CDK Global (CDK) is a gem of a business that operates in a niche of the software industry known as Dealership Management System (DMS) software. To put it simply, CDK sells (licenses) mission critical software to retail auto dealerships that enables dealerships to run their operations.
CDK’s business benefits from durable competitive advantages that include a recurring revenue stream governed by long term contracts, resiliency to recession, high switching costs, slow industry life cycles and high operating leverage.
CDK sits in a coveted position with dominant market share among its industry peers but there are some business risks to keep on the radar screen. Key risks to the business include historical auto industry cyclicality, relatively low margins of the Digital Marketing Solutions segment, fragmentation of DMS providers and recent consolidation of franchised dealership groups. However, there are strong mitigants to most of these risks that provide comfort to this investor.
The recent purchase activity by hedge funds Sachem Head Capital Management and Fir Tree Partners has driven up CDK’s stock price significantly and it no longer trades at the discount to intrinsic value that it did post spin off from ADP. I believe that these funds may try to push for cost cutting strategies that can boost margins and bring operating margins more in line with their comps. Transformative acquisitions/mergers are unlikely in the near term due to tax implications.
Based on a DCF that assumes operating margins can meaningfully improve from where they are today (over 10 years) as CDK improves its cost structure, I value the company at ~$42/share versus a current share price of ~$46/share. I would wait for a more attractive entry point closer to and optimally below intrinsic value. Market volatility should hopefully provide this opportunity in the near to intermediate term. I am a buyer of this wonderful business when that opportunity presents itself.
In an earlier post a few months ago (this post is way overdue) I talked about digging deeper into a spin-off opportunity. From a field of several recent spin-off announcements, I decided that CDK Global (CDK) was the most compelling opportunity.
CDK creates software for the automotive dealership industry and was spun off from ADP (which focuses primarily on HR outsourcing and staffing) on October 1st. Well, a lot can happen in a few months and it has – two large hedge funds have taken sizable positions in the company since I first discussed CDK back in October. Sachem Head Capital Management (known to have an activist bent) now owns ~9.8% of the company and Fir Tree Partners (a value investing focused shop) now owns ~8.7% of the company for a combined ownership stake of ~18.5% between the two funds. The buying activity of these funds has driven the price of the stock up significantly over the last few months, but that shouldn’t stop us from analyzing this opportunity in the traditional Scuttlebutt Way.
CDK plays in a space that is relatively unknown (usually great hunting grounds for underpriced investments). It operates in a niche of the enterprise software industry that I like very much as a risk averse value investor. At its core, CDK provides operating software to retail automotive dealerships (think your local Smith Toyota dealership). The software often referred to as Dealer Management System (DMS) software can be likened to the central nervous system of an automotive dealership. CDK’s software automates and integrates all the critical workflow processes that happen at an auto dealership including advertising, sales, financing, insurance, parts and repair and maintenance of vehicles. The business is organized into three segments but I analyze the business primarily as two segments because two of the segments provide nearly the same solution but to different geographic markets.
CDK generates revenue from two key businesses: 1) Selling the aforementioned DMS software to dealerships in North America and International markets; 2) Selling marketing solutions (web development, hosting and ad placement) to dealerships. For reporting purposes, CDK reports the North America DMS and International DMS businesses separately but I will discuss them together here.
Automotive Retail Solutions ($1.6bn or ~80% of revenue) – North America and International
Automotive Retail Solutions focuses primarily on mission critical Dealer Management System (DMS) software that helps franchised automotive dealerships manage every component of their businesses from sales and inventory to maintenance and parts. CDK’s primary DMS platforms for dealerships are called CDK Dash and CDK Drive and are sold both in North America and in International markets. While both North America and International markets offer similar DMS solutions, they are organized as two separate reportable segments due to geographical focus. CDK provides solutions to approximately 26,000 retail locations and most Original Equipment Manufacturers (OEMs or car manufacturers like Toyota, Ford, etc.) worldwide. In North America specifically, CDK serves about 8,400 auto retail locations, which equates to about 40% market share. Furthermore, in the US, CDK’s clients include 7 of the top 10 largest automotive retailer groups.
More specifically, the automotive retail software solutions that are provided by CDK include: Dealer Management System (DMS): In many cases, the DMS encompasses many standalone features such as Customer Relationship Management, Financial Management, Vehicle Sales Management, Fixed Operations Management, Document Management, Network Management, Data Management and Business Intelligence
Digital Marketing Solutions ($370mm or ~20% of revenue)
The Digital Marketing Solutions segment is largely built on top of the acquisition of Cobalt in 2010. The segment provides solutions to drive and manage demand. More specifically, it helps automotive dealers manage their advertising spend and their websites. In this part of the business, endorsements from the OEMs are critical and CDK has endorsements from 10 of the major car brands. While digital marketing solution software is not as mission critical to the operations of a dealership as dealership management software, it does play an important purpose if you understand the modern car buyers’ path to purchase. Approximately 94% (http://business.time.com/2013/10/10/buying-a-car-online-is-about-to-get-way-easier/) of car buyers research online before they buy a car. Thus it is essential that automotive dealerships advertise on online platforms where prospective car buyers are doing their research.
