate through the open internet adtech firms. Recent privacy initiatives by the Walled Gardens may pressure the open internet players in the near-term, while inviting increased political scrutiny. Advertisers and publishers seeking transparency continue to be supportive of the open internet, while politicians/consumers increasingly advocate breaking up these tech giants, which may keep them from being overly aggressive.

Inflationary Environment Risk

With interest rates inflecting upwards as inflation and inflation expectations rise, the impacts to MGNI should be negligible to modest. Digital ad spending is more secular and driven by nominal GDP and the business itself is capital-light with the possible ability to raise costs (take rates) over time. However, variable interest rate debt ($250M-$300M) may cause the company’s interest costs to rise until the balance is paid down.

Capital Structure Risk

The company incurred ~$650M in debt when it acquired SpotX in early ’21, which raised its net-debt-to-EBITDA to over 4x and introduced a variable interest-bearing term loan. MGNI recognizes this debt structure concern and plans to pay down its term loan promptly, lowering net-debt-to-EBITDA to ~2x by ’22 as its revenue grows/cash flow margins expand as well.

Competition Risk

The SSP side of the adtech industry is less consolidated than the DSP side and competition, like PubMatic, are still trying to compete on price. As the largest SSP, we anticipate these risks to be mitigated going forward as consolidation accelerates.

5) Business Valuation

Magnite should exhibit almost 30% revenue growth per annum over a number of years as the CTV industry expands and the company exerts its influence as the largest independent SSP. MGNI generates revenue by charging fees on winning ad bids run through their platform on behalf of publishers (take rate = bid fees/total ad dollars spent), as well as modest software sales from Demand Manager.  

A summary of the Base Case assumptions for the company is below:

1)      Short-Term Guidance 2Q/21 Revenue: $92M-$96M, EBITDA margin: ~28%.

2)      Long-Term Guidance: Organic Revenue Growth: ~25% per annum (expected to be raised to ~30% this year), EBITDA Margins: 30%-35%.

3)      Total revenue should grow ~28% to ~$545M in ’22 and average ~28% growth per annum to ~$1,480M by ’26. This assumes no change in take rates (12%-15%) and no impact from consolidation.

4)      EBITDA margin of ~32% in ’21 should improve to ~45% by ’26, equivalent to a ~50% incremental EBITDA margin (below company’s 75%-80% ability).

5)      Over $700M in NOLs should be consumed by ’25, followed by a ~25% tax rate thereafter.

6)      Capex: ~8% of revenue in ’21 scaling down to ~3.75% by ’25.

7)      SpringServe value: ~$450M (~$4.25B Freewheel comp, ~20% of Freewheel, ~50% MGNI ownership)

8)      ~23.5x normalized EV/EBITDA multiple for Magnite. Reasoning behind the multiple is in the Peer Analysis section below.

9)      Discount rate at ~10% (large-cap).

Five-Year Operating Model

A simple five-year operating model is utilized to determine value.

Base Case:

Base Case assumes the company grows ~28% per annum from ’22 through ’26 and expands EBITDA margins to ~45% with MGNI’s interest in SpringServe valued at ~$450M (~$3.50/share).

-           Base Case Valuation: $78/share.

Upside Case:

Upside Case assumes the company grows ~35% per annum through ’26, generates a ~65% incremental EBITDA to achieve ~60% EBITDA margins by ’26 valued at ~32.5x EV/EBITDA with SpringServe fully acquired and equivalent to the size of Freewheel (~$4.25B, ~$34/share).

-           Upside Case Valuation: $196.75/share.

Downside Case:

Downside Case assumes the only remaining asset is the CTV segment growing 40% per annum valued at ~20x EV/EBITDA with MGNI’s interest in SpringServe valued at ~$175M (~$1.50/share).

-           Downside Case Valuation: $30/share.

Option Value Case:

Option Value Case assumes Magnite competes against a ‘broken-up’ Google and generates roughly half of Google’s open internet ad revenue (~$18B for MGNI) with 30%+ free cash flow margins growing at ~4.5% per annum with SpringServe fully acquired and equivalent to the size of Freewheel (~$4.25B, ~$34/share).

-           Option Value Case Valuation: $505.50/share.

A summary of the valuation cases is below ($):

Peer Analysis – Trading Comps

Magnite’s direct public comparables are the adtech companies PubMatic (SSP) and TradeDesk (DSP). However, PUBM is inferior to MGNI on several levels and while TTD is a comparable as the largest DSP, they operate in a more consolidated part of the market with better take rates.

