LLenny's Podcast
← All frameworks
MarketingEthan Smith (Graphite)

Head vs Tail in Answer Engines

Why AEO rewards citation frequency and hyper-specific long-tail questions

Difficulty
Moderate
Time to result
~days to results
Steps
3
Confidence
90%

A mental model contrasting how answer engines differ from Google search at both the 'head' (broad, high-volume queries) and the 'tail' (specific, low-volume queries). At the head, winning means being mentioned most often across citations rather than ranking first. At the tail, the volume of ultra-specific questions is far larger than in search, and much of it has never been asked before — so early-stage companies can win immediately by answering questions no one else covers.

Origin

Ethan Smith's framing, contrasting his 2007 long-tail SEO era (a page per keyword) with how the long tail has 'come back' in chat.

Core principles

  • 01Head: LLMs summarize many citations, so mention-count beats rank position
  • 02Tail: chat questions average ~25 words vs ~6 for Google, so specificity is far higher
  • 03Many tail questions have never been searched before because search can't support hyper-specific queries
  • 04Anyone can win the tail quickly by being the only citation answering a niche question

How to run it

  1. 1

    Attack the head by maximizing mentions

    For broad questions like 'what's the best website builder,' focus on being cited across as many sources as possible rather than holding one top URL, because the answer is a summary of the most-mentioned options.

    Pro tip The option mentioned most across the citations tends to become the first answer.

    Watch out A #1 URL that is only cited once will lose to a competitor mentioned many times.

  2. 2

    Map the long tail from real conversations

    Enumerate the highly specific follow-up questions people ask — about use cases, features, integrations, and languages — by mining sales calls, support tickets, and Reddit, since these mirror what people ask in chat.

    Pro tip Turn your existing search terms into questions as a starting proxy, then extend with mined real questions for the parts of the tail search can't show you.

  3. 3

    Own niche questions no one else answers

    Publish content answering ultra-specific questions (obscure but non-zero use cases) so you become the sole citation and win that answer outright.

    Pro tip Early-stage companies should do only citation optimization plus long-tail and skip mid-funnel SEO entirely.

    Watch out These questions have low individual volume, so value them by intent/LTV, not raw traffic.

In the wild

Newly launched startups winning obscure questions

Ethan describes early-stage companies that launched a very specific AI-enabled payment-processing API and showed up in answers immediately because they answered a question that had never been answered before.

They appear in LLM answers the day after launch, without domain authority, by owning a previously unanswered question.

Common mistakes

Applying Google's rank-first logic to LLMs

Being first in the citation list doesn't win the answer; the model blends many mentions, so frequency across sources matters more than position.

Is it for you?

Best for

Early-stage founders and growth marketers who lack domain authority but can move fast on specific content and citations

Not ideal for

Established brands whose primary goal is broad-head traffic through classic domain-authority SEO

From the transcript

the head and the tail are different

10:30

in the LM because the LM is summarizing many citations. And so you need to get mentioned as many times as possible

11:00

the average number of words I think perplexity said this... was around 25 words where versus Google words it's around six words

questions that have never been asked or searched for before that are now being asked and then you can win that

14:00

early stage my recommendation is don't do SEO at all. For uh for answer engine optimization definitely do AEO and only do citation optimization and…

40:30

From the episode

The ultimate guide to AEO: How to get ChatGPT to recommend your product

Ethan Smith (Graphite)