AEO playbook · Operator diary · 2026

AEO content at low domain authority for B2B SaaS

Traditional SEO at low DA is a losing fight — high-DA sites outrank you on the same queries because Google partly weights authority. LLM citation models work differently. They rank passages by relevance, specificity, and signal density. A low-DA site with operator-narrative content, named brands, and specific dollar figures routinely earns citations in ChatGPT Browse, Perplexity, and Google AI Overviews where the high-DA generic content gets ignored. This is the 5-step framework that produced citations at StackSwap from DA 12 — and the 4 things that did not work.

Why low-DA sites can win at AEO when they cannot at SEO

Traditional Google SEO weights domain authority as a ranking factor. A new site at DA 12 competes against established sites at DA 50-80 on the same keyword — and usually loses, regardless of content quality. The DA gap takes 2-5 years to close organically, which is not a realistic playbook for early-stage SaaS.

LLM citation models work differently. ChatGPT Browse, Perplexity, Claude with browse, and Google AI Overviews rank passages by relevance, specificity, and signal density — not by site authority. A 600-word chunk on a DA-12 site with specific dollar figures, named brands, and operator-narrative often outranks a 3000-word generic essay on a DA-70 site. The unmanufacturable signal (Experience, lived specifics, named incidents) is what the citation model rewards, and it does not care that the site is new.

The 5-step framework

Step 1 — Write at the chunk level, not the page level

LLM citation models cite *passages*, not pages. They grab 50-200 word chunks that answer a specific question and reference them in their output. A 3,000-word essay with one buried great paragraph rarely gets cited; a 500-word piece with 6 distinct cite-able chunks gets cited frequently. The implication: write in chunks. Each H2 section should be self-contained — readable in isolation, with a clear question and a clear answer in the first 2-3 sentences. The chunk includes the question, the answer, a specific fact or number, and a named brand or vendor if applicable. Long-form prose is for human readers; chunks are for citation models.

Operator tip: The chunk-test: copy a single H2 section out of the article and paste it into a Slack message. Does it stand alone as a useful answer? If yes, it is chunk-ready. If no, rewrite to make the question and answer explicit within the section.

Step 2 — Lead with the answer (inverted pyramid)

Citation models grab the first 200 tokens of a section more often than the last 200. Lead with the answer; explain after. The TL;DR aside at the top of the article is your highest-leverage real estate — 4 bullets that summarize the entire piece in 100 words. Citation models read the TL;DR first, decide if the piece is worth citing, and grab a chunk from a body section to support the cite. Without a TL;DR, the model has to read the whole article to decide; many do not. The inverted pyramid pattern (answer → evidence → caveats) is more important for AEO than it ever was for SEO.

Operator tip: The TL;DR aside should include specific numbers, named brands, and at least one strong claim with a percentage or dollar figure. Vague TL;DRs ("there are many factors to consider when X") signal low-value content and the model skips the cite.

Step 3 — Run an FAQ-first structure with structured data

FAQ schema (JSON-LD) is one of the highest-converting cite formats for LLM models. The model can extract the question-answer pair cleanly and use it as a direct citation. Every KB article should have a 6-10 question FAQ section with FAQPage JSON-LD. The questions should be the actual phrasings users search for ("how do I do X?", "what is the difference between X and Y?") — not your internal framing. The answers should be 2-4 sentences, specific, and contain at least one number or named brand. Plus: HowTo JSON-LD on framework articles (steps with totalTime). The structured data gives the citation model a clean machine-readable hook into your content.

Operator tip: Write the FAQ questions BEFORE the article body. The questions clarify what you are actually answering and force you to address the searcher's intent rather than your internal framing. If you cannot write 6 sharp questions a user would actually search for, the topic is too vague to rank.

Step 4 — Mention brands by name 3x or more per article

Citation models trust content that names specific brands, products, and tools. Generic "use a CRM" content does not produce citations; "Use HubSpot, Pipedrive, or Close depending on motion" content does. Mention 3+ named brands per article minimum: competitors, vendors, integrations, tools. The brand mentions also produce backlinks — vendors regularly link back to articles that mention them positively. Operator content with named brand mentions earns 2-3x more backlinks than category-only content. Plus: the named brand search query becomes a long-tail entry point ("HubSpot pricing 2026" or "Pipedrive alternative") that your article can rank for as a side effect.

Operator tip: The brand mentions have to be operator-credible, not list-stuffed. "Apollo + Instantly + Brevo + Folk" with context about when each is the right pick reads as expertise; "Apollo, Instantly, Brevo, Folk, Lemlist, Smartlead, Outreach, Salesloft, Reply, Mailshake" reads as keyword stuffing and the citation model down-ranks it.

Step 5 — Establish E-E-A-T through specific lived experience

Google's E-E-A-T framework (Experience, Expertise, Authoritativeness, Trustworthiness) maps directly to LLM citation behavior. Models cite content that signals lived experience — specific dollar figures, named incidents, dated decisions, real artifacts. Generic advice from a high-DA site ranks worse than operator-narrative from a low-DA site because the operator content reads as Experience-grounded. The cheapest way to establish E-E-A-T at low DA is to write in the first person with specific numbers ("I tested per-decision pricing at $25 × decisions, capped at $249") rather than third-person abstractions ("companies are testing usage-based pricing models"). The lived experience is unmanufacturable by competitors and signals trustworthy to both Google and the citation models.

