Why We Recommend Accelerated Mobile Pages

When Accelerated Mobile Pages (AMP) were first introduced in 2015, the primary audience was publishers, and more specifically, news sites. The introduction coincided with the announcement that search queries on mobile devices now outnumbered search queries on desktops, and the touted benefits of AMP included a better user experience, which translates into an increase in time spent on page, number of return visits and click-through rates (CTRs). Since then a number of studies on sites using AMP have confirmed these benefits, while an ongoing survey shows that users are more inclined to click on an AMP link than a regular link.

The benefit to users is a page that loads almost instantly (usually in under a second), without it being nominally different to the regular page. We already know that page load speed is extremely important, especially on mobile devices, but Accelerated Mobile Pages load even faster than perfectly optimised mobile pages. And this is one of the reasons users are more inclined to click on an AMP link and spend more time on the page. The benefits of AMP can be further explained as offering:

• Much faster load time on mobile devices for all AMPs – by reducing or eliminating unnecessary elements, visitors don’t need to wait before being able to browse the page content. But you don’t need to resort to having barebones AMPs with no navigation options and branding; these can – and should – still be included.

• Improved rankings and visibility on SERPs – AMP is not a ranking factor, but the faster load times, and potential for higher click-through rates and lower bounce rates could lead to better mobile rankings. Additionally, AMPs on SERPs have a small AMP icon displayed next to the URL, which serves as a signal to users that the page will load much faster. Some AMP results also appear near the top of SERPs in a carousel with images, making them even more visible.
• Own analytics – the are two tags available for inclusion on AMPs that allow you to still track visitors to your site, so you can differentiate between visitors to your normal website, and those to your AMPs, along with any behaviour that follows. This does require setting up though, but the benefits make it worthwhile.

Metromile performed an isolated test with a single landing page converted to using the AMP framework, and traffic from a search ad then split between a normal landing page and the one using the AMP framework. Results showed the CTR to the AMP page to be 13 percent higher than to the non-AMP mobile landing page, and the overall cost per conversion on mobile was 23 percent lower for AMP traffic than for non-AMP mobile traffic.

A Forrester study commissioned by Google in 2017 had e-commerce vendors reporting an average 10 percent increase in site traffic following the implementation of AMP, with a 20 percent increase in conversions, and a 60 percent increase in page views per user visit. The biggest increase in site traffic generally comes from mobile, as would be expected for technology that is – at its core – mobile-first.

The image above shows how different brands implemented AMP, along with their reported results, and notably almost all indicated a double digit drop in bounce rates, and increases in traffic, session time, and/or conversions. And though AMP is not a ranking factor, CTRs, time spent on page, and other metrics that are affected by the use of AMP do influence how pages rank.

Once a brand has decided to implement AMP, the biggest challenge is in deploying AMP properly. While there are WordPress plugins that simplify the process for sites using the WordPress platform, these are often quite basic and result in Accelerated Mobile Pages that aren’t too similar to the mobile version of the site.

The use of AMP delivers the best results when the pages are – functionally – almost identical to the mobile site and don’t impair the UX in any way. A custom deployment of AMP would see many design elements stripped from the pages, but not those that relate to the UX, such as some branding, navigation, and the appearance of the content itself. Depending on the number of pages that need an AMP version, a custom implementation should take two to four weeks from start of development to deployment, using either your in-house team of developers, or a professional digital agency.

AMP is not a static technology and has continued to evolve since it was launched, but we fully recommend a customised deployment over adopting pre-built plugins. It is no longer aimed primarily at publishers, and now even includes support for e-commerce sites and advertising. It’s also not a mobile-only solution, but rather a mobile-first solution, something that – following the roll-out of the Speed Update – is more important than ever before.