A little more than half of the Digital Marketing revenue is generated helping OEMs and dealerships to manage their digital ad spend with the remaining coming from website management.
Think of the advertising side of the business as a large media buying agency that buys and places digital ads for dealerships. CDK has a proprietary advertising technology platform that dynamically adjusts content and spend across multiple digital marketing channels. CDK is one of the largest purchasers of automotive retail advertising inventory. That scale enables them to leverage strong relationships with the likes of Amazon, Edmunds, Google and Yahoo.
The website business provides solutions and helps manage approximately 7,800 websites for retail dealerships. From an OEM perspective, it is important that all dealerships use similar layout and thematic for their websites because they want to convey brand consistency to the end consumer whether they are browsing at Fremont Chevrolet or Berkeley Chevrolet. Thus, in many cases, the OEM exerts control over or “gently” recommends the software that dealerships need to use to manage their websites.
For an industry that seems to fly under the radar, CDK surprisingly has many competitors. So many competitors that the industry seems fragmented and ripe for consolidation. More on that later. In the meantime, let’s delve a bit deeper into the competitive landscape. There are competitors in every aspect of the retail automotive industry that focus on specific workflows such as inventory management or parts and service but there are also comprehensive solution providers that provide an end to end to Dealer Management System similar to CDK. For our purposes, we’ll focus primarily on the latter. The chart above from the CDK Roadshow Presentation provides the primary competitors in the DMS space.
Reynolds and Reynolds (R&R) is one of the largest competitors to CDK. In fact, R&R and CDK were for the longest time seen as a duopoly in the domestic DMS provider market. While they both have dominant market share in the US, a duopoly isn’t necessarily the case anymore, especially given the growth in International markets. R&R’s primary offering is its Dealer Management System called Retail Management System that, similar to CDK, is a complete dealer management system. R&R also has a secondary forms business that provides printed forms used in every day dealers operations.
R&R is a private company run and primarily owned by Bob Brockman who started Universal Computer Systems and subsequently acquired Reynolds & Reynolds in 2006. Vista Equity Partners and Goldman Sachs Capital Partners were equity investors in this deal but Vista has since exited this investment as of December 2014. R&R is privately owned and public information is difficult to come by but there are some rating agency reports that discuss the business. The information that is available is a good case study as to how much margin improvement can be driven through a DMS business given that PE companies are known to wring out every cost efficiency possible when they acquire a business. In the case of R&R, the business generates operating margins in the low to mid 40% area. This compares to ~28% for CDK’s North America Auto Retail Solutions business, which indicates that CDK may have some room for improvement when it comes to expense management and driving efficiencies in the business. That doesn’t mean that CDK can get to a ~40% operating margin but they may be able to get part of the way there and thereby drive a tremendous amount of incremental value.
DealerTrack Technologies (TRAK) is the only publicly traded competitor. TRAK offers a similar breadth of solutions to CDK including a Dealer Management Software Suite and also a dealer marketing solutions platform through Dealer.com (acquired Dec 2013). The Company has experienced strong revenue growth through both organic growth as well as a multitude of acquisitions. TRAK provides services to ~20,700 dealer customers, but the install base of DMS customers (which is what I primarily care about) is unclear from public filings. From a revenue perspective, TRAK is about a fourth the size of CDK. However, TRAK’s revenue mix differs significantly from CDK in that TRAK generates a large portion of its revenue from its transaction fee based services such as lending and registration/titling. Approximately 58% of TRAK’s revenue is derived from per transaction-based services and only about 38% of revenue is derived from subscription-based software. Thus the business model is not as preferable as CDK, given that transaction based revenue is far more susceptible to cyclical swings in consumer car purchases while subscription revenue is typically contracted over several years. We will discuss the revenue model at more length in the Revenue Model section.
Autosoft is a smaller player (which I believe is family owned) that has gained some share from the bigger players by being very competitive on price. Flex DMS is the company’s DMS offering that is used by approximately 2,000 dealerships in North America. This equates to approximately a 10% share. Similar to R&R, the company also has a forms business that provides printed forms that dealers use everyday.
Auto/Mate is also a smaller player that provides a DMS solution to approximately 1,150 dealerships in the US. Similar to Autosoft, Auto/Mate also competes aggressively on price. More recently the company’s software has been fully configured to communicate seamlessly with Nissan. I wouldn’t be surprised to see other OEMs allowing integration with Auto/Mate’s DMS as it provides a less expensive alternative to their franchised dealers versus CDK or R&R. Auto/Mate is owned by a private investor.
