Monday Note

The ARPUs of the Big Four Dwarf Everybody Else 2019-02-11 07:20:27

by Frederic Filloux

Ready for the ARPU squeeze. Photo by Alex Loup on Unsplash

Let’s give a look at the most important metric of all: the Average Revenue Per User.

This week, I wanted to write about the recent shifts in the business model of news: the end of the advertising model, the rise of subscriptions, the difficulty of substituting the disintegration of local media with small digital outlets, and the limits of the non-profit system. Through my research across multiple financial statements and industry reports, I’ve discovered that the most compelling piece of data is the evolution of ARPU over the last year and since 2011.

In case anyone doubts it: the FAANG (Facebook, Amazon, Apple, Google) has been making a killing thanks to their domination of the advertising business (Netflix earns a special mention for its mastery of the subscription model.)

Let’s start with some context by looking at the period 2011–2018 on the US market:

  • Over the past seven years, Digital ad spending in the United States rose from $32B to $111B; a 3.5x increase.
  • Google, which had already a stronghold on the US market seven years ago rose from $18B to $63B in revenue, also a 3.5x increase.
  • Facebook, which was created eight years after Google, catapulted from $2.1B to $27B, a 13x rise.
  • The New York Times digital advertising revenue rose from $233m in 2011 to $259m, a modest 11 percent increase. See the chart below (the perimeter has evolved: for the FY 2011, I took the NYT “News Media Group”, which excluded properties such as the Boston Globe):

To sum up the growth differential for the last seven years in the US market:

I included the New York Times because it is supposed to be the gold standard for digital advertising with a vast range of products from programmatic to sophisticated branded operations.

As expected, these numbers show that for the past seven years the duopoly snatched most of the growth of digital ad spending:

For each dollar of digital ad growth between 2011–2018, Google took $1.00, Facebook $3.70, while the NYT took only 32 cents.

Now, let’s turn to the Average Revenue per User (ARPU) per year with this chart:

Now let’s dive into some details:

Amazon is the king of ARPU with its giant retail operation which brought last year $752 per customer on average for its global 310 million customers. There is a significant imbalance in Amazon revenue due to the difference between the casual Amazon user and the Prime subscriber, the latter paying a $100 membership per year and buying many more products. But the most interesting part is the growth of Amazon’s advertising business: it produced an ARPU of $15, which is five times the ad revenue of most news outlets, including large audience brands.

Google’s global advertising machine is the most sophisticated and profitable in the world. Its ARPU is sky-high and shows no signs of erosion. Too bad the company doesn’t provide a detailed breakdown by line of business and geography. As a comparison, domestically, a publisher would need to charge $21 per month to make the same revenue as Google.

Facebook’s revenue per user is roaring. For Q4 2018 vs. Q4 2017, it’s global ARPU increased by 19 percent to $7.37, and the US-Canada ARPU by 30 percent to $35. Between 2011 and 2018, the social network global ARPU rose nearly 6x, while the US-Canada grew 11x. With a domestic revenue per user of $112, Facebook is the equivalent of a publisher charging $9 a month. It does that with a free service.

Apple’s master analyst and writer is Horace Dediu from Asymco. Here is what he explained recently:

“In the latest letter to investors, Tim Cook wrote that the total number of active devices has increased by more than 100 million in the last twelve months [of 2018]. Given that the figure a year ago was 1.3 billion, the current total must be 1.4 billion. We can also estimate that Apple sold about 303 million devices in 2018. This implies 2 out of 3 were replacements and 1 out of 3 increased the base.

Services revenues were about $40 billion in 2018 (up 28%). Total revenues were $261.5 billion.

Thus for the whole of 2018:

  • $194 per active device of which…
  • $30 in service revenues per active device”

To put it another way, with its service business (music, apps, etc.) Apple is making 10x more than a decent publisher makes on advertising. And given the growth (+17 percent Y/Y), Horace Dediu does not exclude a service-related ARPU of $50 per device owner and per year in the near future.

The New York Times is doing just fine with its digital subscription business with 3.3 million customers paying for its digital services, including 2.7 million who pay only for the news service. The news product generated $139.50 per subscriber last year. However, while the Big Four manage to have a double-digit growth of their ARPUs, the NYTimes’ revenue per user has eroded by 4 percent last year ($146.10 in 2017). Interestingly enough, while the revenue of the Times’ digital news subscription grew by 16 percent to $378M (details here), its other digital product segment, while much smaller ($22M in revenue) grew by 54 percent last year. Diversification works when it is carried by a brand like the NYTimes.

