A survey of 131 search practitioners released on September 9, 2026, ranked relevance, backlinks and content quality as the three signals with the greatest positive effect on Google organic rankings, with search intent match rated the single strongest factor out of 103 tested and the meta description rated so weak that most respondents assigned it no impact at all.

In Short

A group of 131 search specialists was asked to score more than a hundred things that might help or hurt a web page's spot in Google results. They agreed that matching what a searcher actually wants matters most, that links from trusted sites still count for a lot, and that the little summary text under a search result does almost nothing. If you run a website, it means the old checklist of technical tweaks matters less than giving people the answer they came for and getting other real sites to talk about you.

What the survey measured

The study, published under the title Google Ranking Factors Expert Survey 2026 by Cyrus Shepard and Dawn Shepard through the Zyppy Signal publication, collected responses from 131 practitioners. According to the authors, contributors supplied 13,665 ranking-factor data points, a figure they describe as one of the largest and most current external collections of expert opinion on how Google Search operates.

Each respondent scored more than 100 potential factors. The instrument asked how much of a positive or negative effect each factor has on organic rankings, and answers were recorded on a seven-point scale running from +3, strongly positive, to -3, strongly negative. A separate open-ended question asked each participant to name the three most important ranking factors, with no fixed answer set. According to the survey, contributors spent a combined 2,620 minutes, or 43.6 hours, completing it, and the authors reached out to more than 200 practitioners to assemble the panel. The final questionnaire narrowed a larger pool of candidate signals down to 103.

The methodology matters for how the numbers should be read. The percentages attached to each factor below represent the share of respondents who named that factor in their open-ended top three, not a measured weight inside Google's systems. The scale scores, by contrast, capture average sentiment across the full panel for each individual signal.

The top ten

According to the survey, the ten factors most frequently named in the top three were the following, with the percentage of respondents selecting each shown alongside.

Relevance led at 57.1 percent. The authors define content as relevant when it matches and satisfies the search intent of the user, semantically matching what is sought rather than merely sharing keywords.

Backlinks followed at 54.8 percent. The report states that links from trusted, relevant sites were rated among the most important factors, and that many respondents stressed links from pages carrying real visitors over raw link counts.

Content quality came third at 47.6 percent, covering originality, accuracy and freshness. Nearly half the panel placed some element of content quality in its top three.

Authority and trust reached 36.5 percent, defined as how much Google trusts the website, the creator or the brand. Behavior and click signals scored 29.4 percent, described as how users select results and interact with pages. Brand signals registered 27.0 percent, capturing the value of being an entity that people search for by name. User satisfaction, framed as task completion, drew 19.8 percent. Technical SEO health, the crawling and indexing foundation, reached 17.5 percent. Topical authority scored 14.3 percent, and internal links closed the list at 11.1 percent.

Search intent match, and the meta description at the bottom

On the individual-signal scale, the clearest verdict concerned a factor the authors call search intent match, defined as how well a page satisfies the type of result or task the searcher appears to want. According to the survey, it was the highest-rated of all 103 signals. The report frames the distinction as going beyond keyword and semantic matching, arguing that a page must deliver the specific outcome a searcher is after.

One contributor put the point in blunt operational terms. According to a comment published in the report, "You can have technically perfect content with pristine headings and a gorgeous URL structure, but if the person searching wanted one thing and you answered another, none of that saves you." A second respondent proposed information gain as the tiebreaker once intent is met, observing that when ten pages all cover the same ground, what decides ranking is whether a page adds something the others lack.

At the opposite end sat the meta description. The authors report it was rated so ineffective that most respondents do not believe it carries any ranking impact. The finding is consistent with Google's own long-stated position that the meta description is not a ranking factor, though it can influence the snippet a searcher sees on the results page, the SERP surface examined in PPC Land's glossary entry.

A more surprising note concerned exact-match domains. One contributor, expressing surprise, wrote that an exact-match domain was ranking for a competitive term "with a fraction of the effort" required for their main site targeting the same query.

The survey's second-ranked factor pushes against a recurring claim that links no longer matter. According to the report, links from highly trusted domains and from topically relevant pages were two of the highest-rated signals in the entire study. Spam links, meanwhile, were rated among the largest negative factors, a scoring the authors note despite Google's public position that it ignores most such links.

The emphasis fell repeatedly on links that carry human traffic. One comment in the report states, "A hundred links from real audiences beat a thousand links nobody ever clicks." Another connects link value to visibility inside AI systems, describing a client that was mentioned and linked in a Reuters article and, according to the contributor, saw its visibility in ChatGPT rise fourfold over the following week. A third respondent addressed anchor text directly, arguing that exact-match anchors drive gains up to a point and suppress performance past it, with the deciding variable being the share of the anchor profile they represent rather than the raw count.

Content quality and the cost of AI abundance

Content quality is where the 2026 results diverge most from prior editions, according to the authors. Original research and first-party data topped the quality sub-list, which the report reads as a response to the flood of machine-generated text. Respondents viewed AI-generated content negatively, but the report specifies the objection applies when such content is scaled with little added value rather than to AI assistance as such.

One contributor framed the economics plainly. According to the comment, "AI has made content cheap. Anyone can spin up a thousand articles overnight. And Google knows it. That's why quality is becoming the filter. When the volume of content explodes, the cost of publishing drops to zero... but the cost of trust goes up." Another respondent reported positive results from high-quality AI generation at scale, on a site with substantial unique content and user-generated content, where AI output passed through intensive human editorial review. A third warned of a narrow band between two failure modes, writing that "Too much consensus, and you're invisible for not being relevant enough. Too much information gain, and you're shadowbanned for being too contrarian."