Incadea is a company based in Germany that competes primarily in international markets. The company was once owned by R&R but was spun off into an independent company in 2007. Incadea has experienced strong revenue growth over the last several years driven partly by acquisitions and partly by strong organic revenue growth in emerging markets. Incadea provides software and services to ~2,400 dealerships worldwide, but it does not compete in the US market. In January 2015, DealerTrack acquired Incadea. The news is so recent that I am still discussing Incadea separately here.
Major Industry Trends
Historically, OEMs only allowed communication and integration with a select few larger DMS systems (typically the industry heavyweights- R&R and CDK), but more recently, some large OEMs have broadened the number of DMS solutions that their franchised dealerships can integrate and interoperate with. In other words, the OEMs have allowed the DMS of other smaller players (e.g. Autosoft) to “talk” to their corporate headquarters and factories to ensure that things like parts, sales and accounting data can be exchanged in real time. The additional options are expected to create more competition among DMS providers and drive down pricing for dealers.
In direct contradiction to the trend noted above, some OEMs (e.g. BMW) are doing exactly the opposite of broadening the DMS’ that they integrate with – they are implementing global initiatives to standardize their DMS systems (which are currently fragmented on a worldwide basis– meaning that dealerships around the world are all using different DMS providers). In the case of BMW, they are standardizing to two systems – incadea (now DealerTrack) and CDK.
Both of the aforementioned trends seem contradictory and the reality is that different OEMs are approaching their DMS strategies differently. However, as the undisputed largest player in the industry, both of these trends could actually benefit CDK. In the case of standardization, CDK (as the largest player) will continue to be one of the approved DMS systems and in the case of newer players entering the approved systems lists, CDK will continue to play a major role given the high switching costs inherent with changing to a different DMS provider.
The great recession and resulting OEM bankruptcies have resulted in significant dealership consolidation and closures in the US. US franchised dealerships have declined from about 21,000 in 2008 to about 17,600 in 2013. The industry consolidation has presented both an opportunity and a risk. Dealerships that have a common owner have a desire to run all of their locations on a common platform and have shifted their business to a single DMS solution.
The great recession also had some unintended positive effects on the industry. While it resulted in significant dealership closures, it also helped improve industry pricing and profitability by reducing dealership competition.
Auto dealerships are one of the largest print (e.g. newspapers) advertisers. The continuing shift from print to digital advertising will benefit players that have a digital marketing business like CDK and DealerTrack.
Let’s think about the revenue model as it relates to the different business segments
The ‘Automotive Retail North America’ and ‘Automotive Retail International’ segments comprise approximately 80% of total company revenue and are very similar in terms of revenue composition. Both segments make money by selling (licensing) DMS software to dealerships. CDK receives monthly subscription fees from customers for software licenses, ongoing software support and maintenance of Dealership Management Systems. Revenue generated by the Automotive Retail segments is highly attractive as it demonstrates the characteristics that any prospective investor loves to see in a business: revenue is reliable (contractual), recurring (subscription based) and visible (can be forecasted in the future with predictability). These are revenue characteristics that are present in many software businesses and partly why this investor loves software businesses (I have said that I don’t like businesses that are on the cutting edge of technology here, but CDK is anything but cutting edge. CDK is boring technology and that is a wonderful thing. We will discuss why shortly). Retail dealership clients typically sign long term contracts of several years and pay monthly subscription fees to license the DMS software. Thus, the next month’s revenue or next year’s revenue is not dependent on making the next sale, but rather it is already contracted. This makes it much easier for the CFO to forecast revenue and thus plan for the business. It also provides tremendous comfort as a potential investor. Furthermore, CDK contends that it has a diverse client base with high client retention rates (they don’t however provide the data behind the retention rates).
Despite all the similarities, there are a few key differences in the revenue models of the International and North America segments:
The North America Automotive Retail Segment also generates revenue from transaction fees for processing credit reports, vehicle registrations and internet leads for automotive retailers, primarily in the US. While public filings do not break out the transaction revenue from the subscription revenue, I am led to believe that the transaction-based revenue is much smaller component of the revenue. The International segment does not generate revenue from transaction-based sources. Side note: Transaction revenue is not as desirable as subscription revenue. Transaction revenue lacks all of the characteristics that we love about revenue – it is not recurring, not reliable and not visible. Furthermore, transaction revenue is much more susceptible to cyclicality inherent in the automotive industry. In other words, when consumers don’t buy cars, you don’t make money. This investor prefers to make money whether or not consumers buy cars.
The tenor of DMS contracts in North America is typically longer than in International markets. The average contract length in North America is ~5 years versus shorter contract lengths of ~3 years in International markets. This may just be a result of competitive dynamics in international markets versus domestic markets. While 3 years isn’t as good as 5 years, we will still take a revenue stream contracted for 3 years any day and twice on Sunday. The figure below highlights the typical contract terms for each segment.
