June 10, Chicago — I was at IRCE (Internet Retailer Conference and Exhibit) for three days. It’s a large show for ecommerce on the Internet. There were 650-700 booths and perhaps 10-20,000 visitors. One afternoon, I walked up and down every aisle and looked at every booth to see what is going on in ecommerce.

I spend so much time on SEO and PPC (Google Adwords) that every once in a while, I look up and realize there are entirely new kinds of things on the web. That’s one reason that I go to tradeshows: to learn new things.

The new thing is marketplaces. Some of you already know about these. But some of  you don’t. I talked with various friends in Silicon Valley this weekend and only a few knew a bit about these.

“Marketplaces” are large websites that allow other companies to set up shops within their website.

For example, you make shoes so you set up your shoe store website inside Amazon. The Amazon marketplace handles product descriptions, photos, the catalog, warehouse storage, ordering, transaction (payment), shipping, and returns, all for a small fee. You can use WooCommerce and other tools in WordPress, Wix, Mailchimp, and so on to build and connect your store to the marketplaces. There are a bunch of tools and services, including agencies for this.

Marketplaces grew 19% in the past year and now account for 33% of all US ecommerce sales. Yes, they’re big.

Where Do Companies Advertise?

US companies spent around $60B in digital marketing in 2015. Where is that money going? Here is data from eMarketer (Sept. 2015):

  • 40% Google
  • 13% Facebook
  •   5% Microsoft
  •   4% Yahoo!
  •   2% Twitter
  •   2% Amazon
  •   1% Yelp
  •   1% LinkedIn
  •   2% AOL

Google got 40% of the $60B. That’s good, right? No, wait, that’s not good.

How Do People Buy Stuff?

The issue is how people buy stuff. According to Bloom Research (2015):

  • 81% of US purchases start with online search
  • 44% of consumers start at Amazon
  • 34% of consumers start at search engines
  • 21% start at store sites

Do you see the problem? Only 34% of consumers start with a search engine. Google has 64% of the US market, which means Google gets 22% of US purchase searches. Not good. This means Google’s 40% share of ecommerce ad spend is reaching only those 22% of consumers who go to Google.

Meanwhile, 44% of consumers go straight to Amazon: that $22B that was spent in Google didn’t reach them. Google Adwords looks big but it actually has only half the reach of Amazon. Amazon Merchant Services accounts for 40% of Amazon’s revenues.

There’s more bad news.

How Big Is Amazon?

Amazon Prime is Amazon’s member service. You pay $99/year and get unlimited, two-day free shipping, free unlimited movies, free unlimited music, unlimited file storage, 20% discounts on many new things, and more stuff. According to Internet Retailer Survey (April 2016):

  • 60% of US households are members of Amazon Prime
  • 50% of their shopping is at Amazon
  • 40% go first to Amazon
  • 30% buy only at Amazon and nowhere else

Whoa. 30% of consumers don’t even bother to look at Google or go to a store? $22B in Google advertising isn’t reaching them.

And yes, there’s more bad news.

The Ten Largest Marketplaces

Amazon isn’t the only marketplace.  Amazon Merchant Services has two million stores with 365 million products (SKUs) and $224B in annual sales. eBay has 900m products and 25 million vendors. NewEgg has ten thousand vendors, 27 million products, and $2.9B in annual sales. Here’s a table from Internet Retailer (April 2016):


The top marketplaces in April 2016. SKU = Stock Keeping Unit, or a product, which is pronounced “skew”. Just so you know, a large supermarket has around 50,000 SKUs. Bluefly, the smallest marketplace with 100,000 SKUs, has twice as many products as a large supermarket. Jet is the size of 200 large supermarkets.

Look carefully at that list of the ten largest marketplaces. What’s missing?

Yep… Google. Have you ever heard of Bluefly or Jet? They’re bigger in ecommerce than Google?

Okay, a side note here. I knew that there were ecommerce stores in China. Merchants set up shops in Alibaba, Taobao (80m SKUs), WeChat, and so on. I knew this has been going on for five years or so, but I thought it was just one of those things they do only in China. I wondered why merchants could set up in WeChat but there was nothing like that in Facebook. We didn’t notice it was happening in the US as well.

New kinds of companies are entering this space. For example, Marketplacer.com from Australia builds marketplaces for companies. If Safeway wants to start a marketplace, Marketplacer.com builds the infrastructure.