Brand, and the popularity contest

Brand signals surfaced as a primary route through which Google decides whom to trust, according to the survey. Online reputation and branded search volume, meaning how many people search for a business by name, were rated among the most important factors in this group. Google ad spend registered a barely positive influence, which the authors attribute to its potential to boost visibility rather than any direct ranking effect. One comment qualifies the point: "No direct effect of ad spend - it can affect other signals, though, so it can result in benefit."

The strongest language in the report came from contributors on brand. One respondent called ranking "a popularity contest" and argued that content pitched too far toward expert level can actually harm visibility, describing a Goldilocks zone of broad appeal that attracts shares, links and mentions. Another stated that brand signals are the strongest factor in their view, contending that a site in poor technical shape with unoptimized content can still perform as long as people search for and discuss the brand. A third traced the idea to a Google patent, writing that "Branded Search is the new link, since 2012, if a Google patent is to be believed."

User signals and the antitrust record

User-based signals took what the authors describe as a large leap in this edition, which they tie to evidence that emerged from Google's antitrust trial and the 2024 leak of internal Content Warehouse documentation. That leak, which PPC Land reported on at the time, outlined more than 2,500 modules and 14,000 attributes describing how Google Search processes and stores information. Google confirmed the documents' authenticity shortly after they surfaced.

Within the user-signal group, satisfaction and task completion rated highest, while return to SERP, where a searcher goes back to Google's results after visiting a page, rated negatively. One contributor centered the Navboost system, the click-based mechanism described in Department of Justice testimony and named in the 2024 leak, writing that "Navboost is still a monster" and that sites should build a satisfaction metric by page type because a ranking ceiling exists when click signals trail SERP competitors. Another distinguished bounce rate, which the contributor calls a poor metric, from pogo-sticking to another result, which they read as a signal Google clearly takes. A third raised a forward-looking question that the survey does not resolve: when most browsing becomes agentic, how will these signals be measured at all?

Technical SEO as table stakes

Most respondents described technical SEO as table stakes, according to the report: bad technical health can hurt, but good technical health does not lift otherwise mediocre content. One comment captures the framing, calling technical SEO "a commodity at this point" that "won't necessarily win the race, but it can absolutely lose it." A contributor illustrated the downside with a case in which canonical tags pointed at unrelated pages; once corrected, the site saw an almost immediate rise in impressions and clicks the respondent estimated at roughly 60 percent.

Internal links drew warmer assessments than their tenth-place finish might suggest. One respondent argued that "a good internal link is as valuable as a very good external backlink," and another described internal linking as the one ranking lever a site controls end to end, deciding which pages accumulate equity and which Google reads as important.

On user experience, respondents flagged mobile usability as mattering most, while site speed and Core Web Vitals emerged as two of the most contested signals in the whole survey. Views ranged from a contributor calling speed and Core Web Vitals "the most over-rated things in SEO" to another insisting that for a large e-commerce site, speed genuinely matters.

Why this matters for the marketing community

The survey lands at a moment when the economics of organic search visibility are under measurable strain, a strain PPC Land has documented across the year. Ahrefs analysis of 300,000 keywords, published in April 2025, found the first organic link loses an average of 34.5 percent of clicks when an AI Overview appears. A randomised field experiment reported in July 2026 produced causal evidence that AI Overviews cut outbound organic clicks by 39.8 percent while lifting zero-click searches by 34.5 percent. Seer Interactive measurement across 42 organisations showed organic click-through rates for AI Overview queries falling 61 percent between June 2024 and September 2025.

Against that backdrop, the panel's emphasis on brand, real-audience links and task completion reads less as SEO orthodoxy than as a description of the signals that survive when the click itself is disappearing. Several of the survey's own comments make the connection explicit, tying trusted links to citation inside ChatGPT and framing brand recognition as the durable asset when raw traffic erodes. The verdict on GEO, or generative engine optimization, remains contested elsewhere: a survey of 45 academic studies covered by PPC Land concluded that the strongest available evidence shows content already retrieved by a generative system can influence how it is cited, while far weaker evidence exists that any technique reliably improves whether a page is retrieved.

The panel's framing of technical SEO as a floor rather than a lever also aligns with the direction of Google's own public guidance. Comment spam links have no effect on rankings, per John Mueller's January 2026 confirmation, and court testimony from Google's own expert has indicated that most of the company's quality signal is derived from the webpage itself rather than from external linking. The survey's contributors, working from outside Google, arrived at a broadly compatible hierarchy: satisfy the intent, earn trust from real sources and sound infrastructure, and treat the rest as maintenance.

The authors note that a companion study on ranking inside Google's generative AI responses is planned, addressing how content gets cited by AI Overviews and AI Mode. That release was not available as of the September 9 publication.

Timeline

Summary

Who: Cyrus Shepard and Dawn Shepard, publishing through the Zyppy Signal newsletter, surveying 131 search practitioners including many well-known names in the field.

What: The Google Ranking Factors Expert Survey 2026, which collected 13,665 ranking-factor data points across 103 tested signals and ranked relevance, backlinks and content quality as the three most positive factors, with search intent match rated the single strongest signal and the meta description rated as carrying no meaningful ranking impact.

When: Published on September 9, 2026, with contributors spending a combined 43.6 hours completing the questionnaire.

Where: An online survey of practitioners based across the search industry, released through the Zyppy Signal publication.

Why: For marketers, publishers and SEO teams, the results describe which ranking signals still hold value as AI Overviews and zero-click behavior erode organic traffic, pointing toward brand recognition, links from real audiences and task completion over technical tweaks and metadata that the panel rated as low-impact.