Spotify Here I leave the comment to Tim Ingham, who wrote an excellent piece in Rolling Stone last month.

“According to an F-1 SEC filing, Spotify’s monthly ARPU had officially fallen to €6.84 at the close of 2015; a year on, it tumbled again to €6.20, and in 2017 it hit €5.32. And in Spotify’s latest fiscal update, for the third quarter of 2018, ending September 30th, the company posted an official monthly ARPU figure of just €4.73 ($5.50).

This means that the average Spotify subscriber around the world is now paying more than $20 less per annum than he was four years ago. And the record labels fear this situation is about to get worse.”

Netflix has, without any advertising, made $131 per viewer globally last year vs. $116 in 2017, an increase of 13 percent. In 2011, the streaming giant was making $133 per user, but the acquisition costs since then have soared from $16 in 2011 to $180 in 2018. The competition from other (Disney, Amazon, Apple…) is looming.

Buzzfeed is an example of a digital outlet that was supposed to revolutionize online advertising with a clever social propagation of content. Their plan didn’t work as expected. The $3 in ad revenue per year and per user is a generous estimate. After off-loading a sizable part of its staff, the fancy news outlet now wants to get married to others of its kind. I am certain that, business-wise, the journalistic brand (BuzzFeed News) has been dragged down by the mediocrity of the main outlet (listicles and al.) and the pressure to make clicks. Too bad.

The ARPUs of the Big Four Dwarf Everybody Else was originally published in Monday Note on Medium, where people are continuing the conversation by highlighting and responding to this story.

Read more

50 Years In Tech. Part 16: Be Fundraising Misadventures 2019-02-11 02:36:10

by Jean-Louis Gassée

Sneering at venture investors, calling them names such as Vulture Capitalists, is a long-standing Valley pub topic. But, as we’ll see today, cynical pros are less dangerous than a naive entrepreneur taking money from friends.

Summer 1991. Be, the company that Steve Sakoman and I had started the year before, is ramping up, but I’m merely the CEO writing checks, not code, so I take a short family vacation to Arcachon in southwest France.

I’m daydreaming past the seaside shop fronts when something catches my eye. I back up, peer in the window of the antique store, and there they are: Two ceramic pigs, about a foot tall, dressed as butchers. The perfect avatars of my disdain for the Venture Capital profession. Unsubtly, I christen them Victor and Charles:

After all these years — and even after becoming a member of the VC brotherhood — they’re still on my desk. If only they could talk…

I wanted to keep Be out of the vulture capitalists’ talons, so to fund the company in its early years, I put my own money into the venture. That was the first of a series of fundraising mistakes.

My first excuse is that I used to be French, from a distrustful culture. In France, the entrepreneur were often viewed as a kind of mountebank who is keen to take advantage of investors’ funds. To prove their good faith, French entrepreneurs were expected to put their own money on the line. (It’s better now.)

So I thought I was doing the right thing, but, as I found out, professional investors in the US are suspicious of self-funding. They prefer a clean division of labor: The entrepreneur provides the idea, the psychic energy, the leadership; the pros supply the financial fuel.

When my personal coffers began to run low, I accepted investment money from friends and business acquaintances. I’m still moved by the memory of a friend signing a second check on the trunk of his rental car after visiting Sakoman’s lab in Scott’s Valley. But taking money from friends can be lethal: The money always runs out, friends don’t have the pros’ deep pockets and the company becomes vulnerable in a way I’ll explain a bit later.

But, first, the Law of Professional Venture Investing.

A Pro invests as much money, as many times, for as long as required for the situation to attain Clarity. This Clarity allows only two outcomes, Dead or Liquid. When a Pro declines the invitation to invest in a follow-on round, it means they’ve reached Clarity: In the eyes of the Pro who declines to invest, the company is Dead, their initial stake is worthless. Write it off and move on, no tears, no recriminations.

But if another Pro is willing to keep the company alive by putting more money at stake, they do so seeing an elevated risk, they pay less, often much less for the new shares they buy. As a result, original investors will be diluted by the new money: 10% share in a previous round might now be worth 2% or less.

