Article No. 80
Long-Tail vs. Short-Tail vs. Head Terms: The Query-Length Taxonomy Explained
Abstract
Most explanations of this taxonomy define the three tiers by word count: one or two words is a head term, three or four is mid-tail, five or more is long-tail....
On this page
Most explanations of this taxonomy define the three tiers by word count: one or two words is a head term, three or four is mid-tail, five or more is long-tail. That’s a rough proxy at best, and it breaks down constantly. “New York personal injury lawyer” is five words and enormous search volume in its market. “Zyzz” is one word and vanishingly rare. Word count correlates loosely with the traits that actually matter; it isn’t the definition.
What Actually Defines Each Tier
The real distinguishing traits are search volume concentration, specificity, and competition level, not literally counting words in the query string.
A faster, more reliable way to place a keyword than counting words: look at the SERP itself. Head terms tend to surface broad category pages, major brands, and Wikipedia-style overview content. Mid-tail terms tend to surface a mix of category pages and more specific comparison or service content. Long-tail terms tend to surface individual blog posts, forum threads, and highly specific product or FAQ pages, often from smaller sites that would have no chance ranking for the head term in the same space. What’s actually ranking tells you more about a query’s real tier than its word count does.
Head Terms
High individual search volume, high competition, and typically ambiguous intent. A search for “shoes” or “insurance” could mean dozens of different things depending on the searcher: browsing, comparison shopping, researching a specific sub-category, or looking for a specific brand’s storefront. Ranking for it usually requires substantial existing authority or brand recognition, built over years, not a single well-optimized page. Head terms are largely brand- and authority-dependent to rank for; a new site has essentially no realistic path to page-one visibility on a competitive head term regardless of content quality, because the sites currently occupying that space typically have a scale of backlinks, brand search demand, and topical depth that a new or smaller site simply hasn’t accumulated yet.
Short/Mid-Tail
Moderate individual volume, more specificity than a head term, and a realistic ranking target for most established sites. “Running shoes for flat feet” or “small business liability insurance” sit here: specific enough to signal real intent, but not so narrow that volume disappears. This tier is usually where category and service pages live.
Long-Tail
Low individual search volume per term, but high specificity and, in aggregate across the enormous number of distinct long-tail phrases that exist, meaningful total volume. “Best running shoes for flat feet and plantar fasciitis marathon training” is a long-tail query: nobody else is searching that exact phrase very often, but a huge number of similarly specific, similarly narrow phrases exist, and collectively they represent real traffic. Long-tail terms also tend to carry higher conversion propensity per visitor, because the specificity of the query usually reflects a more developed, more decided searcher.
The “Long-Tail Search Share” Statistic Problem
You will see it stated as fact across SEO content: “70% of searches are long-tail.” Sometimes it’s paired with a companion claim that “15% of searches every day are new, never-searched-before queries.” Both numbers get repeated constantly, almost always without a live, checkable source attached.
Digging into where these figures actually come from turns up a mess, not a clean citation. Different sources cite wildly different percentages for essentially the same claim: some put long-tail’s share of search volume around 70%, others (using their own analysis of tens of millions of keywords) put it closer to 90%, and the definitions of “long-tail” being measured aren’t consistent between them. Some of the oldest versions of this figure are commonly traced back to comScore- and Hitwise-era web analytics studies from over a decade ago, though that attribution itself isn’t cleanly sourced either; at this point the claim has been cited so many times, by so many secondary sources, that the original methodology has effectively been lost. Newer analyses exist but use their own definitions and datasets, which produce different numbers that aren’t directly comparable to the older figures still circulating.
The honest position, and the one this guide takes: a large share of total search volume is long-tail, distributed across an enormous number of individually low-volume queries, but no single precise percentage of that share can currently be traced to one clear, current, methodologically transparent source that the rest of the industry has actually verified rather than just re-quoted. If you encounter a piece of content asserting a specific percentage with confidence and no named, checkable source, treat that percentage as folklore, not fact, regardless of how many other sites repeat it.
What’s genuinely well-supported, without needing a specific percentage: the number of unique long-tail phrases vastly exceeds the number of unique head and mid-tail phrases, and it’s this sheer count, not any single term’s volume, that gives long-tail traffic its aggregate weight. That’s a structural, definitionally true statement about how search query distributions work, independent of whichever contested percentage gets attached to it.
This matters practically, not just as a citation-accuracy exercise. If you’re building a content strategy or a client pitch around a specific claimed percentage of search volume being “long-tail,” that number is doing more work in the argument than it can actually support. Build the argument instead on what’s verifiable: long-tail queries are numerous, individually low-competition, and collectively meaningful, which is a strong enough case on its own without needing a precise, unsourced figure attached to it.
The Monetization Angle
Long-tail terms often convert better per visitor than head terms, and the mechanism is straightforward: a more specific query usually reflects a more decided searcher. Someone searching “shoes” could be browsing, researching, or ready to buy. Someone searching “waterproof steel-toe work boots size 11 wide” has largely already decided what they want and is closer to a purchase decision.
That said, this doesn’t hold as a universal, quantifiable conversion-rate lift you can apply across industries. Conversion behavior varies enormously by category, by price point, and by what “conversion” even means for a given business (a $12 e-commerce purchase and a $40,000 B2B contract have completely different decision paths). Treat the conversion advantage as a real, directionally consistent pattern worth acting on, case by case, not a fixed percentage lift to plug into a projection. Where commercial intent signals actually live within a query, independent of its length, is its own classification method with its own dedicated guide.
Query length and commercial intent correlate but aren’t the same thing, and conflating them causes real strategic mistakes. “Buy iPhone 15” is short and highly commercial. “What’s the difference between iPhone 15 and iPhone 15 Pro camera specs” is long-tail and still largely informational, closer to research than to a purchase decision. Length tells you about specificity and likely competition; it doesn’t by itself tell you where someone is in a buying journey.
Practical Takeaway: You Need All Three Tiers
A real site’s keyword portfolio needs representation across the taxonomy, targeted differently by tier:
| Tier | Typical page type | Role |
|---|---|---|
| Head terms | Brand pages, top-level hub pages | Long-term authority-building target, not a realistic near-term ranking win for a newer site |
| Short/mid-tail | Category and service pages | The realistic backbone of most sites' organic visibility |
| Long-tail | Deep, specific content and FAQ-style pages | Individually small, collectively substantial, and often the fastest realistic wins for newer or lower-authority sites |
A site built entirely on long-tail content can generate real traffic without ever cracking a competitive head term, but it usually caps out below what a mix including strong mid-tier category pages can achieve. A site chasing only head terms, without the mid- and long-tail content that actually converts and builds topical depth, tends to struggle to rank for the head terms in the first place, since search engines weigh a site’s overall topical coverage when deciding whether it deserves visibility on the most competitive, ambiguous queries in that space.
The practical sequence for most sites without existing authority: build the mid-tail and long-tail layer first, since it’s realistic to rank for sooner and it establishes the topical depth that eventually supports competing for head terms later.
None of this is a fixed formula that applies identically across every industry. A local service business might realistically never need to compete for a national head term at all, and its entire viable keyword portfolio could sit in the mid- and long-tail tiers permanently, mapped to specific services and specific service areas. A national e-commerce brand, by contrast, may have head-term competition as a genuine, achievable near-term goal once it has enough backlinks and brand recognition behind it. The taxonomy is the same in both cases; what changes is which tier represents the realistic ceiling for that specific site.