Notes on Demand Media and Rightside

These companies do not seem compelling to me as investments (long or short).  However, I feel like these companies are somewhat misunderstood because some analysts don’t understand the industry.

Demand Media is slightly parasitic

Demand Media provides content on the Internet and makes money from advertising.  However, it has engaged in practices where it has tried to game search engines into driving more traffic to their websites.  The only search engine that matters is Google so I will start talking about Google.  In general, Google spends a lot of resources fighting people who try to game its search engine.  The general pattern is this:

  1. Somebody figures out a way to game Google’s search engine (e.g. link spam, reciprocal links, paid links, link farms, content farms, etc.).
  2. More and more people join the bandwagon.  Suddenly, the tactic begins to dominate search results.
  3. A mini arms race erupts as Google implements new algorithms to identify “web spam”.  The people trying to game Google’s search try to figure out ways to get around these new algorithms.
  4. Because the algorithms aren’t perfect, there will be false positives and negatives.  Over time, Google tweaks its algorithms and improves them.
  5. Ultimately, Google wins and the particular technique used to game search results no longer works.
  6. The era for a particular “search engine optimization” technique ends.  (Mostly.)

Demand Media’s ‘content farm’ strategy is not something that will continue to work given that Google is trying to kill that strategy.  Its quality of earnings isn’t that great whenever it is pursuing “fast money” strategies that will not lead to sustainable cash flows.

Why isn’t Demand making money?

Demand’s origins are in venture capital.  The problem with venture capitalists is that many of them are trying to make money by selling a high-growth business to stock market investors. Because of this, VC-backed companies often focus on revenue growth over activities that will create long-term profitability.  I don’t think that the content business is one where a revenue growth strategy makes sense.  I do not think that it is a good idea to grow an unprofitable business; scale will not help them become profitable.

Rightside Group

Rightside Group owns a wonderful business called eNom, which is the #1 wholesale registrar in the world.  eNom should be a cash cow, yet Rightside Group has managed to see its GAAP earnings go from positive to negative in the past few years according to its form 10 filing.  It loses money despite having a wonderful business.

Industry overview

There are different models in selling domain names to retail customers:

  1. Almost all web hosting companies offer highly discounted prices on domain names.  For example, I have web hosting with icdsoft.com and can buy a .com domain through them for $5/yr which is roughly half the cost I would otherwise pay.  One reason why web hosts offer discounted domain registration is because it makes it more difficult for their customers to switch web hosts.  Almost all web hosting companies buy domains from a wholesale registrar such as eNom (owned by Rightside Group).
  2. Some eNom resellers sell domain names directly to retail customers.  While I pay icdsoft.com for web hosting, I transferred my domain to namecheap.com, an eNom reseller.
  3. Some domain registration companies such as godaddy.com are vertically integrated.  They act as their own wholesale registrar.
  4. Network Solutions.  When the Internet first started, Network Solutions (owned by WWWW) was charging around $60/yr for domain registration.  Its prices have now dropped to the $35/yr range (versus competitors charging around $10/yr).  Many customers stay because they are too lazy to switch or they don’t want to risk disruption to their website.  Network Solutions will slowly lose customers over time but this is otherwise a highly profitable business.

If you want to buy just a domain name (without web hosting), big players in the field are GoDaddy and Namecheap.  I believe Godaddy is clearly #1 and eNom resellers like Namecheap are #2.  When it comes to wholesale registrars, eNom is #1.

Pricing strategies and competition

Godaddy is generally cheaper than Namecheap.  If you register a domain, you may wish to purchase privacy protection to reduce spam emails and to avoid people stalking you.  My writeup on CNIT demonstrates why privacy protection can be a good idea.  If you include the cost of privacy protection, then I believe that Namecheap is cheaper than Godaddy.  Some people prefer Namecheap over Godaddy due to better technical support and security (e.g. more difficult for others to steal your domain via social engineering or by gaining access to your email).  At the end of the day, Namecheap and Godaddy largely compete on price.

