We'd like to address what we perceive to be the bear case.
First, we do think there are varying opinions on how much growth Ancestry.com can achieve during the next several years.
However, let's look at Ancestry.com this way. What's the company worth if what the bears say is true (no growth after year 2, flawed business model, etc). This is where things get really interesting -- there's significant upside with little downside risk at these price levels. In order to achieve little to no growth, the firm would likely turn off marketing and ad spend, resulting in earnings explosion.
Specifically, in a no-growth scenario, we're looking at over $3.50 in earnings power NEXT YEAR (2012), ex marketing and advertising. Please see the P&L below. The takeaway here is not that the company is going to reduce marketing and advertising expense to 0 in 2012, but instead the tremendous opportunity it presents in terms of earnings and cash flow generation when the firm turns it down after the high-growth period -- whether it be 5, 10, 15 years from now (or after year 2 if you're a bear).
Here's an example P&L with no marketing and advertising spend - the bottom line is very impressive. Please do not assume that this is my base case. I'm simply illustrating a point revealing the tremendous earnings potential in this cash-rich business model. Please forgive me for the presentation of the data, as my model would not paste easily into the entry box.
Ending Subs 1,813,383 2,176,060
YoY % Growth 30.0% 20.0%
Sub Rev 375,325 459,398
YoY % Growth 33.3% 22.4%
Prd Rev 24,021 27,794
YoY % Growth 24.7% 15.7%
Tot Rev 399,346 487,192
Tot Rev YoY% 32.7% 22.0%
Sub COGS 56,711 66,379
SubCOGS%SubRev 15.1% 14.4%
Prd COGS 7,470 8,726
PrdCOGS%PrdRev 31.1% 31.4%
Total COGS 64,181 75,106
TotCOGS%TotRev 16.1% 15.4%
Gross Prof 335,165 412,086
Gross Margin 83.9% 84.6%
Tech and Dev 56,707 68,207
Tech/Dev%TotRev 14.2% 14.0%
GrossSub Adds 1,494,546 1,450,706
Net Sub Adds 418,473 362,677
Sub Acq Costs 84.00 85.00
Total Cancel 1,076,073 1,088,030
Mark Ad Spend 140,029 0
Gen & Admin 42,291 49,903
YoY % Growth 19.5% 18.0%
Tot SG&A 182,320 49,903
Tot SG&A%ofrev 45.7% 10.2%
Amort of intang 17,080 17,080
Tot op expense 256,108 135,190
Income from op 79,058 276,896
Op Margin 19.8% 56.8%
Int expense -428 -428
Other income 0 0
Income bef tax 78,630 276,468
Tax expense -27,717 -96,764
Eff tax rate 35.3% 35.0%
Net Income 50,913 179,704
Dil Shares Out 50250 50250
Diluted EPS $1.01 $3.58
Again, this P&L is for 2011 and 2012; 2011 includes marketing and advertising expense and 2012 does not. The firm's current $42 price per share divided by ~$3.58 earnings per share in 2012 (ex mark & ad) = under 11 times earnings. Translation: the company is fairly valued right now only if it fails to attract any new subscribers (assuming a market multiple meaningfuly below 15 is justifiable for no/low growth company). That said, an argument can be made that given its cash-rich biz model, it could still warrant a 15 times multiple even with no/little growth. 15 x 3.58 = ~$55. We've run the valuation across a number of different scenarios now -- PEG below 1; bear-case P/E below 15 times next year; DCF basis over $55 per share, etc.
The shorts are betting on a thesis that has already been proven wrong via valuation. Having a conversation about growth is just icing-on-the-cake for the longs. When we're talking about growth, we're talking about whether this is a double, triple, or more from these levels.
Fair disclosure: Including marketing and advertising expense of $138 million in 2012, we expect the firm to earn over $1.75 per share.
Ancestry.com's has minimal downside risk with significant upside potential.
Thank you for the clarification. We've been modeling such a roll off and do not expect to make any changes to our adjusted EBITDA or EPS forecasts at this time.
Ancestry.com presents investors with substantial upside opportunity with little downside risk. We'd still be buyers at these levels in anticipation of a break out to new highs within the next few weeks (we hope).
So with EPS for 2012 of $1.75 what earnings multiple would you use? At 20 times EPS the share price is $35. Higher than 20 multiple implies the EPS growth rate to be magnificient on an ongoing basis.