What This Means for Google

Okay, here comes a bit of search engine theory. Just take notes.

In the early 2000s, Altavista researchers found there are three kinds of searches:

  • Informational: You want to learn about something. Who was the president of France in 1934? What’s the best way to clean a silk shirt? About 80% of searches are informational.
  • Navigational: You want the official website for a person, organization, company, product, city, national park, and so on. You want the website for Yosemite National Park, so you search for Yosemite. 10% of searches are informational.
  • Transactional: You want to get something. Either buy something or download a song, movie, book, PDF, app, software, and so on. You search for it and the search engine shows you the webpage that sells it. 10% of searches are transactional.

This three-part model of search is basic user behavior at all search engines.

You see now? Yes, a very big problem.

In 1998, Google was just a small startup that built a search engine which they wanted to sell so they could go on to do something worthwhile (really). They went around to the big companies and offered it for $1m. Yes, less than the price of a house in Palo Alto. And everyone turned them down. “There’s no money in search engines.”

Many people think this was crazy: of course there was money in search engines: Google makes $50B every year. Don’t they?

But it’s true: there’s no money in search. Most searches are informational or navigational and Google doesn’t make much money on those. Who cares about the president of France in 1934? There’s little advertising for Yosemite or Justin Bieber. Any search engine just points to those. Which is why DuckDuckGo and all of the other search engines make very little money.

Google makes money on transactional searches, which is that ten percent of searches where people are looking to buy something. Only a few companies can get to the top of the search page, so to make sure they’re at the top, they pay Google Adwords for advertising. Google makes money on ads, not the search. Google is not a search engine. Google is an ad publisher.

From 2002 (the beginning of Adwords) up to 2015, if you want to buy something, you Googled it, saw the ads, and clicked the ads. That made Google #1.

But no more. Amazon Prime is #1. More people buy at Amazon than search for products at Google.

Want to hear something funny? Amazon Prime was a lucky idea. In 2005, a staffer at Amazon suggested it and Amazon tried it. 30% of consumers now buy at Amazon and don’t bother to look anywhere else, not in Google or local stores. Amazon memberships created tremendous customer loyalty.

So the other marketplaces are trying the same thing: Build a platform to allow hundreds of thousands of vendors to sell tens of millions of products.

MailChimp, Amazon, WordPress, Adobe, IBM, FedEx, Visa, MasterCard, UPS, and many others were at the IRCE trade show in Chicago. But Google, Facebook, and Microsoft were not there.

Facebook has a chance in the game. They could set up a marketplace. Facebook has the audience, the advertising platform, and relations to companies.  They had a marketplace back in the late 2000s, but they gave up on it. What about Google? Google could buy one, such as eBay or Etsy. eBay would be a good target: it fell 2% in a market that’s growing 19% (yes, that’s really bad). But this would put Google in competition with its advertisers which would kill its ad engine.

And then there are the Chinese. Alibaba is twice as big as Google and Amazon combined and will soon be the first trillion dollar company. WeChat has 600m users, which is twice the population of the US. These Chinese companies could buy the large US companies.

So, if you’re selling products, look into marketplaces. Set up your store in all relevant marketplaces (not just Amazon). Learn about advertising in the marketplaces (for example, Amazon has its own ad platform).

What about SEO and Adwords? Keep doing them. After all, it still reaches 22% of US shoppers.

A New Web?

So what did I mean by “a new web”? I thought about why I hadn’t yet noticed marketplaces. In Silicon Valley, we pay attention to search engines and social networks, i.e., Google, Facebook, and the social sites. The goal is to get many to make money on advertising. As for actually selling something? That’s for other people.

So much, okay, too much, of the Silicon Valley web is clickbait: creating sites, pages, and stuff to get people to click. In the 90s, we called it eyeballs. Clickbait is junk. Just as the chase for eyeballs led to the dotcom crash, clickbait will also crash for the same reason: the number of people doesn’t change, but the amount of ads and clickbait keep increasing, which lowers the value of the clicks.

Ecommerce however is based on reality: People pay real money to buy real products. Amazon’s revenues for 2nd quarter in 2016 are around $30B. That’s $120B per year, which is more than double Google’s annual revenues. Google is big? No, Amazon is bigger. And Amazon sells real stuff, whereas Google is advertising.

So maybe we’re entering a new phase of the Internet.