Publicly-traded domain companies

GoDaddy may go public.  Their SEC filings can be found here.  The company has been GAAP unprofitable.  Blame it on venture capitalists?

Rightside Group is currently being spun off from Demand.  It is currently in when-issued trading so its ticker symbol is NAMEV.

Web.com (WWWW) owns Network Solutions (a profitable melting ice cube) and other domain-related businesses.  Its 2010-2013 fiscal years have been GAAP unprofitable.

Tucows (TCX) is a wholesale registrar similar to eNom.  It has been GAAP profitable in the past decade, despite having lower scale than eNom.

Versign (VRSN) runs the domain registry for .com and .net (among other top level domains).  It has a disturbingly profitable monopoly.  However, its monopoly and profitability depend on contract renewals that are negotiated with ICANN, a non-profit body.  I don’t understand ICANN politics so I don’t understand the sustainability of Verisign’s future cash flows.  It has renewed its contracts as part of competitive bids in the past.  As the incumbent, it has an advantage in these bids because it has a proven track record of running its domain registries well.

Neustar (NSR) has domain-related businesses under its (poorly-named?) “Security Services” segment.  The domain-related businesses are a small part of this conglomerate.

The future winners and losers

I think the winners in this field will be the companies that manage to drive their costs down and become the low-cost provider.  The larger players (GoDaddy, eNom, Namecheap) naturally have an advantage as their scale allows them to have their fixed costs (software development) spread over a large number of customers.

There is some room for value creation in providing value-added services such as privacy protection (as well as email, web hosting, DNS services such as dynamic DNS, domain marketplaces, etc. etc.).

Rightside’s new gTLD strategy

ICANN has opened the floodgates to new “generic top level domains” (gTLDs) such as .ninja, .guru, etc.  The first round of the new gTLD program could create a maximum of 1,409 new TLDs (source).  The reason to apply is because there is the potential to wind up with a quasi-monopoly like Verisign.  So far, some gTLDs have been significantly more profitable than other gTLDs.  It is likely that a handful of gTLDs will make almost all of the money… if there is money to be made.  There is debate over whether or not:

  1. New gTLDs will be profitable (in aggregate).
  2. Whether or not new gTLDs are good for the Internet.

So far, Rightside has spent $18.2M on gTLD application fees.  (The all-in costs are a little higher because it has to hire people to write the applications, to run the domain registry, etc.)  While Rightside has been GAAP profitable in the past, it is currently running GAAP losses.  GAAP may obscure the true economic picture because:

  1. Rightside has made acquisitions in the past (these are arguably bad acquisitions).  These acquisitions cause Rightside to record amortization of intangibles as a non-cash expense.  I think that GAAP should be changed so that companies do not have to do this.  If you think that Rightside has made dumb acquisitions (and that goodwill will ultimately be impaired), then perhaps the amortization of intangibles does not distort earnings much.
  2. It did not capitalize the costs of preparing gTLD applications and the start-up costs of running gTLD registries.

Current success of Rightside’s gTLDs

I believe that Rightside’s most popular new gTLD to date seem to be .ninja.  There are currently 12,947 .ninja domains (and rapidly growing).  The revenues on .ninja domains are higher than .com domains.

Ntldstats.com has information on the number of domains registered for the new gTLDs.  Here is their page for .ninja.

As a point of comparison, zooknic.com has some information on the growth of older TLDs such as .info and .biz.

Suppose that .ninja grows from 12,947 domains to 150,000 domains (more than 10X), revenue of $13/domain, and will have Verisign’s operating margin of ~55%.  This equals to an operating profit of $1.0725M/year.  That isn’t quite a home run.

The bottom line

DMD:  Bad business and bad management.  However, some of its assets are worth something so this is not a zero.

NAME:  eNom is a wonderful business.  Unfortunately, management may do dumb things and manage to lose money.

Links

There is some interesting discussion about this industry at the Corner of Berkshire and Fairfax forum.

The Web Value Investor blog has some very good analysis of this industry given that the author runs easydns.com.

