How Google ripped us all off

I wrote a few days ago about the Texas AG's lawsuit against Google for monopoly practices in the Advertising market and the academic Stanford Technology Law Review article that clearly forms the basis for that law suit. I also put out a podcast about it and one particular graph in that article.

Google revenue share graph.png

This graph looks very simple, but it needs to be explained carefully. I'll also put the original footnote from the academic paper by Dina Srinivasan down below.

In their Dec 31st 2004 Annual Report, Google was able to say this:

We generate revenue by delivering relevant, cost-effective online advertising. Businesses use our AdWords program to promote their products and services with targeted advertising. In addition, the thousands of third-party web sites that comprise our Google Network use our Google AdSense program to deliver relevant ads that generate revenue and enhance the user experience.
We share most of the revenue generated from ads shown by a member of the Google Network with that member—creating an additional revenue stream for them. Web sites can also license our Google WebSearch product to offer the Google search experience to their users. The key benefits we offer to web sites in the Google Network include:

Access to Advertisers. Many small web site companies do not have the time or resources to develop effective programs for generating revenue from online advertising. Even larger sites, with dedicated sales teams, may find it difficult to generate revenue from pages with specialized content. We believe that Google AdSense enables Google Network members to generate revenue from their sites more effectively and efficiently. Google AdSense promotes effective revenue generation by providing Google Network members immediate access to Google’s base of advertisers and their broad collection of ads. As soon as a web site joins the Google Network, our technology automatically begins delivering ads for posting on the member’s web site. The automated nature of our advertising programs promotes efficient revenue generation. Our online registration systems enable web sites to easily join the Google Network and our ad serving technology allows automated delivery of ads for posting on the member’s site. The Google Network member determines the placement of the ads on its web site and controls and directs the nature of ad content.

In that year of 2004, Google's share of the total revenue it generated in advertising came 50% from it's own properties, at that time limited mostly to search. The other 50% came from the "Google Network members", i.e. websites who were suppliers of eyeballs to Google's adverts. However by 2004 they were already "enhancing" search to make you stay on Google longer.

If Google were a normal business, buying and selling something tangible, you'd say they were paying about half their revenue to their suppliers for the goods they sold.

There were two top level ways to make more money:

  • keep people on Google properties for longer and
  • pay Google's suppliers less.

Google did both of those AND a third way: rig the market.

The first scheme was already underway in "enhancing search". Obviously the less often someone has to leave Google to look at another website, the more time Google gets to show advertising to them from its own website and the more it can chip away at the 50% of its revenue that were leaving the building.

So nice that they gave so much away for "free". And just so you understand, this is the size of the pie they were playing with in 2004:

Google revenues 2004

Total revenues were a little over $3bn in 2004 and as the graph showed, split roughly 50/50 between Google and 3rd party sites giving Google eyeballs to monetize with advertising. As the following breakdown shows, that $1.554bn in revenue came at the following cost:

Google Traffic acquisition costs 2004

The $1.228bn listed as Traffic acquisition costs is the amount Google paid out to Google Network web sites. That's a total of 39% of their revenue. In 2005 that number was $2.115bn but only 35% of their revenue.

Jumping forward to 2019 that number is now 18.6% and in real numbers that's $30,089,000,000 being paid out to the SUPPLIERS of eyeballs and viewers out of Google's total revenues of $161,857,000,000. Google doesn't really make anything. They resell the content of others with a huge mark up.

In fact, Dina Srinivasan was too kind to Google with her graph. My version plots this number instead and I went back to 2002, that's important because I think it starts to show what went on inside Google.

% of Google Ad Revenue Paid to 3rd Parties

What Happened in the run up to 2004?


Communication is another such area. People increasingly rely on the Internet to communicate with each other. Gmail, our new email service (available in a limited test), offers a gigabyte of free storage for each user, along with email search capabilities and relevant advertising. Delivering an improved user experience in Gmail has similar computing and software requirements as our search service.

So here's what I believe happened. Google grew rapidly with its search engine but soon hit a limit to how much money it could make monetising only its own website properties. It had to get more real estate to show the adverts it was so adept at selling, programatically in bulk.

It branched out to running adverts all over the web and paying those other websites quite well. Too well. It watched as its own share of the revenue started falling (the dip in the red line above from 2002 to 2004.

Boosting Revenue by cutting costs

How do you boost revenue? Cut the amount you pay to suppliers and stop needing to pay suppliers. Google did both. It extended the "free" services and information offered when you ran a search on Google. Slowly but surely you found more of the information you wanted right there on the Google page, no need to click through to a news site or a sports result site as you already had the headline or the score.

Boosting Revenue by swapping expensive inventory for cheaper

Then Google added services for which you needed an account, first up Gmail, again for "free" at a time when this was a paid service. This time you logged in, suddenly you are trackable at all times. This becomes hugely important when we get to how they cheated. Quickly they followed up with the whole suite of Google Docs and a myriad of others, all of which required logged in users.

The percentage of time spent on Google products went up, the rest of the web went down. And the dollars flowed along with attention.