Given the evolving nature of the adtech industry, we triangulate our adtech valuations by also comparing to three large scale payments companies, MasterCard (MA), PayPal (PYPL) and Visa (V). To contrast the similarities to the adtech industry, these payment companies operate in a highly consolidated marketplace, generate fees as a % of spending on their platform and are conduits for billions of consumer/business financial transactions. The payment firms have higher margins, switching costs and pricing power, but lower growth compared to the adtechs. The analysis below utilizes a normalized EV/EBITDA metric for the adtech and payment comparables to account for the margin implications of scale and market consolidation.

Public Comparables – EV/EBITDA

1)      PubMatic (PUBM) – ~20x normalized EV/EBITDA; currently at ~23x FTM.

2)      TradeDesk (TTD) – ~45x normalized EV/EBITDA; currently at ~47.25x FTM.

3)      MasterCard (MA) – ~22.5x normalized EV/EBITDA; currently at ~27.5x FTM

4)      PayPal (PYPL) – ~25x normalized EV/EBITDA; currently at ~31.5x FTM.

5)      Visa (V) – ~22.5x normalized EV/EBITDA; currently at ~25.5x FTM.

PubMatic is an SSP, which recently went public. The firm is sub-scale compared to MGNI and lacks true exposure to the CTV industry, while also under-investing in technology on their PnL. Notably, this is the only direct competitor utilized for our comparison analysis. Growth should average ~20% per annum over the next five years as they underperform compared to Magnite.

TradeDesk is a DSP and an industry leader in the open internet adtech space. TTD is MGNI’s biggest customer as they are counterparts in the industry, each representing either side of an ad transaction. The company is one of four DSPs that cover 90% of ad requests and its take rate is closer to ~20% given lower levels of competition and a highly developed third-party data marketplace/software offering. Growth should average ~28% per annum over the next five years even though the DSP side of the market is more mature than the SSP side.

MasterCard provides financial transaction processing services for credit/debit cards, electronic cash, ATMs, and checks. The company generates the majority of its revenue (~80%) from transaction processing and associated fees based on card balances. MA’s competitors of scale are Visa, American Express and Discover. Growth should average ~17% per annum over the next three-to-five years.

PayPal is a technology platform that enables digital and mobile payments for consumers and merchants Approximately 90% of its revenue is derived from processing fees (~13 billion transactions per annum). Growth should average ~20% per annum over the next three-to-five years as more consumers pay digitally.

Visa operates a retail electronic payments network processing ~550M transactions a day for credit/debit/prepaid cards for consumers, payroll government and healthcare. The company generates its revenue from transaction processing and associated fees based on card balances. V’s competitors of scale are MasterCard, American Express and Discover. Growth should average ~13% per annum over the next three-to-five years.

The payment groups’ normalized EV/EBITDA valuation multiple equates to ~23.5x and has ranged between 10x-40x from ’15 to ’20 and currently trades at ~28.25x FTM assuming ~17% per annum revenue growth. The faster growing (25%+ per annum) adtech group has a normalized ~32.5x EV/EBITDA valuation multiple, currently trading at ~35.25x FTM. Magnite currently trades at ~21x ’22 EV/EBITDA on our expectations.

Interestingly, our normalized ~23.5x EV/EBITDA valuation multiple equates to an implied ~11x EV/Revenue multiple for Magnite based on our expectations for revenue and EBITDA margins in ’26, below the adtech groups’ implied ~13.5x EV/Revenue valuation multiple.

Peer Analysis – EBITDA Margins

EBITDA Margins

1)      PubMatic (PUBM) – ~30% normalized EBITDA margin; currently at ~29% FTM.

2)      TradeDesk (TTD) – ~40% normalized EBITDA margin; currently at ~35% FTM.

3)      MasterCard (MA) – ~60% normalized EBITDA margin; currently at ~58% FTM

4)      PayPal (PYPL) – ~32.5% normalized EBITDA margin; currently at ~30% FTM.

5)      Visa (V) - ~70% normalized EBITDA margin; currently at ~68% FTM.

PubMatic’s operating margins are expected to modestly increase over the next five years as the company grows less effectively than Magnite. Capital expenditures should average ~6.5% of revenue.

TradeDesk’s operating margins have increased with a ~35% incremental EBITDA margin since ’15, which is expected to inflect upward to ~50% over the next five years given their scale and leading market position. Capital spending has averaged ~5% of revenue.

MasterCard’s operating margins have stabilized just short of 60% since ’15 and should minorly improve going forward. Capital expenditures have averaged ~5% of revenue.

