When AMP was announced, I did not see it only as a Google project.
I saw it as a public sign that the mobile web performance problem had become too visible to ignore. For years, developers had talked about page weight, scripts, advertising tags, images and slow mobile connections. Those conversations often stayed inside development teams. AMP brought the issue into a wider business conversation because it framed speed as something that affected publishing, search, advertising and the way people read on their phones.
Whether AMP becomes the right answer for every publisher is a separate question. The more useful question for me is why something like AMP felt necessary in the first place.
Mobile Pages Had Become Too Heavy
A lot of mobile websites were not slow because mobile devices were weak.
They were slow because the pages were asking too much. Large images, multiple scripts, adverts, tracking code, custom fonts and social widgets could all arrive before the visitor had read a single line of content. Each individual addition might have had a reason, but the combined experience became heavy.
That is the pattern I recognise from normal website work. A page rarely becomes slow in one decision. It becomes slow because every small addition feels justified at the time. A marketing script is added, then a carousel, then a larger image, then a third-party embed. A few months later the mobile page is carrying far more than the original design suggested.
AMP Forced Restraint
The useful part of AMP is that it forces certain decisions.
It limits what can be added, expects content to be structured in a particular way and pushes pages towards a faster loading model. That restraint can feel uncomfortable because developers and publishers are used to having control. At the same time, control is often what made the page slow. When every department can add something to a page, the visitor eventually pays the cost.
I do not think restraint needs to come from AMP specifically. A well-built responsive site can also be fast. The issue is that many sites do not maintain that discipline over time. AMP’s stricter rules are interesting because they expose how much of performance work is not technical difficulty, but decision control.
The Concern Around Ownership
I would still be cautious about handing too much of the web experience to any one platform.
If a publisher’s fast version of a page depends heavily on a specific distribution system, there are questions to ask. Who controls the format? How does analytics work? What happens to the original site experience? Does the visitor understand where they are? These questions matter because performance is not the only value on the web.
That does not mean AMP should be dismissed. It means it should be understood as a trade-off. Faster content is valuable, especially on mobile. But businesses still need to think about control, maintainability and how the faster version fits into the rest of the site.
What This Means For Normal Websites
Even if I never use AMP on a project, the discussion around it is useful.
It gives me another reason to challenge heavy pages. Does this script need to load immediately? Does this image need to be that large? Does this advert or widget help enough to justify the delay? Those are not only developer questions. They are business questions because slow pages affect enquiries, reading, trust and revenue.
The mobile web does not need special treatment because it is secondary. It needs care because for many people it is the main way they use the web. Designing the desktop experience first and then hoping the mobile version behaves well is no longer enough.
What I Took From AMP
AMP makes the performance problem harder to avoid.
I am not convinced every site should immediately adopt it, and I would not treat it as a replacement for building properly. The stronger lesson is that mobile pages need less waste, clearer priorities and more discipline around what gets loaded.
If AMP pushes more businesses to ask why their pages are slow, that alone is useful. The answer may be AMP for some publishers. For others, it may simply be a better website with fewer unnecessary decisions getting in the way of the content.