Then YouTube, Chrome and more and more free trinkets. Oh how benevolent was Google, look what we got for "free". When YouTube started giving a small share of advertising revenue to the real content creators who drove the traffic, it was always an unequal bargain. This was never a major TV network negotiating with advertisers, all the power rested with Google and they always knew they could turn the dial and decide who much money to keep and how much to give to the suppliers of their eyeballs. Those eyeballs (and our private information) being the product which Google sells day and night, billions of times a day, to their only true customers, the advertisers.

And Android, it's a side story but of course it was essential that Google get ahold of that market.

The Third Way

But the final trick in the Google playbook was the one that is hardest to explain. It brings me back to Dina Srinivasan's paper and to the Texas vs Google law suit.

In 2009 Google bought the across the web advertising market place Doubleclick. That was the start of allowing Google to run what I think of as the largest, fastest and most corrupt trading market in human history.

I will lay out the details of this in another post, this one is already too long, but just imagine quickly a stock market or a crypto trading market which takes unspecified and unknowable commissions on every trade. Where ever single web page load can result in 20 or more trades happening in 300ms to 1500ms. Where Google runs the market and the sales brokers and the buy side brokers and has the lions share in every step of the trade. A level of insider trading unimaginable in a reputable stock market.

You do need to know one vital point: knowing who is looking at a web page can increase advertising bids by more than double. Which is why having you logged in to your Gmail or YouTube or anything else matters greatly. And at the same time, on the pretext of your "privacy" they've been steadily blocking third parties from accessing information on their advertising markets that would help independent markets and advertising suppliers compete. Brilliant.

All of this was built by Google with some cooperation from Facebook when it looked like Facebook might be able to challenge the central market where most of Google's money is made today.

And Texas's AG is definitely heading toward a huge show down with Google.

But after all this I'll tell you something else: our comparatively tightly focused #CryptoClassAction @jpbliberty lawsuit against Google and Facebook is absolutely relevant and probably easier to fight and we have a head start.

This is Dina Srinivasan's footnote accompanying her graph. I've gone through the Google Annual reports 2002 - 2019 myself and confirmed this work.

In Google’s annual 10-K SEC filings, Google breaks down its advertising revenue as going to “Google properties” or “web sites of Google Network members.” The term “Google Network members” refers to non-Google websites on which Google places advertising. In its 2017 10-K, Google explains that it generally accounts for third-party revenue on a gross basis: “For ads placed on Google Network Members’ properties, we evaluate whether we are the principal (i.e., report revenues on a gross basis) or agent (i.e., report revenues on a net ba-sis). Generally, we report advertising revenues for ads placed on Google Network Members’ properties on a gross basis, that is, the amounts billed to our customers are recorded as revenues, and amounts paid to Google Network Members are recorded as cost of revenues. Where we are the principal, we control the advertising inventory before it is transferred to our customers. Our control is evidenced by our sole ability to monetize the advertising inventory before it is transferred to our customers, and is further supported by us being primarily responsible to our customers and having a level of discretion in establishing pricing.” In 2004, Google buying tools allocated approximately 50% of advertising revenue to Google's proprietary properties, such as Search, and the other 50% to non-Google websites selling their ads through Google's buying tools and advertising exchange. Google Inc., Annual Report (Form 10-K) (Mar. 30, 2005), It was in 2006 that Google acquired YouTube. Andrew Ross Sorkin & Jeremy W. Peters, Google to Acquire YouTube for $1.65 Billion, N.Y. TIMES (Oct. 9, 2006), In 2005, Google's share of advertising revenue increased to, approximately, 55%; 2006, 60%; 2007, 65%; 2008, 68%; 2009, 68%; 2010, 68%; 2011, 71%; 2012, 71%; 2013, 73%; 2014, 75%; 2015, 77%; 2016, 80%; 2017, 81%, 2018, 82%; 2019, 84%. Google Inc., Annual Report (Form 10-K) (Mar. 16, 2006),; Google Inc., Annual Report (Form 10-K) (Mar. 1, 2007),; Google Inc., Annual Report (Form 10-K) (Feb. 15, 2008),; Google Inc., Annual Report (Form 10-K) (Feb. 13, 2009),; Google Inc., Annual Report (Form 10-K) (Feb. 12, 2010),; Google Inc., Annual Report (Form 10-K) (Feb. 12, 2011),; Google Inc., Annual Report (Form 10-K) (Apr. 23, 2012),; Google Inc., Annual Report (Form 10-K) (Jan. 29, 2013),; Google Inc., Annual Report (Form 10-K) (Feb. 11, 2014),; Google Inc., Annual Report (Form 10-K) (Feb. 6, 2015),; Google Inc., Annual Report (Form 10-K) (Feb. 11, 2016); Alphabet Inc., Annual Report (Form 10-K) (Feb. 2, 2017),; Alphabet Inc., Annual Report (Form 10-K) (Feb. 5, 2018),; Alphabet Inc., Annual Report (Form 10-K) (Feb. 4, 2019),; Alphabet Inc., Annual Report (Form 10-K) (Feb. 3, 2020),

brianoflondon hive footer.png