Pricing appears to be much lower in International markets. Average Monthly Revenue per Client ranges from ~1,000 to $15,000 in North America versus an average of ~$2,000 in International markets. This may also be a result of competitive dynamics in those markets. Thus, while some of the international markets are growing way faster than the US market, they are likely not as profitable as the US/North American market.
‘Digital Marketing’ segment revenue consists of two component revenue streams.
A little more than half of the revenue is generated by the digital advertising business. This segment has experienced strong revenue growth over the last several years as dealerships shift more of their advertising spend online. CDK receives revenue for placement of automotive retail advertising – likely a small percentage of the advertising dollars that are placed. This revenue stream, however, is not subscription based and is highly susceptible to economic cycles. Typically, during times of economic weakness, companies (especially dealerships) reduce or completely eliminate their advertising spending as consumers begin to curtail purchases (especially large purchases like automobiles).
A little less than half of the Digital Marketing revenue is generated by website management for retail automotive dealerships. Dealerships sign contracts to outsource their website management, maintenance and hosting to CDK. In return, CDK receives monthly recurring fees. These revenues are typically contracted with contract lengths in the range of 1-3 years. While contracted, these revenues are not as highly prized as the DMS revenues, because they don’t benefit from the high switching costs that the DMS revenues do. In other words, a dealership is much more willing to switch its website management provider versus switch the mission critical software that runs its entire operation.
Only 80% of the revenue stream (the DMS portion) is recurring/contracted and visible- not 100% a skeptic might say, but this is a good thing. A revenue stream that is 100% contracted would be able to provide very little upside in a stellar economy. The 20% or so of the revenue stream that is more cyclical can help drive upside during good economic times supported by the stability of the 80% of revenue stream that is contracted and recurring. I couldn’t build a better revenue stream if I tried.
CDK has demonstrated strong revenue growth over the last several years. From 2010 through 2014, revenue grew at a 12.4% CAGR. The composition of this growth is a combination of organic growth and acquisitions.
For the year ending 6/30/14, CDK revenue was $1,974, 7.3% versus the year ending 6/30/13. The overall revenue growth was driven by 5% growth in the Auto Retail Solutions North America segment, 3% growth in the Auto Retail Solutions International segment and 20% growth in the Digital Marketing Solutions segment. Further analysis reveals that the key drivers of revenue growth for all the segments are generally healthy.
Auto Retail Solutions North America segment growth of 5% was driven by both pricing and volume. DMS client sites grew 3% to 13,600 sites and Average Revenue Per DMS client (often referred to as ARPU in other industries) grew 5% driven by a combination of increased sales to existing customers, growth in transaction based revenue and higher pricing.
Auto Retail Solutions International segment growth of 3% was driven primarily by increased sales or expanded solutions to existing customers (not new client sites). We will need to continue to evaluate this segment's ability to increase client sites among what may be a more competitive dynamic in international markets.
Digital Marketing Solutions segment has experienced particularly strong growth over the last few years. The segment has benefitted from the website revenue growth in addition to a secular shift of dealership advertising from offline mediums (i.e. newspapers) to the internet. Segment revenue growth of 20% for the year ending 6/30/14 was driven by an 18% in Average Monthly Revenue per Website and a 26% increase in OEM advertising.
Total company operating margins have increased consistently from 14.4% in 2011 to nearly 18% in 2014 driven primarily by increased operational efficiency and operating scale - a dynamic I expect to continue as CDK grows the topline. It is worth pointing out, however, that segment margins differ dramatically from one another. For the year ending 6/30/14, Auto Retail North America operating margin = ~28%, Auto Retail International operating margin = ~14% and Digital Marketing Solutions operating margin = 7%. I believe that the competitive dynamics and geographic dynamic between the segments drive the variance in the segment margins. North America DMS is likely less price competitive than international markets. And Digital Marketing Solutions is a drastically different business where the value added is clearly not enough to justify the robust margins that the DMS business delivers.
I believe that there is room for improvement in all the segments, especially DMS. CDK's primary competitor, Reynolds & Reynolds (R&R) generates operating margins in the neighborhood of 40%. While it is known that R&R makes trade offs in customer service and technology to generate margins this high, there is a realistic opportunity for CDK to meaningfully increase margins while maintaining high quality service levels, even if they can't get to R&R margin levels. The DCF which is discussed in the valuation section assumes continuous margin improvement driven by both operational efficiency and operating leverage over the long term (10 years).
Investment Merits and Risks
Now that we have covered the business model and the industry landscape, let’s get to my favorite section. What I like about CDK’s business (investment merits) and what concerns me (investment risks). A graphical interpretation will serve us well here.
