Question 1

Why don't traditional case studies show up in AI search recommendations?

Accepted Answer

Traditional B2B case studies are written as persuasion documents for human buyers at the bottom of the funnel — they lead with a narrative, bury the quantitative outcome, gate the full document, and use prose structures that AI retrieval systems cannot cleanly extract. AI assistants cite case studies when they need to answer queries like 'what results have companies seen from X' or 'does Y work for companies in Z industry.' To serve those answers, the model needs a named company, a specific metric, a methodology description, and a clearly bounded outcome — all surfaced in the first 300 words of an uncrawlable page. Most traditional case studies deliver none of these requirements. The company name is sometimes anonymized, the headline metric is buried three paragraphs down, the full document is behind a form, and the page itself is JavaScript-rendered and invisible to AI crawlers. Fixing these four structural failures transforms an invisible case study into a high-citation asset.

Question 2

What structure makes a case study citeable by ChatGPT and Perplexity?

Accepted Answer

The AEO-optimized case study opens with a data hook in the first sentence — a specific company name, a specific percentage improvement, and a time period. It follows with a 100-150 word summary section that stands alone as a complete answer (company name, problem, solution deployed, result, timeframe). It includes a structured results table with metric names, baseline values, outcome values, and percentage change. It describes the methodology in a dedicated H2 section with named steps. And it contains at least one pull-quote from a named executive with a job title. These structural elements match what retrieval-augmented generation systems look for when chunking a document. The summary section becomes a self-contained citation chunk. The results table gets extracted for quantitative queries. The methodology section answers 'how did they do it' queries. The executive quote provides the social proof signal. AI assistants cite documents that make extraction easy — the AEO case study is designed for machine consumption first and human persuasion second.

Question 3

Should case studies be gated or ungated for AEO?

Accepted Answer

For AEO purposes, case studies should be ungated — full stop. A case study behind an email-capture form is invisible to AI crawlers and therefore contributes zero citation value. The lead-generation argument for gating is real but increasingly weak: gated assets produce a small number of high-intent leads now at the cost of all AI-search citation value forever. The better model is to publish the full case study as an indexed HTML page and use behavioral signals — retargeting, intent data from visitor tracking, direct outreach triggered by firm-level identification tools like Clearbit or 6sense — to capture demand without a form gate. For companies that cannot let go of gating entirely, the minimum viable compromise is to publish a full-length, fully indexed summary page (600-1,000 words with all the key data) alongside the gated PDF version. The summary page builds citation authority; the PDF captures the leads who want the deeper version. Any case study that is only available as a gated PDF is not an AEO asset — it is a brochure.

Question 4

What specific data points should a case study include for AI citation?

Accepted Answer

Six data categories appear most frequently in AI-cited case studies. First, a primary outcome metric with percentage improvement and time period — '43% reduction in time-to-close over 6 months.' Second, a baseline-to-outcome comparison — 'from 22 days average to 12.5 days.' Third, a scale signal — company size, revenue range, or transaction volume — that tells AI models which reader this applies to. Fourth, an implementation timeline — 'deployed in 8 weeks' or 'saw first results within 30 days.' Fifth, a named methodology — 'using the three-phase onboarding protocol.' Sixth, a direct executive quote with full name and title that attributes the outcome to a specific person. Data points without a named company are cited significantly less often than data points attached to a real organization — anonymized case studies produce almost no AI citations because AI assistants need verifiable entity references to validate claims. If clients insist on anonymity, use the industry and company size rather than the company name: 'a Fortune 500 healthcare distributor' outperforms 'a large company.'

Question 5

How do you use schema markup to make a case study more visible to AI crawlers?

Accepted Answer

The most effective schema type for AEO-optimized case studies is a combination of Article schema (with articleBody, datePublished, and author fields) and a nested Review or Claim structure for the quantitative outcomes. The Article schema ensures the page is treated as authoritative editorial content rather than a product page. The datePublished field provides the freshness signal AI models use to weight currency. Adding an Organization schema block for the client company — even with minimal fields like name and industry — connects the case study to the entity graph that AI models use to validate citations. For case studies describing a software or service implementation, adding HowTo schema to the methodology section dramatically increases citation probability for 'how does X work' queries. The full schema stack for a case study should include: Article (top-level), Organization (for the client), Person (for quoted executives), and HowTo (for the implementation methodology). This four-schema stack is implemented by fewer than 5% of case study pages in the wild, which means teams that implement it have a structural citation advantage.