131 SEO experts scored 100+ Google ranking factors in 2026. See the full breakdown of what moved up, what didn't, and what the survey says about AI-generated content.
On September 9, 2026, SEO consultants Cyrus and Dawn Shepard published the Google Ranking Factors Expert Survey 2026 through their Zyppy Signal newsletter. They asked 131 well-known search practitioners — collectively contributing 13,665 ranking-factor data points across roughly 43.6 hours of work — to score more than 100 potential Google ranking signals on a seven-point scale, from strongly negative to strongly positive.
The headline finding is almost anticlimactic: the same three factors that have topped every credible version of this survey for a decade — relevance, backlinks, and content quality — still top the list in the age of AI Overviews, AI Mode, and agentic search. But a few details inside the data are genuinely new, and worth knowing before you plan your next quarter of SEO work.
131 Experts, 13,665 Data Points, One Familiar Answer
Both the scale of this survey and its central finding are worth sitting with: a huge, current dataset of expert opinion pointing back to fundamentals that have held for a decade. Let’s break down what’s inside it.
What the Survey Actually Measured
- 131 respondents, including well-known names across the SEO industry
- More than 100 individual ranking factors tested, grouped into 9 major categories: Relevance, Backlinks & Authority, Content Quality, Brand & Reputation, User Signals, Technical SEO, Internal Linking/Site Architecture, UX Signals, and Google AI Ranking Factors
- Each factor scored on a 7-point scale (+3 strongly positive to -3 strongly negative)
- A separate open-ended question asked each respondent to name their top 3 most important ranking factors, with no fixed list to choose from
- 13,665 total data points, described by the authors as one of the largest and most current external collections of expert opinion on how Google Search works
The Top 3 Haven’t Changed
When asked to name their top 3 most important ranking factors, respondents converged hard on three answers:
| Rank | Factor | Named in Top 3 By | What It Covers |
|---|---|---|---|
| 1 | Relevance | 57.1% | Whether the page truly answers the searcher’s actual intent |
| 2 | Backlinks & Authority | 54.8% | Links from trusted, topically relevant domains with real visitors |
| 3 | Content Quality | 47.6% | Depth, originality, and first-party data or research |
Search intent match specifically received the highest individual rating of any single factor tested. Several respondents pointed to “information gain” — offering something genuinely new when multiple pages already satisfy the same intent — as what separates a page that ranks from one that doesn’t.
On backlinks, the survey pushes back directly against a popular narrative. Google’s Gary Illyes has said publicly that links aren’t among Google’s top three ranking signals — yet 54.8% of these 131 practitioners still named backlinks in their own top 3. As Cyrus Shepard put it in the original report, “the rumor of the death of backlinks has been greatly exaggerated.” Spammy, low-quality backlinks, meanwhile, rated as one of the strongest negative signals in the entire survey.
The AI Content Verdict
This year’s survey added a new question specifically about AI-generated content, and the answer was more nuanced than a simple “AI content is bad”:
- AI-generated content that received substantial human review and added genuinely unique value was rated favorably
- AI-generated content published at scale with little added value — the classic “content farm” pattern — was rated as a clear negative signal
In other words, the tool used to draft a page doesn’t appear to be the issue in respondents’ eyes. What matters is whether a human still added judgment, verification, and real value before it was published.
The Surprise Loser: Meta Descriptions
Near the bottom of the rankings, meta descriptions stood out as one of the clearest “doesn’t matter much” verdicts in the whole survey, reported in detail by Search Engine Land’s coverage of the results. Most respondents said meta descriptions carry little to no direct ranking effect — a useful reminder to spend that time elsewhere. They still matter for click-through rate in the search snippet itself, just not for where a page ranks.
Where Brand and User Signals Fit In
- Branded search volume and online reputation received some of the highest ratings within the Brand & Reputation category
- User satisfaction and task completion rated highest among User Signals — the idea that Google can tell, in aggregate, whether people got what they came for
- Quickly returning to the search results after clicking a page rated as a negative signal
- Technical SEO factors — crawling, indexing, overall site health — were rated consistently positive but described by respondents as foundational requirements rather than a ranking lever on their own
Full Ranking-Factor Cheat Sheet
Here’s how all 9 survey categories broke down, based on the report’s framing:
| Category | Standout Positive Signal | Standout Negative Signal |
|---|---|---|
| Relevance | Search intent match | Keyword-matching without answering intent |
| Backlinks & Authority | Links from trusted, relevant, real-traffic sites | Spammy or low-quality backlinks |
| Content Quality | Original research, first-party data | Scaled AI content with no added value |
| Brand & Reputation | Branded search volume, online reputation | Weak or negative brand sentiment |
| User Signals | User satisfaction, task completion | Quickly returning to search results |
| Technical SEO | Clean crawling, indexing, site health | Crawl or index errors |
| Internal Linking / Architecture | Clear, logical internal linking | Orphaned or buried pages |
| UX Signals | Positive on-page experience | Poor UX or high friction |
| Google AI Ranking Factors | Content structured for citation and extraction | Content unreadable or inaccessible to AI systems |
What This Means for Your Site
- Audit your top 10 pages and ask one honest question: does this page fully answer what the searcher came for?
- Don’t chase new AI-era tactics at the expense of backlinks — earning links from real, relevant sites remains one of the two most-cited factors.
- If you use AI to help draft content, budget real time for human review, fact-checking, and adding something the AI didn’t already know.
- Stop optimizing meta descriptions for rankings — write them for click-through instead.
- Track your brand’s reputation and branded search volume as an SEO metric, not just a marketing one.
FAQs
Is this an official statement from Google about ranking factors?
No. This is an independent survey of practitioner opinion, not a specification from Google. The authors describe it as the profession’s collective read on what moves rankings, based on real-world testing and experience.
Are backlinks really still important in 2026?
According to this survey, yes — 54.8% of respondents named backlinks in their top 3 most important ranking factors, second only to relevance.
Does using AI to write content hurt my rankings?
Not inherently. The survey found AI-generated content reviewed and improved by a human was rated favorably; it’s unedited, scaled AI content with little added value that was rated negatively.
Should I still write meta descriptions?
Yes, for click-through rate in search results — just don’t expect them to move your rankings directly, since most respondents said they have little to no ranking impact.
How is this survey different from Google’s own ranking factor documentation?
Google doesn’t publish a ranked list of its own factors. This survey instead aggregates the collective experience of 131 working SEO practitioners, based on what they’ve observed moving rankings in real client and business work.
Closing
None of this points to a revolution. Relevance, links, quality, and trust have topped every serious version of this exercise going back through Panda, Penguin, mobile-first, Core Web Vitals, and now AI answers. What’s changed is what surrounds them: AI content needs a human hand to be trusted, and reputation now shows up as a ranking factor practitioners can feel, even if Google won’t publish the formula.
Written by mootkur9
17 years in digital marketing. Founder of Sadhyam. Writing about AI search, AEO/GEO, and the agentic web. More about me →