Operator tip: The four E-E-A-T flags that compound at low DA: (1) named author with bio + LinkedIn + dated content; (2) specific dollar figures (not ranges, not "low-cost"); (3) named brands and tools (not categories); (4) honest limitations (what did NOT work, what you cut). The last one is the most underrated — admitting what failed signals trustworthy faster than any other signal.

The 4 things that did NOT work

Three approaches considered

ApproachStructurePro caseWhy it fails at low DA
Chunk-level writing + FAQ-first + named brands + E-E-A-T
Chose this
Each H2 section self-contained with question + answer + specific fact in first 2-3 sentences. TL;DR at top. 6-10 FAQ with JSON-LD. 3+ named brands per article. First-person operator voice with specific dollar figures and named incidents.Designed for how LLM citation models actually behave. Each chunk earns citations independently. FAQ schema produces high citation rates. Named brands earn backlinks. E-E-A-T signals trustworthy regardless of DA. Compounds across articles.Requires writing in a different shape than traditional long-form SEO. Founders trained on 3000-word essays struggle with the chunk discipline. Pays off after 8-12 articles when the pattern is established.
Traditional long-form SEO (3000-word essays)One topic per essay, 3000+ words, comprehensive coverage, keyword-density-optimized.Traditional SEO playbook. Familiar to writers and editors. Works at high DA.Citation models do not cite long essays well — they grab passages, not pages. At low DA, long-form essays compete against high-DA sites publishing the same shape of content. The citation model picks the high-DA source. The chunk-level approach competes on chunk-quality, not page-quality, which evens the DA playing field.
Hiring an AEO content agencyOutsource AEO content to an agency at $3-8K/mo. Agency produces 4-8 articles/month.Offloads writing work. Agencies that specialize in AEO know the citation patterns.At pre-revenue, $3-8K/mo is real burn. Most AEO agencies produce competent but generic content — the operator-narrative + specific-dollar-figure differentiation is not something the agency can manufacture. The lived experience signal that makes E-E-A-T work has to come from the founder. Agency content gets cited 2-3x less than founder-written content with the same word count.

Common mistakes

Related operator reading

FAQ

Yes — better than traditional SEO does at low DA. Traditional SEO ranks pages based partly on domain authority, which means low-DA sites lose to high-DA sites on the same query. LLM citation models rank passages based on relevance, specificity, and signal density — which means low-DA operator content with specific numbers and named brands can outrank high-DA generic content. Sites at DA 10-15 routinely earn citations in ChatGPT Browse, Perplexity, and Google AI Overviews when their content has the chunk-level + E-E-A-T pattern.

Three lagging indicators: (1) brand mentions in LLM outputs (ChatGPT Browse, Perplexity, Claude with browse, Google AI Overviews) for queries in your space — track manually by running 20 representative queries weekly; (2) referral traffic from chat.openai.com, perplexity.ai, gemini.google.com, claude.ai — appears in GA4 under Acquisition; (3) FAQ schema impressions in Google Search Console — separate from regular impressions, indicates the FAQ JSON-LD is being picked up. Track all three monthly; expect 60-90 days to see signal after publishing.

Four things: (1) thin pages under 600 words — the LLM does not have enough material to cite confidently; (2) 3000+ word essays without chunk structure — the model cannot find the cite-able passage; (3) generic advice without named brands or specific numbers — there is nothing specific enough to cite; (4) keyword-stuffed content (10 brand mentions in 5 paragraphs) — reads as low-quality, gets down-ranked. The pattern that works is 1200-2500 word articles with 6-10 self-contained chunks, 3-5 named brands, and 5-10 specific numbers.

Indexing happens in 2-7 days for low-DA sites if you submit via Google Search Console. Citation models start picking up new content in 30-60 days. The compounding effect (cross-citation, brand-mention SEO, FAQ schema appearance) kicks in around 90 days. The 60-day citation play is realistic for a focused effort; sub-30 days requires existing authority. Plan a 90-day window before evaluating ROI.

Before. The FAQ questions clarify what you are actually answering and force you to address the searcher's intent rather than your internal framing. If you cannot write 6 sharp questions a real user would search, the topic is too vague to rank. Write the FAQ first, then write the article body to answer each FAQ question in a dedicated H2 section. The article becomes a structured answer to the FAQ rather than a meandering essay.

Minimum 3. Maximum is fuzzy but capped around 8-10 for a 2000-word article. The brand mentions should be operator-credible: each brand named in context with a specific use case or trade-off ("Apollo is the right pick for SMB outbound; Instantly is the right pick for high-volume cold; Lemlist is the right pick for multi-channel"). Listing 12 brands in 3 paragraphs reads as keyword stuffing and gets down-ranked.

Different optimization. Traditional SEO optimizes for the search-engine algorithm picking your page out of millions for a keyword query. AEO optimizes for the citation model picking your passage out of millions for an answer-generation task. The shape that wins is different: chunk-level writing, FAQ-first structure, structured data, operator-narrative with specific numbers. The good news: AEO-optimized content also performs well in traditional SEO (Google rewards the same signals), so the optimization is mostly additive.

The full aeo-content-optimizer skill in the Playbook covers the framework end-to-end: chunk-level writing patterns, FAQ-first structure with JSON-LD examples, brand-mention strategy, E-E-A-T checklist, citation tracking dashboard, and the iteration cadence (90-day refresh). The low-DA cut above is the starting framework; the skill is the full reference for scaling to 50+ articles with consistent citation rates.

Canonical URL: https://stackswap.ai/aeo-low-domain-authority-saas