A little less than two months ago, we decided to discontinue the publication of our content using Google’s Accelerated Mobile Pages (AMP). The main reasons were to simplify our reporting and to stop hosting our content on third-party servers.
Since making this change, we’ve observed minimal disruption to our traffic and gained the advantage of having a clearer picture of our audience analytics.
Traffic Analysis: It’s challenging to draw definitive conclusions about traffic changes since we turned off AMP. As a media website focused on journalism, our traffic spikes and dips are closely tied to major news events. We typically see traffic surges during significant news like core updates or major changes in Google Ads, while traffic tends to decline during the holiday season. Thus, year-over-year benchmarking is generally more effective.
We did not observe any year-over-year traffic declines attributable to the turn-off of AMP, except for a few pages that used to spike for organic traffic. For instance, an article about Google SERP Easter Eggs, which often spikes around Easter, previously directed mobile traffic to its AMP version. After turning off AMP, we didn’t see this traffic transition back to our native page. However, this page wasn’t driving quality traffic, so the loss wasn’t concerning.
Safeguarding Measures: When we decided to shut off AMP, we implemented a few strategies to safeguard against potential traffic declines. We increased our publishing volume and adjusted our newsletter strategy to optimize for click-through rates. This change also responded to Apple’s iOS 15 privacy update, which has made open rates a less reliable metric.
The Big Win: One of the main motivations for turning off AMP was to gain a better understanding of our metrics. Despite several attempts at implementing AMP stitching in Google Analytics, we couldn’t accurately track how our audience moved between AMP pages and our native ones. Users were likely being double-counted as unique in both dashboards. Since turning off AMP, we’ve seen a 30% increase in sessions by return visitors, providing a clearer picture of our most valuable audience segment.
Why We Care: We acknowledged the risk going into this experiment but haven’t found any reason to reconsider our decision. One major concern was the Page Experience Update, as AMP pages were exceptionally fast. There was worry that our native pages, which don’t benchmark as high, would suffer. While we noticed a decline in the percentage of pages with “good” Core Web Vitals scores in Google Search Console after turning off AMP, we don’t believe it negatively impacted traffic or rankings. This aligns with the experience of many SEOs who struggle to directly correlate their performance with the Page Experience Update.
We’re committed to this change and not looking back. If you have a story about turning off AMP, we’d love to hear it.
