For reasons I will explain below, I think there is a good short-term long opportunity in this stock as its very high short interest is likely to come down meaningfully and relatively quickly. Even if I put in very conservative assumptions into a DCF, I cannot generate a valuation that is not AT LEAST 100% upside from here (assumptions for DCF highlighted in the last paragraph below).
AVG Technologies (AVG) is a consumer and enterprise software company that operates using a "freemium" model. In 2012, the company grew Adjusted EPS 37% on 18% sales growth. Despite this growth and consistently beating expectations (the average EPS beat vs. Street over the last 5 quarters has been close to 50%), the stock currently trades at less than 10x next 12 months Street Adjusted EPS. Free cash flow per share has historically been higher than EPS primarily because of the company's growing deferred revenue balance (subscribers generally pay ahead of revenue recognition). As of the middle of June, the short interest as a % of the float stood at 33%.
The company's main product offering is free antivirus for desktop computing. AVG primarily monetizes its user base by upselling subscriptions to a portion of its users and through a product called "secure search." Users of the antivirus software can download a toolbar so that AVG becomes the default search provider. As users click on advertisements, AVG gets a revenue share with the search engine company that is providing the search results. The bear thesis largely revolves around this search business. A good write-up that summarizes the bear thesis can be found here.
Immediately after this article came out, AVG's short interest jumped from just 751k shares at the end of January to over 8.5m shares in mid-April. The Company significantly beat Q1 expectations when it reported results on April 24th, and as I will argue below, the short thesis seemed to have largely played out but the short interest remains very high at 7.9m shares.
In Q4 2012, almost all of AVG's search revenue came from its Google partnership. In February, Google (GOOG) made a policy change so that all new downloaded toolbars needed to be "opted in." Bears were correct in concluding that this change would force AVG to move all new users to Yahoo (YHOO), which still has an "opt-out" policy. However, they were incorrect in the financial impact this would have. Bears argued that since Yahoo monetizes search significantly (up to 50%) lower than Google, AVG's search related revenue would have a sharp fall. This argument made sense at the time, but the evidence does not support the bear case. In Q1, the company reported that Yahoo represented 9% of search related revenue. Moreover, AVG said that Revenue per Thousand Searches (RPM) stayed flattish sequentially and guided for this to be the case again for Q2. The bears are failing to recognize the fact that while Yahoo likely monetizes less, Yahoo's new management views getting increased scale in search as strategic and is very likely giving up more in revenue share than Google was. I estimate that AVG's revenue per search is only about 15% less than what it was getting with Google. If it is much lower than that, then RPM would not have stayed relatively flattish sequentially last quarter. This estimate has also been confirmed by talking to other toolbar companies. Moreover, if one looks back to the 1st half of 2010 when almost all of AVG's search business was with Yahoo, the RPM was in the $16-17 range vs. the $21 RPM it had with Google in Q4 2012. Based on conversations with several people in the industry, I think it is highly likely that AVG's current RPM with Yahoo is much better than it was in the first half of 2010. First, Yahoo has improved monetization since then. Second, AVG has much more leverage since it has many more users. Third, Yahoo has new management who wants to take share from Google and who views increased scale in search as strategic. While it is true that AVG would have been economically better off if Google did not do the policy change, it does not appear that the lower monetization with Yahoo is enough for the company to miss estimates, and at sub 10x FCF, the company needs to miss estimates significantly for the shorts to make money. When the company beat EPS expectations in Q1 by over 50%, AVG left full year guidance largely unchanged. This effectively brought down search revenue expectations for the rest of the year and was the impact of the Google policy change. This should become even clearer when the company reports results for Q2 which will be the first full quarter following the Google policy change. Perion Network (PERI) and IAC/InterActiveCorp (IACI) are impacted by the same Google policy change and both also still guided for growth.
While the new Google policy has already impacted Street numbers, there are several sources of large upside optionality that will also become clearer to bears as we progress through the year.
1) AVG has 36m mobile users as of Q1. These mobile users grew 100% y/y (83% organically if you adjust for an acquisition). The company is currently not monetizing these users and the Street does not seem to be incorporating any revenue from mobile even for next year. AVG has said that it will start to monetize these users in Q4. AVG's mobile products have gotten great reviews for products such as AVG Security and AVG Battery Saver & TuneUp, both of which received 4.5 out of 5 stars on Google's app store. In addition, given the privacy climate created by the government's PRISM project, AVG's Do Not Track (DNT) functionality has attracted a lot of users to their mobile offerings. I estimate that the company could have close to 95m mobile users by the end of the 2014. In Q1, 2013, the company generated $56.6m of subscription revenue from an average of 117m desktop users or close to $2 in annual subscription revenue per desktop user. Even if the company can monetize mobile users at just 10% of the rate it does on desktop, this could imply at least 3% upside to Street revenue in 2014, and because the incremental margin on this revenue would be extremely high, it could imply well over 10% upside to Street 2014 EPS.
