In Semrush's 2024 study of US mobile results, half of page one ran under 1,069 words. So my posts cover the topic, then stop.
Most of the confident numbers repeated about blog posts trace back to another blog post, not a study. I went back to the research on eighteen questions and 142 sources. Three findings held up: length does not drive rankings, attention peaks in the first few hundred words, and citations build trust. Where the research ran out, the rule says it is my call.
Sources checkedOctober 2026
Last revised2026-10-12
Next reviewJanuary 2027
ByChuck Aikens
one-person consultancy for consumer product brands
15 years
Shorter posts. Answer first. Citations judged.
The rules
All 7 rules. Three shown in full.
1RULE 1 OF 7 · MEASURED
Length
Write 1,000 to 1,200 words for a listicle, 1,200 to 1,500 for a supporting post, and 1,500 to 1,800 for a pillar post. Cover the topic completely, then stop.
What backs it
Semrush's study of 16,298 US keywords, mobile results, puts the top-10 average at 1,451 words and the median at 1,069. It calls the link between word count and rank “minimal at 0.02”. A vendor study; the median sits in the report's benchmark tables.Semrush, Ranking Factors Study 2024 (vendor study)
In Backlinko's 11.8 million Google results, “word count was evenly distributed among the top 10 results,” the first-page average was 1,447 words, and it found “no direct relationship between word count and rankings.” Top-10 results only, and a correlation study. Backlinko discloses it is owned by Semrush, so their agreement counts once.Backlinko, We Analyzed 11.8 Million Google Search Results (2025) (vendor study)
3 steps to apply this are in the full Guideline.
2RULE 2 OF 7 · MEASURED | EXPERT PRACTICE
The opening
Section one answers the question the reader searched for. No context first, no warm-up. The answer, then the rest.
What backs it
In a 2015 reading study of 254 people reading texts screen by screen, “the deepest engagement is in the first 200-400 words, then noticeably decreases and slowly decreases thereafter”.Goedecke et al., Breaking Off Engagement (EDM 2015)
In Nielsen Norman Group eye-tracking of 120 people on 1920x1080 desktop screens, users spent about “57% of their page-viewing time above the fold. 74% of the viewing time was spent in the first two screenfuls”. That is viewing time across page types, not memory. Answering first is my judgment about where readers look.Nielsen Norman Group, Scrolling and Attention (2018)
3 steps to apply this are in the full Guideline.
3StructureIn the full Guideline
4LinksIn the full Guideline
5RULE 5 OF 7 · MEASURED
Citations and facts
Cite with inline links, up to three per post and at most one per paragraph, with no footnotes and no Sources section. None is fine. Judge the source and the page it sits on, and verify every link live before publishing. Health and wellness brands carry at least one. Facts run one per section, about five across a post.
What backs it
In a 2001 Stanford survey of 1,410 people rating 51 site elements from −3 (“Much less believable”) to +3 (“Much more believable”), “The site links to outside materials and sources.” averaged +1.25. “The site links to a site you think is not credible.” averaged −1.53.Fogg et al., What Makes Web Sites Credible? (CHI 2001)
The 2002 Stanford-Makovsky follow-up, 1,481 people on the same scale, found +1.17 and −1.38 for the same two items. Linking to a site readers think is not credible costs more than linking out earns. These are self-reported perceptions of believability, so I make no ranking claim.Stanford-Makovsky Web Credibility Study (2002)
4 steps to apply this are in the full Guideline.
6Reading levelIn the full Guideline
7Calls to actionIn the full Guideline
The short version
Cover the topic, then stop.
Shorter posts that open with the answer, links and calls to action in deliberate positions, and citations judged instead of counted. You get the same amount of content. Each piece is asked to do a clearer job.
How this Guideline gets made
Eighteen questions and 142 sources, studies over summaries
Three rules and one myth of seven. Clients read the whole Blog Guideline on app.tymoo.ai, updated quarterly. When the research moves, I update the rules and say when.