93 months of ranking data behind every recommendation.
AwesomeSEO reads the TopSEOFactors monthly top-200 correlations from Jan 2019 to Sep 2026, maps each factor to Koray Tuğberk GÜBÜR's topical-authority pillars, and labels every practice with how strong its evidence is.
- months of data
- 93
- Jan 2019 – Sep 2026
- distinct factors
- 619
- 200 ranked each month
- factor-months
- 18 600
- one row per factor per month
- practices, each tiered
- 106
- 7 evidence tiers
Correlation, read carefully.
What we import every month
- Window
- Jan 2019 – Sep 2026 · 93 months
- Factors
- 619 distinct · 200 ranked per month
- Rows
- 18 600 factor-months
- Statistic
- Rank correlation between a factor’s value and Google position
- Imported
- 2026-10-20
Position 1 is the top, so a factor that rises as pages climb correlates negatively. Number of Referring Domains has r = −0.28: pages with more of it tend to sit higher. Closer to zero means a weaker link, not a negative effect.
3 of 5 AI answers is 60%, but the 95% interval runs from 23% to 88%. At 105 of 175 it narrows to 53%–67%. Every rate shows its interval and n.
Going from 3 of 5 to 4 of 5 looks like +20 points. The test gives p = 0.49, so the product marks it n.s. instead of calling it a win.
Test 40 factors at p < 0.05 and about 2 look significant by chance alone. In a seeded simulation of 40 factors with no real effect, 2 passed; after the false-discovery-rate correction, 0 did.
To detect r = 0.20 with 80% power you need 194 pages. For r = −0.12, the weakest factor in the top 200, you need 543. One SERP of 20 results proves nothing on its own.
- Correlation is negative because rank 1 is the top result: more negative means a higher factor value goes with a better position.
- Correlation is not causation. Several factors are proxies (page weight, include counts) that rise with richer pages.
- The data is global and mostly US/English SERPs. Treat page-1 averages as benchmarks for Nordic markets, not quotas.
- The latest month may be month-to-date.
- Pillar mapping is keyword-rule judgement (Koray framework), not measured data.
The 200 strongest factors in September 2026.
Showing 25 of 200 factors
TopSEOFactors · top 200Sep 2026r is negative because position 1 is the top. Proxy marks factors that rise with page weight: keep the page fast, do not copy the number.
106 practices, each labelled with how sure we are.
7 tiers · 106 practices
Confirmed: 18 practices
Google states it in documentation or official guidance.
Do: Reference each page’s main entity by a stable @id in JSON-LD and link it to Wikidata, Wikipedia and official profiles with sameAs.
Evidence: Google’s structured-data guidelines support sameAs and consistent identifiers for organisations and people.
How AwesomeSEO measures it: Entity graph with @id coverage and sameAs completeness.
Topical mapDo: Label how AI helped, have a named person review every page, and check AI drafts for accuracy and originality.
Evidence: Google accepts AI-assisted content that helps people; scaled low-value content is a spam policy violation.
How AwesomeSEO measures it: Approval log on every agent draft.
AgentsDo: LCP ≤ 2.5 s, INP ≤ 200 ms, CLS ≤ 0.1, measured on real users.
Evidence: Part of Google’s page-experience guidance. INP replaced FID in March 2024.
Site healthDo: Server-render or pre-render pages that should rank; check that text, links and schema are in the HTML response.
Evidence: Google renders JavaScript, but later and at a cost; many AI crawlers do not render it at all.
Site healthDo: Every structured-data claim must be visible on the page.
Evidence: Google’s structured-data guidelines; mismatches can lose rich results or trigger a manual action.
How AwesomeSEO measures it: Schema parity check in the crawler: share of JSON-LD values found in visible text.
Site healthDo: One indexable URL per piece of content, 301 for moves, 404/410 for removals, no chains.
Evidence: Google documentation on canonicalisation and redirects.
Site healthMore from the data.
Priority = new-site control (1–3) × factors in the latest top 50 × |average r| × 10. The mapping is keyword-rule judgement, not measurement.
TopSEOFactorsSep 2026Share of the top 200 by category, 2019 → 2026
Average monthly share · open dot 2019, filled dot 2026
- Relevancy−5.8 pts
- Other+10.9 pts
- Size+2.4 pts
- Authority−8.5 pts
- Technology+0.6 pts
- Diversity+3.0 pts
- Trust+0.4 pts
- Unclassified−3.0 pts
- Performance±0 pts
Average |r| by year
Higher means a stronger link with position · 2026 covers 9 months
| Year | Top 10 factors | All top 200 |
|---|---|---|
| 2019 | 0.27 | 0.16 |
| 2020 | 0.25 | 0.16 |
| 2021 | 0.24 | 0.16 |
| 2022 | 0.28 | 0.17 |
| 2023 | 0.28 | 0.18 |
| 2024 | 0.24 | 0.16 |
| 2025 | 0.23 | 0.16 |
| 2026 | 0.23 | 0.16 |
Biggest gains, year on year
- 1Number of Social Accounts+96
- 2Has Meta Charset+91
- 3Number of H1 Tags+90
- 4Has Form+78
- 5Number of H2 Tags+76
- 6Is Home Page+74
Biggest drops, year on year
- 1LSI Words in Title Tag−110
- 2Clean Keyword Density in the HTML Tag−98
- 3Variations in Script Tags−91
- 4Entities in Meta Description Tag−88
- 5Variation Density in Top 30KB−73
- 6TF/IDF−71
Most durable, by average r
- 1Number of Distinct Entities Usedr −0.27 P1
- 2Number of Factors Usedr −0.24 P1
- 3Variations in HTML Tagsr −0.24 P2
- 4Variations in Body Tagsr −0.23 P2
- 5Variations in Div Tagsr −0.22 P2
- 6Variations in H1-H6 Tagsr −0.22 P2
The average tells you what a typical first-page result serves. The data is global and mostly US/English SERPs. Treat page-1 averages as benchmarks for Nordic markets, not quotas. For your market, SERP Lab measures your own top 20 instead.
Open the same library inside the product.
In the workspace, every factor links to the practices it backs and to your own pages measured against it.