The End of AMP
I am hopeful that 2021 will be the beginning of the end for two of my least favorite things – the pandemic and Accelerated Mobile Pages (AMP).
For the past few months, I’ve been focusing on Google’s Page Experience update due to launch in May and what it means for publishers. The largest and most talked-about item in the update is Google announcing that sites with passing core web vitals will receive a ranking boost on mobile. However, there is another important item in the update – the end of special treatment for AMP pages.
AMP is an open-source framework that Google created with the stated goal of making mobile web pages fast. AMP pages must follow a specific set of rules and use a fixed set of features and functionality.
While the stated goal of AMP – making the web fast – sounds noble, AMP also came with the unfortunate requirement that publishers let traffic sources such as Google cache their content and serve it from their domain, such as google.com. In practice, this meant that when a user clicks on a search result from the BBC, they didn’t go to BBC’s website; they view a copy of the BBC’s content on google.com. AMP created all kinds of problems, from analytics to ad serving to logins. Perhaps most importantly, it created a temptation that even Google couldn’t resist – encouraging users to stay on Google instead of consuming more content on a publisher’s website.
Google provided a distinct advantage to sites using AMP – priority placement on the world’s largest traffic source – Google search. I’ve had the pleasure of working with more than twenty thousand publishers in the five years since AMP’s launch, and I don’t believe I’ve ever heard a single reason that a publisher uses AMP other than to obtain this priority placement. Let me package that up for you – Google, the most dominant search engine globally – used that dominant market position to encourage publishers to adopt technology so that Google could store and serve publisher’s content on Google’s domain. How is that legal? Well, I’m not a lawyer, but it possibly isn’t.
The good news is that, in May, this is all about to change. Part of the Google update is that all pages with high Page Experience scores are eligible to be in the featured top news carousel. This effectively means that publishers will no longer be forced to use AMP and can instead provide fast, rich experiences on their own domains.
Ok, so AMP loads a bunch of junk my site doesn’t use, but how does it do with core web vitals? I tested two websites, one which is normally ok-to-poor on web vitals and speed and the famous lafoo – which is perfect. 🙂
Both sites are measurably worse on their AMP versions. The amount of data transferred is higher, the script execution is worse, and the page takes longer to load. So much for the promise of speed and a better user experience.
The good news gets even better; non-AMP pages make considerably more revenue per pageview than AMP pages. Initially, I assumed this was due to the nature of how ads load on AMP, however, recent Antitrust lawsuits have proposed that hindering ad competition was a feature and that all non-amp ad tags, such as my company, Ezoic’s, were delayed by 1 second to make them less effective. It is also alleged that Google let their own exchange win, even when someone else bid more!
25 thoughts on “The End of AMP”
In advertising ecosystem, Google says AMP increases 17% of revenue. But according to user experience, it doesn’t do so. Besides, AMP destroy the site design thought it has some amazing features such as speed. Will AMP be more improved in future?
I work with tens of thousands of sites and I can say without question that AMP earns less revenue per user. Google cites numbers like that because they are only talking about Google revenue which has an advantage on AMP.
Nice information, actually I was wondering because I am looking for someone who can provide me AMP for my blog, but after reading this article I am looking forward to write some useful Post on my blog. Which can be more fruitful for helping me rank on search engines. Thanks for sharing.
So you’re saying I should forget about AMP?
Why aren’t these monopolists ever punished? AMP was the most blatant attempt at destroying the web I’ve ever seen. Break up Google, fine them hundreds of billions of dollars, and put their executives, product managers, and engineers who conspired to destroy the free and open internet in prison.
Good riddance. Controlling both search and content is a massive conflict of interest that gives a search engine the ability to prioritize content under its own control.
In the meantime, there exist some Firefox addons that redirect AMP pages to their non-AMP counterparts.
AMP is unfair – locking publishers by offering speed.
I’m building superblog.ai to make blogs load as fast as AMP if not faster than it. Let’s hope it will a minor step 🙂
Ah, but it isn’t always the CMS/Wordpress’ fault. This site is wordpress and it scores pretty well. It’s the bloated themes and 150 plugins that people add.
That’s true. WP-Optimisation is whole other thing.
Also, I just did google and gtmetrix audit for this page. It scores a 100/100 for performance on both, amazing!!
But looks like the complete page load takes 5.6 seconds. https://gtmetrix.com/reports/www.lafoo.com/Mz3uSRMe/
Superblog loads in less than 1 second 🙂
Despite all the “benefits”, I never used AMP, but I really was thinking about it because that was the “next big thing”. But just the thought that I must give up from UX and functionalities that I find that are the best for a client was enough to stay away from promised “rankings, speed and simplicity”. There are another ways to achieve that – through design, technical optimization and good hosting choice.
AMP isn’t going to end. I am sure it will continue for years.
We attempted implementing our Pigeon Paywall plugin with AMP in the early craze days. We quickly realized the fundamental flaws in the framework’s approach and maintainability. We proceeded to advise our clients to steer clear of AMP and focus on better responsive design, which in my opinion is one of the best “standards” approach to interface form factor flexibility since HTML became a thing.
Haha, just last week I got AMP working because of the promised “17% increased revenue”.
(It turned out I just had to tick the right box in my caching plugin.)
I have to say my speed test values improved significantly, but then the bad values had more to do with how Google delivers ads anyway, so it might be some behind the scenes effect from Google changing their algorithms.
I guess I now have a month to see if the revenue increases or if I will un-AMP my site again.
Thanks for the update!
I use AMP, and the results are rather in my favour. I position more urls overall in the SERP. I am surprised that this will stop in a while. I felt like I was staying ahead of the competition.
Nice information, actually I was wondering because I am looking for someone who can provide me AMP for my blog, but after reading this article I am looking forward to write some useful Post on my blog.
I get most of my programmatic ad income from AMP pages, although I still run desktop and mobile versions of my website.
Yes, that means I maintain three versions of the website which is not ideal… But for several reasons, I have no other choice.
What about other AMP components like AMP for Email, AMP Stories? They will continue and rule the world
AMP is worse than COVID as there’s no vaccine for it
You are awesome . Thanks for sharing this information
this was really in depth i was playing to use amp
thanks for the help
Awesome article nice information thanks