PayPal’s operating margins have improved with a ~45% incremental EBITDA margin since ’15, though are expected to only moderately increase the next five years. Capital spending has averaged ~4% of revenue.

Visa’s operating margins have stabilized just short of ~70% since ’15 and should minorly improve next five years. Capital expenditures have averaged ~3.25% of revenue.

Magnite’s closest comparable based on business model and scale with similar absolute and incremental margins is TradeDesk, which generates ~40% normalized EBITDA margins with ~50% incremental margins. Magnite should generate more than ~35% EBITDA margins over time given conservative 50%+ incremental EBITDA margins (capable of 75%-80% incremental EBITDA margins).

Peer Analysis – Conclusion

For Magnite, we apply a ~23.5x EV/EBITDA normalized valuation multiple, in-line with the comparable group of payment companies for conservatism as the normalized ~32.5x EV/EBITDA multiple for the adtech group embeds higher expectations for growth/margin improvement/industry developments. Lastly, we did not include a strategic acquirer analysis as we anticipate Magnite being more valuable as an independent company and a consolidator as opposed to an acquisition target.

6) Market Expectations/Perceptions

Magnite is covered by seven analysts with an average price target of ~$50/share. Five analysts have a Buy rating and two have a Hold. Analysts give little credit to the company’s statements around future financial performance, despite management’s track record of conservative guidance that results in almost continuous beats/raises. Notwithstanding that, management is well regarded, and the CEO has been referred to as a living legend.

Notably on the 1Q/21 earnings call, management raised their long-term annual growth guidance from ~20% to ~25%, despite admitting they have not given full credit for effects stemming from the SpotX acquisition. Additionally, their 2Q/21 revenue guidance integrating SpotX basically assumed no incremental growth for both MGNI and SpotX, disappointing many market participants. However, the company acted similarly last year with their guidance for the first quarter as a combined RUBI/TLRA, they subsequently beat expectations. Lastly, the hidden asset SpringServe has gone unnoticed not only as far as equity value but also the long-term strategic implications for MGNI.

Consensus expects only ~$510M in revenue for ’22 and growth slowing to ~16% per annum by ’23, below company guidance of ‘well over $500M’ revenue in ’22 and ~25% long-term annual growth.  Sell side consensus for EBITDA margins is also anemic with forecasts below 30%.

Lastly, there is an overall fear surrounding Magnite’s competitive dynamics in the SSP market as the industry consolidates and shifts to first-party data, highlighting the opportunity for long-term investors as near-term volatility should give way to improved market dynamics/economics afterwards.

7) Downside Protection – Where’s the Margin of Safety?

The company’s downside is well protected as Magnite provides crucial services as the largest ad inventory aggregator/processor in the industry outside of Google.

At the $25/share, the market assumes two-thirds of the business is worthless and the remaining CTV segment (40%-50% growth, 35% EBITDA margins) is valued at ~6.5x EV/Revenue or ~15x EV/EBITDA.

Adjusting growth and margins to consensus estimates implies a normalized ~16.75x EV/EBITDA valuation for the business based on the current market price of $28.56/share. Which highlights the markets lagging recognition of the changes occurring at both the company and industry level.  

Notably, our normalized valuation multiple appears conservative compared to the representative adtech firms, creating the possibility for a higher multiple as the company proves out its revenue and EBITDA growth. While we do not apply TradeDesk’s normalized valuation multiple of ~45x EBITDA in any of our valuation cases, should the SSP side of the industry consolidate similarly and Magnite’s take rates increase with growth above 30% it may be warranted, valuing MGNI at $136.25/share on our Base Case (including ~$3.50/share for SpringServe).

8) Conclusion       

Magnite is a misunderstood, highly advantaged adtech firm run by a very competent management team, well-positioned for the changing industry landscape. The company’s market positioning as the largest SSP after Google with significant CTV/video/mobile exposure should enable it to grow faster and more profitably than it has in the past, above current market expectations, following the transition to first-party data.

The company’s valuation should inflect upward driven by strong revenue growth and expanding competitive advantages/financial performance as the open internet adtech industry matures/consolidates and Magnite develops into a potential Google adtech competitor.

As of May 21, 2021, the name is trading at $28.56/share. With a Base Case valuation of $78/share, we believe there is ~175% upside, equating to ~22% annualized IRR over five years.  

Catalyst

·         SpotX Analyst Day – increase in long-term revenue growth guidance

·         Revenue/EBITDA growth in excess of historical norms/expectations

·         Deprecation of third-party cookies

·         Industry consolidation

·         Acquisition of remaining interest in SpringServe

·         Break up of Google and other Walled Gardens