This might seem counterintuitive and unfair: Shouldn’t the original investment, the one that meant taking a risk on an unknown, be worth more than those of the newcomers? This is another part of The Law: If a company needs additional rounds of funding because it’s not growing as well or as fast as anticipated, it is now seen as a riskier play. As a reward for taking a risk on a venture that is now shown to be foundering, new investments get a higher percentage ownership per dollar. (None of the current discussion applies to follow-on rounds for a happily growing entity.)

Back to Be.

A good friend introduced me to the COO of Crédit Lyonnais, France’s largest bank at the time. The meeting went very well and the senior exec told the bank’s venture arm to invest in Be.

I was thrilled, we had deep, friendly pockets, and the deal struck a high valuation that I was proud of…for a while. I didn’t realize that it would discourage other potential investors who couldn’t see a way to a significantly higher multiple of the Crédit Lyonnais terms. Another misstep.

But wait, there’s more…

In 1993, Crédit Lyonnais went belly up right at the time when we needed more money to move beyond the end of Sakoman’s Hobbit-based design (see Part 15 of this series). Crédit Lyonnais couldn’t put more money into the company. In effect, our lead investor was dead and its failure almost killed us.

Seeing the bottom of the cash drawer I once again went on a fundraising campaign, this time in a weakened position. Thankfully, friends extended help; my spouse, Brigitte, felt we couldn’t let the product die and agreed we could put more money at stake. Desperate to keep the company going, I even gave a presentation to a French investor three days after I was released from Stanford Hospital, unshaven and with metal clips on my neck where the neurosurgeon had patched my carotide artery. We kept the company alive, barely, with a motley crew of backers.

Then an Apple alumna saved us.

In the Summer of 1995, I had lunch at Menlo Park’s Cafe Borrone with Lia Lorenzano (now Lorenzano-Kennett), the logistics director of a now deceased industry conference called Agenda. Although I protested that Be wasn’t ready for the limelight, she wouldn’t let me go until she had convinced me to present at the next Agenda a couple of months later.

I went back to the Be offices across the street and confessed my just committed sin to the horrified engineers. “We’re not ready!”. It didn’t matter. We were on the Agenda.

I flew down to Scottsdale a weekend days before the conference in order to prepare…but I was blocked, paralyzed with stress. Fortunately, my good friend Jean Calmon, the Apple France Sales Manager who had joined Be to help us with the European side of our business, put together a slide presentation. Be employees and my spouse flew in right before the conference started. On the day, Steve Horowitz, one of our earliest engineers, gave a masterful demo while I stumbled through Jean’s slides. (Later, Horowitz would give the first video demo of Android.)

The response was astonishing: We got a standing ovation, only the second one in the conference’s nine years. I was left close to tears and (almost) speechless. The multimedia performance we had dreamed of when starting Be stunned the crowd. We got good press, including a nice NY Times article. And, at long last, we felt actual interest from the Silicon Valley professional investors who were to give us a new lease on life.

The lead venture investor in the next round was August Capital’s David Marquardt, the only VC who had invested in Microsoft prior to its 1986 IPO. (He was also a Microsoft Board member, a position that would make life “interesting” from day one of the relationship.)

The friends and acquaintances who had invested in Be were thrilled to see support from Valley VCs…but then they saw the terms. I had the humiliating task of telling them that Marquardt wanted a “clean” capital investment that would dilute previous ownership by 300! If they refused to sign off on the terms, there would be no more money, Be would be dead. (I was also told the money I had lent the company “stayed”, meaning it was gone with the other investors’ money.)

To add to our investors’ hurt feelings, the Be team, yours truly included, would receive fresh, undiluted stock options. In a way, this was an application of The Law: We kept investing (ourselves) in the venture and would be treated as “new money”. But it left a sour taste in the mouths of some earlier investors, included Crédit Lyonnais’ venture arm that should have known better: “We supported you and now we’re left with nothing while you go off with glorious new money, fresh stock options, and media attention!”

Everyone reluctantly but helpfully signed off, but any one of them could have killed the deal — and Be. To this day, I’m grateful and embarrassed for the consequences of my naiveté.

Victor and Charles might be difficult, greedy, and whatever other negatives you might want to throw at them, but they’re deep-pocket pros who serve a purpose. And there are many to choose from in the Valley. The trick is to choose wisely.

In the next part of this series, we’ll look the last part of Be’s life.


50 Years In Tech. Part 16: Be Fundraising Misadventures was originally published in Monday Note on Medium, where people are continuing the conversation by highlighting and responding to this story.

Read more