Advertisements

8 thoughts on “Notes on Demand Media and Rightside

  1. Why do you assume Verisign’s margins of 55% for the new gTLD domains? Isnt the whole point that ICANN has less control over pricing/margins on these domains. I believe the mgmt presentation on Rightside showed expected margins greater than 90%.

    What do you think of the stock based comp at Rightside?

    Rightside is effectively moving up the value chain to become more like Verisign, a registry operator which seems like a more interesting business. Of course,whether or not gTLDs become popular and used etc remains to be seen

    • 1- The presentation is misleading. (This is one of the reasons I don’t like management, because I don’t like people who try to mislead investors.) Obviously there are operating expenses when you run a domain registry. The margin they show is before operating expenses. Verisign’s .com business may have higher margins than the new gTLDs due to scale; we’ll see I guess.

      2- I think insiders get paid too much.

  2. Hi Glenn,

    Industry insiders have informed us that the aftermarket business on NAME is basically a 100% margin business, because it costs them virtually nothing to advertise on parked domains and sell domains they already own. Therefore, Service Costs, on the income statement, is almost entirely attributable to the Registrar business. If NAME’s DNS revenues/domain is $10, as disclosed in the Form 10, and Verisign charges $7.85 per .com domain (as disclosed in the VRSN 10-K), this would imply something around $2/domain of gross profit on the registrar business. But looking back at the financials since 2009, it looks like they have never made any gross profit on this registrar business except for a small one in 2011. What else would be going into “Service Costs” that would eat away at the rest of that margin? (D&A is only a small piece of it) Secondly, if there was never any gross profit on the registrar business, was the only reason it existed for the purpose of being a platform for Aftermarket services (which are now in secular decline)?

    Any insights here would be much appreciated.

    Thanks.

    • The 10-Q describes the costs that go into “Service Costs”.
      Service costs primarily consist of fees paid to registries and ICANN associated with domain name registrations, revenue‑sharing expenses, Internet connection and co‑location charges and other platform expenses including depreciation of the systems and hardware used to build and operate our services, and personnel costs related to customer service and information technology. Our service costs are dependent on a number of factors, including the volume of domain name registrations, value‑added services supported by our registrar service and the revenue share agreements we have with our domain name customers utilizing our monetization platform. In the near term, we expect higher overall registration costs as a percentage of revenue due to the recent growth in higher‑volume, lower‑margin customers as well as a mix shift of revenue toward lower‑margin domain name services revenue relative to aftermarket service revenue.

      • 1- The aftermarket business is not 100% margin. Perhaps you are misinterpreting somebody’s comments. I’m guessing the biggest cost there is to pay Verisign (and other registries) for the domain registration fees. For Verisign it’s several bucks a domain. There should be very small costs in web hosting, IT, web design, etc.

        What they do is to take over the registration for domain names that their customers stop paying for. A lot of people will let valuable domain names lapse because they don’t understand that their domain name can be monetized for a few hundred bucks or whatever (on average it is probably less). It may not even make sense for their webmaster to learn how to monetize the domain name because it would take time to learn and to set everything up. But a lot of it is simply because people don’t understand the value of what they have.
        They probably develop some automated system to figure out which domains are worth taking over and which are worth lapsing. Most domains are not worth keeping.

        2- The wholesale registrar business, on a standalone basis, is probably capable of being profitable. Tucows is doing it.

        3- Perhaps the financial statements are unclear because all of the whole registrars are usually involved in adjacent businesses. So if you are in the registrar business, you definitely get into aftermarket.

        Demand Media, the former parent of NAME, was in the web content business. And part of their strategy was using lots and lots of domain names to rank higher in Google. So they registered a lot of domain names for their own use.

  3. Ok thanks for your response. So how should we think about valuing the registrar business (separate from the new gTLD registry business) if it currently isn’t profitable but is a wonderful business? Thanks.

    • It may or may not be profitable. I suspect that it is. Certainly it “ought” to be profitable given its scale and market share.

      I don’t think that NAME is a great long at current prices. I definitely wouldn’t short it. This isn’t a pitch you have to swing at.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s