AMP's traffic record: five publishers, five different outcomes

Ask whether AMP helped or hurt publisher traffic and the honest answer is: both, depending on which real site you look at. Five documented, named cases below span a lead-generation host, a national newspaper group’s two different titles, a 159-publisher academic-style study, and two publishers who later ripped AMP out. None of them agree, and the disagreement is the finding.

Five sites, five results#

Independent.ie gained roughly 55% mobile search traffic when it adopted AMP. Daniel Smullen, Head of SEO at Independent News & Media, reported the increase directly, then supplied the caveat that matters most: he attributed the gain to “the first-mover advantage of additional traffic from the AMP Top Stories carousel, and also because at that time, we had a terribly slow mobile site.” Early adoption and a bad baseline, not AMP in the abstract, did most of the work.

Belfast Telegraph, a regional title in the same publisher group, launched AMP in 2019 and saw no measurable traffic change at all - but a 49% decline in mobile digital ad revenue. Smullen reported both cases in the same article. Same company, same technology, different title, and the opposite result on both axes that mattered: traffic held flat while ad revenue collapsed, largely from AMP’s ad-format and rendering restrictions.

Chartbeat and the Daily Beast ran the closest thing here to a controlled study, and found the mixed picture holds at scale. Chartbeat compared traffic before and after AMP implementation across 159 publishers and found an average lift of +22%, but only 34% of those publishers, about 1 in 3, showed a statistically significant increase at 95% confidence. The Daily Beast’s own A/B test inside the same research found no statistical evidence of higher Google-driven pageviews from its AMP pages, and lower revenue on the AMP format.

Kinsta, a managed WordPress hosting company with no news content, lost 59% of its mobile leads after two months running AMP, alongside a 16.67% drop in mobile newsletter signups and a 10.53% drop in mobile account creations. Kinsta disabled AMP and published the numbers as a warning to other conversion-focused sites: a template built to strip a slow news article down to load fast can just as easily strip out the trust signals and forms a B2B site needs to convert.

Search Engine Land and Future plc both later removed AMP and reported no traffic loss. Search Engine Land saw a 34% drop specifically in AMP traffic between July and August 2021, after Twitter stopped referring mobile users to AMP pages and Google’s Top Stories carousel began admitting non-AMP results; it turned AMP off that November. Future plc migrated 80 to 90% of its pages off AMP starting the prior autumn, and Stuart Forrest, its director of audience operations, said plainly: “there’s been no impact on volume, and all the impact on revenue that we had hoped for.”

Why these numbers cannot be averaged together#

The sites are not comparable populations. A news publisher racing to win Google’s AMP-only Top Stories carousel in 2016, a B2B hosting company measuring form fills in 2018, and a publisher group removing AMP once the carousel stopped requiring it in 2021 are three different businesses answering three different questions. Averaging Independent.ie’s +55% against Kinsta’s -59% produces a number that describes nothing real.

The baseline before AMP explains more variance than AMP itself. Independent.ie’s own account credits a “terribly slow mobile site” for most of its gain - AMP looks best precisely where the existing mobile experience was worst. Kinsta and Future plc, already investing in site speed for conversion or Core Web Vitals reasons, had less room for AMP’s stripped-down template to help and more to lose from its ad and layout restrictions.

The competitive incentive changed underneath every case. Through mid-2021, AMP was the only way into Google’s Top Stories carousel on mobile, which is exactly the incentive Independent.ie and the Daily Beast were responding to. Once Google’s Page Experience update removed that requirement in July 2021 and let any page meeting Core Web Vitals thresholds qualify, the traffic case for keeping AMP mostly disappeared - which is precisely when Search Engine Land and Future plc both left without losing traffic.

Why we care#

AMP’s own relevance has faded since Core Web Vitals became the metric that actually gates Top Stories. Google removed the AMP requirement from mobile Top Stories in July 2021 in favor of Core Web Vitals thresholds any page can meet, which is why the two most recent cases here are both removals with no traffic cost, not adoptions with a traffic gain. A technology-adoption case study from 2016 is not evidence about 2026’s ranking mechanics.

A single AMP traffic number, quoted without its site type and era, is not a citable claim. “AMP increased traffic by 55%” and “AMP cut traffic-driving leads by 59%” are both real, verified, named results. Neither one is “the effect of AMP” - each is the effect of AMP on one specific site, in one specific competitive and technical context, and reporting one without the other misrepresents both.

The pattern that actually replicates is about baseline speed, not AMP. Every case where AMP helped involved a site that was slow before adopting it; every case where removing AMP cost nothing involved a site whose non-AMP pages already met modern performance thresholds. That is a Core Web Vitals story wearing an AMP costume, and it is the more useful takeaway for technical SEO in 2026 than any single case study’s headline number.