Are AMP and PWA really useless for SEO?

What does "direct signal" versus "indirect effect" really mean?

When Google talks about a direct signal, it refers to a criterion explicitly included in the ranking algorithm. The page title, quality backlinks, or the presence of keywords in the content are direct signals: the algorithm reads them, evaluates them, and adjusts the ranking accordingly.

An indirect effect is different. AMP and PWA are not scrutinized by the algorithm as ranking factors. But if they speed up your site or improve user engagement, these improvements can influence signals measured by Google (bounce rate, session time, Core Web Vitals).

Why does Google maintain this blurry distinction?

This position allows them to promote open-source technologies (especially AMP) without committing to a measurable competitive advantage. If AMP were a direct signal, Google would have to prove that it brings a ranking gain. By remaining in the indirect, they evade metric responsibility.

Another reason: to avoid charges of abusing their dominant position. Making AMP a ranking criterion would have triggered antitrust investigations. By presenting it as an optional optimization lever, Google stays in a legally comfortable zone.

Do PWAs and AMP really improve perceived speed?

For PWAs, yes, but only after the first visit. The service worker caches essential resources, drastically reducing loading times on return visits. On the first visit, the advantage is null or even negative (the browser has to register the worker).

For AMP, the acceleration mainly comes from the Google cache and the constraints imposed on the code. However, a well-optimized site without AMP can achieve equivalent performance. The real difference? Pre-rendering in mobile SERPs, which gives the impression of instant loading.

• Direct signal: explicitly evaluated criterion by the algorithm (backlinks, title tags, Core Web Vitals)
• Indirect effect: improvement of an aspect (speed, UX) that can influence other measured signals
• AMP only provides a significant speed gain through Google cache and SERP pre-rendering
• PWA mainly improves repeat visits thanks to the service worker and local caching
• None of these technologies compensate for failing SEO fundamentals (architecture, content, links)

Is this statement consistent with on-the-ground observations?

Partially. Sites that migrated to AMP between 2016 and 2019 often observed gains in mobile traffic, but these gains were linked to two factors: access to the Top Stories carousel (which required AMP at the time) and a drastic improvement in their LCP. Today, Top Stories accepts non-AMP pages that comply with Core Web Vitals.

Regarding PWAs, feedback is more nuanced. E-commerce sites with a lot of recurring traffic (multiple sessions per week) report improved engagement, but no jump in rankings. News sites or those with sporadic content see little impact, as most traffic comes from the first SEO visit. [To be verified]: Google has never published data quantifying the actual impact of PWAs on user metrics at a large scale.

What nuances should be added to this official position?

Google fails to specify that certain SERP formats have long required AMP as a prerequisite. This wasn’t a ranking signal, but a condition for accessing a major visibility lever. The distinction becomes purely semantic when the absence of AMP excludes you from a carousel that drives 40% of mobile CTR.

Another blind spot: Core Web Vitals have become direct signals, and both AMP and PWA can help meet them. But Google doesn’t say that these technologies are often a costly detour. A well-architected site with a good CDN, intelligent lazy loading, and optimized images gets the same results without the technical debt of AMP.

In what cases does this rule not fully apply?

If your site targets low-connectivity markets (sub-Saharan Africa, Southeast Asia), PWAs offer a real competitive advantage. Users on unstable 2G/3G networks see a massive difference between a traditional site and a well-designed PWA. This UX advantage translates into higher engagement metrics, which can influence local ranking.

Another exception: complex web applications (SaaS, dashboards) where navigation is intensive. A PWA improves the experience so much that it alters user behavior (more pages viewed, longer sessions). These behavioral signals can have a measurable indirect impact. But for a blog or a static showcase site? The effect is marginal.

What should you concretely do with this information?

The first strategic decision: do not invest in AMP or PWA solely for SEO. If you are considering these technologies, valid motivations include user experience on mobile, reducing data consumption, or accessing offline features. SEO will be, at best, a collateral benefit.

Focus your resources on Core Web Vitals, which have been direct signals since May 2021. A fast site without AMP will always outperform a slow site with AMP. Prioritize server optimization, image compression (WebP, AVIF), native lazy loading, and reducing blocking JavaScript.

What mistakes should be avoided in interpreting this statement?

A classic mistake: thinking that AMP and PWA are obsolete or useless. They are not direct SEO levers, but for certain uses (high-traffic news sites, web applications with intensive navigation), they remain relevant. The trap is adopting them blindly, hoping for a ranking miracle.

Another mistake: neglecting the ecosystem. If your direct competitors are using AMP and achieving a higher CTR through pre-rendering, you are losing traffic even without a loss in ranking. The SEO equation is not limited to positions: visibility and clicks matter just as much.

How can you verify that your strategy aligns with this reality?

Audit your real performance metrics: LCP, FID, CLS in Search Console. If your scores are in the green, AMP won’t provide any benefit. If you are in the red, fix the fundamentals before considering a heavy technological migration.

Test user impact: compare your bounce rates, session duration, and pages per visit on mobile versus desktop. If the gap is significant and slowness is the issue, a PWA may be justified. But measure before and after to isolate the real effect, not the marketing promises.

• Audit your Core Web Vitals in Search Console and Google PageSpeed Insights
• Compare your mobile page performances with those of your direct competitors
• Measure the real impact on user engagement (session time, bounce rate) before any AMP/PWA migration
• Prioritize classic performance optimizations (CDN, compression, lazy loading) before adopting a new stack
• Document KPIs before/after to isolate the effect of any technological change
• Test AMP or PWA on a subset of pages before a global deployment