The small circles in the illustration below are the key facets of the moat that protects CDK’s business and provide it with a competitive advantage. The arrows are the forces, risks and industry trends that are attempting to pierce that moat. We will walk through each circle and arrow in some more depth because they are the crux of the investment thesis in CDK.
Investment Merits (circles in chart above)
Stable and highly recurring subscription revenue
As previously discussed, CDK’s revenue stream is very attractive because the lionshare (80%) of the revenue stream (the Automotive Retail business) demonstrates the characteristics that any prospective investor loves to see in a business: revenue is reliable (contractual), recurring (subscription based) and visible (can be forecasted with predictability). Much of the revenue is governed by long-term contracts with defined pricing per month. The Roadshow presentation excerpt in the Revenue Model section above shows how contract terms differ between the three segments.
Mission critical software
DMS software is mission critical software that runs all the operations of an entire dealership so it can’t just be turned off to save money during lean economic times when consumers are buying less cars. During the global economic downturn, CDK’s North America revenue (excluding acquisitions) declined 4% between 2009 and 2010 even while US car sales volume declined 21% during the same period. DMS software is a cost of entry and a must have for every franchised automotive dealership. Thus, despite historical auto industry cyclicality, CDK has been very resilient.
High switching costs
There are many DMS providers out there. What’s to stop a CDK customer from switching to a competitor provider like R&R and DealerTrack? High switching costs provide natural barrier to switching. New DMS systems require upfront installation costs and implementation costs in addition to a time investment required to train all the departments within the organization to use new DMS software. Thus, once a dealership implements DMS software and trains their staff, they are not likely to change the software. The time and financial costs of switching are too high. Now this doesn’t mean that dealerships never switch. They do sometimes switch, but it is less likely. CDK’s high renewal rates (as cited by the company) over the last several years support this point.
Scalable software business
Similar to other software businesses, CDK’s DMS business (which comprises the lion share of revenue) is highly scalable and benefits from high operating leverage. In other words, the expense and capital investment in creating and updating the software is largely fixed and can be leveraged across nearly all of CDK’s customers. As CDK’s customer base and in turn revenue grows, the expenses and capital expenditures do not grow at the same pace given that they are largely fixed. This leads to higher margins, as the company grows bigger. CDK has demonstrated this operating leverage and scalability as operating margins have increased from 15.5% in 2010 to nearly 18% in 2014.
Major OEM integration
While the role that this phenomenon plays in the DMS software space seems to be diminishing, OEM approval still provides a competitive advantage to CDK worth discussing. As discussed previously, some OEM software systems only allow communication and integration with a select few larger DMS systems (typically the industry heavyweights- R&R and CDK). This dynamic may be evolving a bit but the entire industry has not changed over. Furthermore, some OEMs (e.g. BMW) are implementing global initiatives to standardize their DMS systems (which are currently fragmented on a worldwide basis– meaning that dealerships around the world are all using different DMS providers). In the case of a limited set of choices, CDK will almost always be one of the limited options that OEMs approve given that it is the leading DMS provider and an understandable reluctance to test, certify and frequently update several different DMS systems. From a dealership perspective, it benefits the dealership to choose a DMS provider that is certified or approved by the OEM so that the respective systems can communicate important information like vehicle inventory and sales data to each other.
Slow pace of enterprise technology
If you know me and have read some of my other articles, you know that I don’t like technology companies. But let’s get a bit more specific, because it’s not technology stocks that I don’t like. It is companies that are on the cutting edge of technology or making a bold bet on the adoption of new technology. It is too unpredictable. This is often the case with new consumer technologies. Let’s leave that to the venture capitalists. Enterprise technology, however, is very different from consumer technology. At Scuttlebutt we love enterprise technology because it is boring and boy oh boy do we love boring. Enterprise technology companies make for good investments for a few key reasons:
Unlike consumer technology, enterprise technology life cycles are long and slow. Enterprises aren’t looking for the disruption caused by frequently changing out their software system. They also aren’t necessarily looking for the hottest new technology. They just need good, reliable software to manage their businesses with the occasional updates. They want to keep their software for at least 5-7 years as long as the price is fair and reasonable. If you want evidence of this, take a look at Microsoft Windows XP. You still have businesses that use Windows XP more than 12 years after it launched in 2002 because it is simple, cheap and it works for their needs. It is only within the last year that Microsoft dropped support for Windows XP.
Switching costs are often even more pronounced when it comes to enterprise technology. The costs to the enterprise of switching technology include upfront installation costs, costs to train employees on new software, losses in productivity as a result of unfamiliarity with the new software, potential loss of historical data or customer records and general friction inherent with any change.
Dominant market share of existing players in a niche industry
While the addressable market across the entire value chain is about $40bn, the retail auto dealership software space is relatively niche versus other larger verticals in the software industry like databases (think Oracle, Salesforce.com) for example. Niche implies that the software is highly customized for a specific use case. And the existing players like CDK and R&R have already established strong market positions. While there are also several other smaller competitors in the space, we’re not likely to see many new competitors enter the space. The niche nature of the retail auto dealership space combined with the dominant market positions of the existing players keeps potential competitors at bay and provides an enduring barrier to entry.
