Why I do not Recommend Google as Search Engine - A Little Experiment
After having created the first articles on this blog together with my personal profile page some weeks ago
I recently did a little experiment: I searched for the article about the RESTEasy Reactive migration in different search engines
because I was curious about the results. Before doing so I waited some time to give the crawlers of the search engines a chance to find and analyze the website. They were pretty fast and at the time of writing this article I see different bots visiting my page daily. Also, although I am not a SEO expert, I tried to make their lives easier by adding the usual things like a robots.txt, a sitemap.xml, metadata on every page and so on.
For the experiment I decided to choose the following search engines:
- MetaGer: The search engine I use daily. It is run by the non-profit German association SUMA-EV which "advocates free access to knowledge on the internet".
- DuckDuckGo: Calls itself "the Internet privacy company" and is the default search engine e.g. for the Tor Browser Bundle.
- Ecosia: Well-known in Germany because they "plant trees with every search". It's based on Bing which I tried to use directly at first, but in the typical Microsoft manner, you can't even submit the simple search form on their start page without activating JavaScript. No comment.
- Startpage: The search engine I used earlier because it's based on Google and I found their search results the best at that time, but didn't want to feed their tracking machine.
- Google: For obvious reasons. It's still the most used search engine in the world with a share of 84.69% in January 2023.
As search term I used "Notes on the Migration from Quarkus RESTEasy Classic to RESTEasy Reactive" (without quotes), the exact title of the article. These are the results:
| Search Engine | Index is based on | Where the article appeared on 10.03.23 |
|---|---|---|
| MetaGer | various | first page, second result |
| DuckDuckGo | various | first page, second result |
| Ecosia | Bing | first page, second result |
| Startpage | last page, last result (domain only) | |
| not at all |
MetaGer, DuckDuckGo and Ecosia all showed the article as second result on the first page. The first result was the official migration guide by Quarkus, so the order looked reasonable. Startpage showed the domain (not the article) at the very end which I found kind of strange because the article is on a pretty specific topic and many of the results further up were fitting worse from my point of view. But the biggest surprise was searching with Google resulting in nothing. The article didn't show up in the results at all although the Googlebot is frequently visiting the website. Additionally, it's not clear for me why Startpage returns at least the domain while Google does not, even though Startpage is based on the Google index. Obviously there are deviations.
The reason for Google not showing the article lies in the dark since it is basically a black-box. Maybe Google's Search Console could shed some light on that but since it requires a login (of course), I won't use it. I deleted my account years ago and have no plans of creating a new one whatsoever. The point is: I checked the page e.g. with Lighthouse which is integrated into the developer tools of Google Chrome/Chromium, and it gives me the maximum score of 100 in the 'Best Practices' and 'SEO' categories. This together with the fact that all non-Google based search engines do show the article appropriately gives me confidence that there is no major issue in terms of SEO. What I can imagine though is that they rank pages down which don't actively use their Search Console (or their bugging device Google Analytics).
Next step was verifying that the page exists in their index by doing a search with site:www.pachamamita.de/posts/quarkus/migration-resteasy-classic-to-reactive/ and indeed it showed up then. However, it doesn't help too much when a page is present in their index, but not shown to people searching for it.
Fun fact: What did show up pretty prominently when using Startpage and Google were several results on groups.google.com and YouTube. The reason is obvious: Google is not neutral and wants to push its own platforms and sites. Also, I can imagine that tracking people - their main business - is easier and more fine-grained when staying on Google-owned pages. Although especially Google Analytics is spreading out literally like cancer in the World Wide Web in recent years, so cross-website tracking should not be a big problem for them anyway. It seems there are no official numbers available on how many pages are using this bugging device, although Google knows them of course. From what I see being blocked by my Ad-Blocker during daily surfing, I assume there are millions and millions of pages who use Google Analytics and feed this huge tracking machine with all kinds of sensitive data, most of the time without visitors even knowing that they are being tracked and monitored throughout the Internet.
Conclusion
In summary, I can say that my decision to stop using Google as search engine some time ago was a good one considering:
- The search algorithm is a black box, no one knows how it works in detail, it's closed source. They give you some general hints what to try when a page is missing in the search results and that's it. Many of these hints link to their Search Console which cannot be used without an account.
- Google doesn't only own the search engine, but also other big platforms like e.g. YouTube, so it's not a surprise that they place many links to these in a prominent manner even though other content might be far more relevant. From my experience also other Big Tech platforms tend to be returned at the top of the results.
- Search terms are sensitive data. People enter everything that concerns them into search engines. It is not a secret that Google's business model is mainly based on targeted advertising (revenue in 2022 from ads alone more than 224 billion U.S. dollars) and that they aggregate all data they can get into large user profiles in order to show the people potentially better fitting ads.
The surveillance capitalism has really messed up the World Wide Web completely in recent years. Surfing the web without an Ad-Blocker is a pain because of all the blinking and attention seeking ads everywhere (not to mention the increasing danger of malvertising campaigns), there are trackers that monitor and record all you're doing on the majority of websites and so on. And Google is one of the biggest players in this misanthropic business. That's why I decided for myself to stay away from Google&Co. as far as possible and use alternatives whenever available, not only in terms of search engines, but also e.g. in using Firefox as browser instead of Chrome, in using OpenStreetMap instead of Google Maps, in using a smartphone with a Google-free Android and so on.
Of course the above experiment is not representative in any way. However, it shows pretty well some of the problems that arise when you trust a black-box search engine whose operator is a major player in surveillance capitalism and essentially interested in your data to serve you profane ads. Using an alternative search engine like metager.de (international version available under metager.org) has several advantages:
- In my experience, it gives at least as good results as Google.
- It is run by a German non-profit organization not trying to extract as much money as possible from people.
- It's open source, everyone can check the source code.
- It's based on Laravel, which is a great framework (just to mention that)
- They are not trying to push other own services.
- A Tor-Hidden-Service is available under http://metagerv65pwclop2rsfzg4jwowpavpwd6grhhlvdgsswvo6ii4akgyd.onion (Tor Browser Bundle required).
- They do use advertising to generate revenue to provide the service, but these ads are not personalized, and they offer an ad-free quota of searches for members of the association.
All in all, I see absolutely no reason to continue using Google as a search engine.
Final note: I have no connection with MetaGer apart from using it, and I am not receiving any money.