Not easily achieved especially in todays market environmewnt.
How many times do I have to tell you and others that institutional investors look at EBITDA in preference to EPS. In this case, specifically it makes sense, because, as I said before, a big chunk of goodwill amortization ends after 2012. The adjusted EBITDA multiple is around 13x and EBITDA is growing at about 40% per year. I asked before if anyone could find a stock growing EBITDA at that rate and yet trading at such a low multiple. No one has responded. I ask it again...
I don't wish to rehash old arguments - we'd just waste time talking past each other.
I will only say this: the bears aren't forecasting stable subs, they're forecasting declines (eventually). And no marketing expense at all is a fool's assumption. Even if they do shift from a growth to an earnings maximization strategy, they'll have to spend some money on marketing just to tread water (again, even their >2 year users are churning off at a >20% per annum rate). I will agree with you though that the company could massively improve its margins quite easily - in fact, I noted as much in my Bear Case response to your original post. But if you assume near-term subscriber growth followed by longer-term decline coupled with massive margin expansion you still only get $15 to $20 in a DCF.
We do understand your position, and while we disagree on the firm's business model and outstanding growth potential, we thought revealing the numbers would present a common language for some debate.
Interestingly enough, it takes quite a while to find an ancestor that served in the Civil War. Did you know that you have 4 grandparents, 8 great-grandparents, 16 great-great grandparents, 32 great-great-great grandparents, and 64 great-great-great-great grandparents...and so on (indeed, the more success one has in searching his/her family history, the more work one has to do to go back further--the longer the subscription). It's probably in the set of 32 that there'd be the possibility of an ancestor serving in the Civil War -- and getting to that point is not an easy task for the first-time user (it takes time). And newbies are just getting exposed to this. Further, many subscribers want to learn about the when, where, and how of their family's past -- 128 great-great-great-great-great-grand parents takes years to discover.
Further, the network effect of having more subs on the site is the most powerful business model out there. It gets stronger as more sign up and share their information/family trees. There may be picture of your great-great-great-great grandfather out there on their site, for example. Now that's interesting!
How many people have visited their site? You mentioned 4-8 million (6 million). Ancestry.com was founded in 1983 – they’ve had about 2.5 million gross sub adds within the past 4 years. 2010: 1.02m; 2009: 674k; 2008: 556k; 2007: 480k. Roughly 2.5m/6m = 42%. That means 42% of gross sub adds from the firm’s 28-year history have occurred during the past 4 years: we have undoubtedly entered into a structural shift in how individuals perform genealogical research based on the trends in this data. As Ancestry continues to upgrade its content, users return – perhaps only the last few years of subscriber adds can be viewed as relevant visitors (2.5 million seems like a better number than 6 million).
A better way of thinking about this: “talking about >500k lost subs a year that have to be replaced by ‘fresh meat’”. The firm’s lost subs can be viewed as largely a percentage of gross subscriber adds for each period. For example, total cancellations divided by gross sub adds were 77% in 2009 and 68% in 2010. Therefore, conversation rates have improved from 24% to 32% during the past few years on new adds to the site. This may be a better way of looking at their business model: Ancestry.com is converting roughly 30% of new sub adds. If we don’t view it this way in a bear case, we may be wrongly assuming that churn will continue to increase (or maintain existing levels) as sub adds decline -- when we know that most cancellations are occurring within the ‘newbie’ group in any given quarter (users less than 3 months).
Unfortunately, I wasn’t able to arrive at a DCF fair value at $20/ per share. Even including subscriber declines of 5% after 2012 over the subsequent 3 years, we cannot arrive at a fair value less than ~$35. What is the P/E on your estimate of earnings in 2015 based on estimates on your DCF? One quick way of checking to see if your DCF makes sense is to triangulate it with earnings, EV multiples, etc. It’s easy to get lost in a DCF model.
What’s probably most intriguing is that Ancestry.com curiously has a revenue line item that it has yet to tap – advertising. With 1.4 million subscribers in a relatively affluent age bracket, this is an opportunity that will be tremendous and recurring (yet cyclical). Plus, subscribers tend to visit its site for hours at a time – this is an advertiser's dream. I don’t think the shorts have given much thought to this opportunity. Even subscription models with 140k paying subscribers (7.5 million registered subscribers) reap tens of millions in yearly advertising sales.
Hope you find this helpful.