Lower oil prices (not a circle in the graphic above)
Lower oil prices are typically a boon for car sales and I don’t see the recent decline in oil prices as any different. I do, however, view the decline as a more temporary performance driver and thus do not include it as one of the components of the moat above. Lower oil prices drive more car sales and thus benefit the dealerships. This is a positive for transaction-based revenue, but also DMS revenue. When dealers are doing well and flush with cash, they are more willing to pay for their DMS systems.
Investment Risks and Considerations (arrows in chart)
Auto industry sales have historically experienced tremendous cyclicality.
The economy in general is very cyclical. Every 5 to 10 years or so there is a boom that is then followed by a bust. This is the nature of the beast and nothing can be done about it. During busts or downturns in the economy, larger discretionary purchases (think that new television or new computer) are the first to be curtailed as consumers skimp and save. Cars are a very big purchase and are no exception to this general reduction in spending. In fact, during the most recent economic recession, new vehicle sales declined by 18% in 2008 and a further 21% in 2009 and still have not recovered to pre-recession levels. See chart below. Remember “Cash for Clunkers”? That was a program devised partly to help struggling auto companies and auto dealerships sell more cars in 2009.
It’s apparent that car purchases are very cyclical and thus would naturally follow that this would have a big impact on companies that provide software to automotive companies or dealerships. While you wouldn’t be crazy to follow that logical path, you would be wrong. DMS software is mission critical software that the dealership needs to run whether they sell a thousand cars or just a few cars. It’s not software that can just be turned off to save money when times are tough. The numbers prove this out (I'll rehash a point I made earlier in the 'Investment Merits' section). During the economic downturn CDK’s North American revenue declined by only 4% (2009 to 2010 excluding acquisitions) while US car sales volume declined by 21% during the same period. The modest decline was likely driven by a decline in some of the transaction based revenue (transaction revenue comprises a small portion of overall CDK revenue) as well as some DMS contracts that renewed around that time with some likely price concessions granted by CDK to their struggling dealership customers.
Low Margins of Digital Marketing Solutions: The Digital Marketing Solutions division is high growth but low margin.
The Digital Marketing Solutions division grew at a 19% CAGR from 2012 to 2014. The division is experiencing tremendous growth driven by two key phenomenon: 1) Dealership advertising levels have recovered tremendously since the depths of the recession and continue to grow in tandem with new car sales. Franchised car dealerships spent a total of $7.6bn in advertising in 2013, up 6.1% from $7.2bn in 2012. This compares to approximately $5.4bn in 2009 during the depth of the recession. Dealer ad dollars are expected to continue to grow for the next few years in lockstep with new car sales (although the ad spend per new unit sold may stay relatively flat). 2) Dealership customers are shifting a greater proportion of their advertising spend to online channels, which is where CDK operates. Internet advertising is taking share from other mediums like newspapers and radio. In 2006, Internet advertising represented 12% of total ad spend and newspapers represented 27% of total ad spend. Since then, the tables have completely turned. Internet is now the number one medium used by dealerships, comprising ~33% of total ad spend (even bigger than TV which is ~21% of total ad spend), while newspapers now represent 16%. See graphs below. In an economic downturn, it is very likely that levels of internet advertising could stay flat or even continue to increase as dealerships continue to shift spend from old media channels like newspapers and radio to the internet despite an overall decline in ad spend. Thus, I believe that growth won’t be an issue for this segment.
The lower margins of the Digital Marketing Solutions segment concern me somewhat. Digital Marketing Solutions EBIT margins were 7% in the latest period (FY 2014), which is down from 9% in 2013. Low margins frequently suggest that a segment is either highly competitive or low value add. Low margins also provide less cushion during economic downturns. The Digital Marketing Solutions segment is by its very nature, the segment that is most susceptible to industry and general economic cycles. According to the Form 10, the margin decline in 2014 was driven by both expenditures to support new contract rollout (admittedly I don’t know what this means but I think it is related to website management) and higher advertising costs. High advertising costs is a bit surprising because I would have thought that there would be more scale in this business as revenue grew $63mm in the same period. The majority of the revenue increase was driven by an increase in website management revenue. In summary, I think there is more to learn about the economics of this segment. I know that it is cyclical but I do not quite understand what is driving margins and how scalable this portion of the business is. I am comforted somewhat by the fact this is a smaller part of the business (<20% of revenue) and that it accelerates financial performance during times of economic boom.
The Auto dealership software industry is much more fragmented than one would think.