2) AVG continues to do well with improving its monetization of desktop users. The company's desktop subscription revenue growth has accelerated over the last 5 quarters and was 54% of total revenue in Q1 2013. This happened despite the fact that desktop users declined sequentially over the last 2 quarters as Google's toolbar policy change also impacted customer acquisition. I expect that desktop user growth will go back to growing sequentially. After all, AVG is offering a good free alternative to existing Antivirus products on the market and still has a relatively low share of the total PC installed base. PC sales have been weak which is great for AVG as new PCs are often loaded with trial antivirus from Symantec or McAfee. An aging PC installed base is also good for AVG's newer Tune-Up products, which appear to be doing well. The company plans to break out Tune-Up users on its next conference call, which implies that it has become material. In Q1 2013, the average paying desktop subscriber paid the company at an annualized rate of $15.09, which was up from $12.43 in Q1 2012. Given that the company seems to be pushing its security suite, which is priced at $40, there seems to be a fair amount of room with this variable to continue to rise.
3) The company recently made an accretive acquisition of LPI which will add $5m of revenue in the 2nd half of 2013. This deal further diversifies the company away from search related revenue and also does not seem to be factored into Street estimates yet.
4) Finally, since the company is domiciled in the Netherlands and has such a large and growing user base that can be cross-sold other products, it is possible that it could be an acquisition target for one of the many Technology and Internet companies that have a lot of overseas cash.
I am not arguing that many of AVG's users likely unknowingly opt-in to AVG's secure search products. However, while the company may do some unpopular things with its search related business, the short thesis appears stale and the short interest has surprisingly stayed high despite obvious evidence that point to the fact that the overall revenues should continue to grow quite healthily. This should become clearer after the Q2 results and if shorts have not covered by Q4, they will likely cover as we get closer to the big upside optionality of monetization of mobile users. The CEO's resignation is concerning but the company is likely past its hyper growth phase and it seems like he just wants to do something more entrepreneurial. He has not sold any shares and has not even initiated a 10b5-1 plan. Moreover, he will stay on at the company as an advisor.
In my DCF, I assume that 2014 search related revenue declines slightly as we cycle through the lower monetization from Yahoo vs. Google (I model a 25% per quarter churn rate for Google toolbar users vs. the 15% churn rate used by the bearish analyst in the article mentioned above). However, search revenue should then go back to growing at least in the low single digits in 2015 and beyond. The company has said that it plans to generate platform revenue in other ways than just search such as ecommerce and advertising. I have not factored any of this. I model desktop subscription revenue growth to decelerate from 21.5% y/y growth in Q1 2013 to the low teens in 2015 and single digits thereafter. If you play around with the key variables (% of active users that pay and revenue per paying subscriber), it is very difficult to even envision this level of deceleration given historical trends. I expect that by 2018, they get to the same number of mobile users as desktop users today - again a very conservative assumption given the current growth rates of mobile users. I also assume that by 2018, the company generates about 40c in subscription revenue per active mobile user (vs. about $2 and rising in subscription revenue per active desktop user today). I use a 30% operating profit margin in my terminal year vs. the 37% operating margin that was reported in Q1 2013 and the 33% average historical operating margin. Finally, I use a 10% discount rate and a 10x terminal multiple. I think a 10x terminal multiple is very conservative especially since I expect in just a few years, the vast majority of revenue will be coming from subscription revenue. The company's auto renewal rate on subscriptions was at 77% last quarter and has been consistently rising (it was just 44% in 2011). With all these conservative assumptions, I still get AT LEAST 100% upside to the stock.
I welcome any comments from people who are short to understand what I may be missing.
Additional disclosure: The author may make trades in securities mentioned without notification. The information contained in this article is impersonal and not tailored to the investment needs of any specific person. You should consult with a professional where appropriate. The author shall not be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages. The opinions expressed in this article are for informational purposes only and should not be construed as investment advice. The article is not a recommendation of, or an offer to sell or solicitation of an offer to buy, any particular security, strategy or investment product. The research for this article is based on public information that the author considers reliable, but the author does not represent that the research or the article is accurate or complete, and it should not be relied on as such. The views and opinions expressed herein are current as of the date of this article and are subject to change. Any projections, forecasts and estimates contained in this article are necessarily speculative in nature and are based upon certain assumptions. In addition, matters they describe are subject to known (and unknown) risks, uncertainties and other unpredictable factors, many of which are beyond the author’s control. No representations or warranties are made as to the accuracy of such forward-looking assumptions. It can be expected that some or all of such forward-looking assumptions will not materialize or will vary significantly from actual results.