Discover more from Growth Memo
#227 - is Google really that good?
🗞️ Documents publicized in the recent antitrust case revealed Google paid Apple, Samsung, Mozilla, AT&T and others a collective $26.3b in 2021 to be the default search engine. That’s 10% of revenues and a third of net income (profits).
It’s hard to divorce from defaults once you’ve been married for a while. Despite cheaper and better alternatives…
… Over 60% of consumers would stick to their current energy provider. (source)
… 75-80% of consumers would stick to their car insurance. (source)
… 70% of consumers would stay with their general practitioner. (source)
Every one of us carries a strong bias toward the Status Quo. We’re afraid of making bad decisions with worse outcomes (think: loss aversion), so we stick to the current state and develop a firm commitment.
It also takes a lot of energy to detach. You have to think about why you want to change, research alternatives, overcome choice overload, evaluate alternatives, and make a conscious decision to switch.
Defaults impact our decisions everywhere:
Email opt-in (forbidden in some countries)
Subscribe instead of one-time purchase (example: Amazon)
Recommended tier on pricing pages
Social media privacy settings
Being the search engine default works so well for Google because the trade-offs are not clear. Searchers don’t know what they lose out on because they’ve used Google for a long time. If presented with a clear description of the differences between Google and other search engines, searchers might make different choices.
Is Google really the best search engine?
There are two options for why Google pays the equivalent of Costco’s, Intel’s, or Coca-Cola’s profits to be the default search engine every year: One, grab as much market share as possible; two, Google’s isn’t that much better than other search engines. The answer: yes.
Last year, Surge AI tasked 250 raters with rating recent queries from their browser history and comparing Google’s with Bing’s results. 20% of search results had the same quality in both search engines, 18% were better on Bing and 28% on Google. Google does better. Somewhat. (source)
The quality of results is better, but the experience on Google has taken a hit and user behavior is changing. Key Google executives know users aren’t happy. (source)
The reasons for the decline of Google’s user experience:
1/ Ad load
The problem with advertising as a business model: if you want your company to keep growing, you need to show more ads or have more users. Most people on the planet already use Google (minus China, Russia, and a few other geos), so the load of ads on Google is growing and cuts into the user experience.
Alphabet’s recent earnings report proves it: Search up +11% Y/Y (as predicted, YouTube earnings beat Google Network for the first time). Nothing short of impressive.
Side note: I can’t help but think OpenAI adopted subscriptions as a monetization model as fast as possible because it saw Google maneuvering itself into a corner with ads that deteriorate the quality of search results.
2/ The web has changed
Marissa Mayer, a former key executive at Google, on the Freakonomics podcast:
“My search results just don’t seem as useful. I feel like I’m seeing more ads, more links that might as well be ads, and more links to spammy web pages.
[...] I do think the quality of the Internet has taken a hit.
[…] The real question is, why is the web getting worse?” (source)
More activity has moved from forums and blogs to apps. Instead of writing on the open web, more people post on social networks. The last big forums, Reddit and Quora, have such an outsized impact that searchers append “reddit” to their queries. Who’s left? Publishers, a few bloggers (hello), indexable newsletters and affiliates.
My take: Google launched Discover and “feeditized” the search results as a response to TikTok and YouTube getting more market share amongst younger audiences.
3/ “SEO farms”
Many critique poor Google results due to “SEO farms”.
“Google search is cluttered with junky results, and S.E.O. hackers have ruined the trick of adding “Reddit” to searches to find human-generated answers.” (source)
“But these changes may not address the fundamental issues behind Google’s decline. Google didn’t get to the position it’s in overnight. As the company ramped up its advertising options, useful search results were demoted down the page in favor of ads, experts say. The company has also been plagued by SEO spam, which crowds out relevant search results, often AI-generated or full of poor quality information, that are designed to rank high on Google’s algorithm.” (source)
The SEO in me wants to shout back. “We’re not all the same”. But even I shiver at some search results.
Despite SEO getting harder and Google using ever more machine learning to train its systems, it seems like Google is losing the battle against low-quality results and spammers. Algorithm updates have become hardcore, but even they can’t prevent bad results in every case.
My take: Google launched the Helpful Content Update to downgrade over-optimized websites. Too many “overoptimized” results also seem to be the reason Google removed FAQ and How-To rich snippets.
The contrast is getting thicker
I’m equally amazed by Google and critical. I owe the company a fulfilling career and a rich life. I tried to let Sergej Brin know when I met him in a restaurant in Menlo Park but picked a bad time. After over 13 years in the game, I also recognize Google is not infallible. People generally try to do the right things but have the wrong incentives, which greatly impacts our default operating mode.
It’s hard to judge the quality of something without a comparison. We need context. For over 20 years, Google’s search results were the best because there was no better alternative. Now, Chat GPT & Co bring a new level of attention to Google’s shortcomings, especially poor longtail search results.
Alphabet was smart enough to invest in Android and Pixel phones. Today, Android owns ~68% of smartphone traffic, and Alphabet can pay for the other 32%. Had they not invested in smartphones, paying to be the default would have cut too deeply into profit margins. Devices are gateways. Microsoft realized it too late, but Meta got the memo: own the device, and you own the default. (source)
Default as a tactic
Defaults can be powerful tools for conversion rate optimization.
1/ Bundling: what products or platforms can you establish an exclusive partnership with to become the default feature?
Netflix has become the default streaming channel on smart TVs. Paypal is the default payment provider on eBay. Stripe is the default payment provider for Shopify.
2/ Pricing: offer visitors 3 pricing tiers and highlight the one you want them to pick.
3/ Onboarding: set a product up as custom to the user as possible so they start with optimal default settings.
4/ Social proof: something can look like the default when lots of other people use it.
5/ Annual subscriptions: people are less likely to cancel subscriptions if they’re annual.
Careful with this tactic. Users can get really frustrated if they feel like you tricked them.
6/ Recommendations: highlight one next article to read that fits best with the topic of the current article.
7/ Research: ask your target audience how they solve the problems your product aims to rectify right now (what’s the default).