As previously discussed, there are a few large players: CDK Global, Dealertrack and Reynolds & Reynolds are the 800-pound gorillas. But there are also a bunch of smaller players that include Auto/Soft, AutoMate, Incadea, Dominion Dealer Solutions and a few others. Incadea operates primarily in international markets and Dealertrack recently acquired Incadea, making DealerTrack an even larger competitor. Fragmentation represents both a risk and an opportunity.
A fragmented competitive landscape is a risk because the persistence and entrance of smaller players drives down industry pricing. These smaller players usually gain market share by offering lower pricing than their larger rivals. For example, a DMS system from a provider like Auto/Mate can cost several thousand dollars less per month than a comparable system from CDK or R&R. While CDK and R&R still dominate, money talks and lower pricing from these rivals can impact their pricing and ultimately operating margins. However, fragmentation is an opportunity here as well. Fragmented industries are often ripe for consolidation. Increasing consolidation can help sustain or drive pricing as smaller competitors that hurt overall pricing are acquired by larger players. We’re starting to see some evidence of this with the recent news regarding DealerTrack and Incadea. Additional consolidation is good for CDK, especially if they are one of the acquirers. Note however, that I don’t expect CDK to be acquired or make a transformative acquisition in the next 2 years because this would likely trigger tax consequences that alter the tax- free spin from ADP.
Increasing consolidation of franchised dealership groups drives DMS pricing down
As of early 2013, there were 17,600 franchised dealership groups and 37,000 used vehicle dealerships in the US. The downturn has driven consolidation of dealerships but there is still a tremendous amount of fragmentation. Over the last several years, we have begun to see ever larger dealership groups form and buy out smaller dealership groups. Some of these are Lithia Motors (LAD), AutoNation (AN) and Penske (PAG)). Even Warren Buffett recently got into the game by buying the fifth largest dealership group in the country called Van Tuyl Group (privately owned). This increasing consolidation among dealership groups is likely to be a negative for CDK and other DMS providers. As dealership groups merge and grow larger, the pricing power shifts in their favor and they are likely to try to extract price concessions from their DMS providers. I haven’t necessarily seen any evidence of price degradation but put it on the radar as a long-term risk if dealership consolidation continues.
New model of direct to consumer sales may limit usage of DMS software
I only need to mention one company here and that is Tesla. Tesla is taking a very different approach to auto sales than any of its automotive competitors. Tesla is eschewing the franchised dealership model in favor of direct to consumer sales. In this new model, Tesla likely does not need to license DMS software. They likely have proprietary software to manage their sales, financing, etc. While I believe that selling cars in this way is likely much more efficient and cost effective, I see it as highly unlikely that larger players like Toyota or GM decide to change their current dealership model in a meaningful way. And even if they do, they cannot eliminate any of their existing dealerships. State franchise laws are very strict and would prevent them from doing so. In summary, I see the risk from the transition to a new sales model in the near to intermediate term as very low, but I did think it was worth highlighting to acknowledge the breadth of risks that are out there.
At a high level, the management team (including the CEO) is a group of ADP veterans that has a wealth of industry experience and has made good capital allocation decisions.
CEO Steve Anenen has been with ADP and CDK for 39 years. He has been at the helm of CDK since 2004 (10 years) and has managed exceptional growth of the base business while making sound acquisitions such as Cobalt, which is now CDK’s Digital Marketing Solutions Business. Anenen has seen many parts of the business and I generally believe that this experience will be valuable to the newly public company.
Many of the other management team members including the CFO and the division heads also have a wealth of experience with CDK and ADP. Chairman Leslie Brun has extensive financial, management and advisory experience that makes him particularly well suited to the role. From corporate governance perspective, CDK gets a gold star for keeping the CEO and Chairman roles separated.
The important thing I like to look at with spinoffs is how much skin management has in the game (in other words, how much of CDK Global’s equity will owned by management). A management team that is overpaid with equity that significantly dilutes the common shareholder’s ownership is not a good thing, but a management that owns a substantial equity stake, has interests that are better aligned to shareholder interests. There is a fine balance between the two, but in summary, we do want management to be owners of the company in a fairly meaningful way.
In the case of CDK, management owns very little of the new company on a pro forma basis. Even if we include additional shares that would be acquired upon the exercise of additional outstanding stock options, the management team would own close to a half million shares in aggregate. While this may sound like a lot, it is well less than 1% of the total outstanding shares. I believe that management’s ownership of the company will grow over time but I wish it were somewhat higher at the get go to ensure that management interests and shareholder interests are well aligned.
Having said that, there are a few things that I like about management compensation at first blush:
Minimum stock ownership guidelines are a key aspect of CDK’s compensation policy. The policy requires that the CEO hold minimum equity ownership in CDK of three times base salary (= ~$1.5mm for 2014) and executive officers hold at least one times base salary. This ensures that management has at least some skin in the game, but this amount of ownership is still relatively low. As of this writing, management satisfied these requirements.
~50% of CEO compensation and ~37% of other executive compensation is comprised of long-term incentive comp like stock options and Stock Awards. I like compensation that is based on long-term objectives. It motivates management to make decisions that are good for the long term.
In mid –November, management declared a dividend of 12 cents per share, which results in a dividend yield of ~1.0% (at current stock price = ~$46). I like the move by management to initiate a dividend. The declaration of a dividend represents a fixed quarterly expenditure ($77mm per year in aggregate in FY 2015) and demonstrates confidence in the business by the management team and a capital allocation policy that is shareholder friendly. I expect continued increases in the dividend as former parent ADP was well known for. Additionally, the Free Cash Flow payout ratio of ~33% represents an amount that is appropriate and provides sufficient cushion relative to annual FCF of approximately ~$230mm (excluding one time separation costs).
Prudent acquisitions that add value are also a way that I qualitatively evaluate management’s strategy. CDK has made 30 acquisitions since 2000 that have ranged from smaller tuck in to more transformative. Two key acquisitions have included: the acquisition of Kerridge Computer in 2005 which formed the foundation for CDK’s International DMS platform and the acquisition of Cobalt which is the basis for CDK’s digital marketing solutions business. Both of these acquisitions have proved to be integral to the current business and generally make logical sense for the company. I don’t have specific information on purchase prices and other tuck in acquisitions, so it is difficult to say if management has been prudent with all acquisitions, but at a high level, I can say that acquisitions have been additive and logical.
A key component of CDK’s business strategy going forward is completing acquisitions that support or complement its existing technology. We will need to continue to evaluate the quality and prudence of acquisitions going forward.
Now we come to the million-dollar question – what’s it worth? Before we go there, let me recap. There are a lot of things that I like about CDK: an experienced management team that is working in the shareholder’s interest, the recurring revenue stream of DMS software, the relative resistance to recession due to mission critical nature of the software, high switching costs, slow pace of change in enterprise technology and high operating leverage. There are some risks/watch outs to be aware of: industry cyclicality could have some impact to the revenue stream, low margins and susceptibility to cyclicality of Digital Marketing Solutions, fragmentation of DMS providers and increasing consolidation of franchised dealership groups. In totality, I would say that the positives far outweigh the risks. And with the risks, there are fairly strong mitigants. So what’s this all worth to us at Scuttlebutt?
Based on a Discounted Cash Flow Analysis, we get to an intrinsic value of ~$42 per share. When the stock was trading around $25 to $27 per share post debut, it was a healthy discount to fair value. However, at a current share price of ~$46, the value proposition is difficult to justify.
There is an important point that I need to make on the origin of the intrinsic value. The intrinsic value of ~$42 is predicated on a reasonable level of margin improvement driven by two key factors. First, I believe that the business is not productive as it can be today when you compare it to peers like R&R that generate operating margins of 42%-44%. However, I believe that R&R may generate those excess margins at the expense of its product and service levels. Thus, my model assumes that CDK can improve margins in its three key segments to competitive levels, but still well below R&R. Second, CDK should be benefit from operating leverage as it grows the top line, thus improving margins naturally.
Sachem Head Capital Management and Fir Tree Partners, which now own ~18.5% of CDK between the two funds have driven the price of the stock up significantly since the debut. I should note that Vanguard and Blackrock, the mutual fund companies, also own large positions of ~6.1% and 6.9% respectively, but neither is known to be activist in any way.
The question is whether or not these hedge funds will continue to buy and if this will continue to drive the price of the stock up. I would call these speculative drivers of price (not value) and could not recommend an investment on the basis of an increase in price due to these drivers. However, these funds may try to drive value– by recommending strategies and acquisitions to management that can drive margins, Free Cash Flow or strengthen competitive advantages. These are the drivers of value that we at Scuttlebutt can get behind. There is another school of thought that there may be a push for consolidation in the industry. I view large scale or transformative consolidation for CDK as unlikely in the near term given that there would be tax implications with doing so. I think that they will be more focused on tuck-in acquisitions in the near term. However, based on the universe of information out there, it is difficult to ascertain precisely what these funds are proposing to CDK management.
My recommendation is to wait out until the stock trades at fair value or a discount to it. I believe that market volatility should provide more reasonable purchase price opportunities. I would be willing to pick up more shares at what I deem to be fair value because the business is a gem and can deliver decent returns (for what I view as relatively low risk) even at fair value, but I will wait until the shares trade at more reasonable levels. Patience is the name of the game. I think Charlie Munger, Buffett’s right hand man at Berkshire Hathaway, called it “sitting on your ass” investing. In other words, sitting on your ass until the right investment or right price materializes.
Disclosure: I am currently long CDK. I am not an investment advisor and this article presents my personal views. While I have conducted a fair bit of research to write this, you should also do your own research and come to your own conclusions.