Their content was technically fine. It just wasn't getting cited.
9 months later,ChatGPT and AI Overviews quote them by name.

The client had a strong content team and a respectable Domain Rating (DR 64) when they engaged us. On paper, their SEO was working — they ranked on page 1 for around 200 queries. But organic-attributed pipeline had been flat for 18 months, and the team couldn't explain why.

Our audit surfaced two compounding problems. First, the AI Overviews layer: 60% of their highest-intent commercial queries now triggered Google AI Overviews that synthesized answers from competitors — but never from them. Second, ChatGPT and Perplexity, which their buyers were demonstrably using (they showed up in attribution data as referral source), almost never named the client when answering category-defining questions.

Both problems traced back to content architecture, not content quality. Their blog was structured for 2020 — long, narrative essays without TL;DR boxes, FAQ schema, citable statistics, or the structural markers LLMs preferentially extract.

We didn't rewrite the blog. We rebuilt 22 priority pages — the ones mapped to the 40 highest-revenue queries — using citation-friendly structure: direct-answer first paragraphs, key-takeaway boxes, comparison tables, FAQ sections with proper schema, and original statistics with explicit sources.

Alongside the content rebuild, we ran a 6-month digital PR campaign focused on getting the client cited in publications LLMs are trained on — Forbes, industry trade press, founder-bylined contributions to category-defining publications. Five Tier-1 placements landed in months 3-6.

Technical foundations got fixed in parallel: schema saturation across the site (Article, FAQPage, Organization, Person, Service), Core Web Vitals work that brought LCP from 3.4s to 1.9s on the rebuilt templates, and an internal linking refactor that tied the rebuilt pages into a coherent topical cluster around the client's category.

By month 6, organic traffic had grown 180%. AI Overview citations started appearing on rebuilt pages in months 4-5 as Google's index caught up to the structural changes. ChatGPT citation rate, measured weekly via direct prompting across 30 priority queries, climbed from 3% to 36% over the engagement.

By month 9, organic traffic was up 347% on the baseline. The client closed $1.4M in pipeline attributed to organic over the engagement window — verified through their HubSpot attribution model, not our reporting alone. The 90-day refund clause was never invoked because directional traffic growth was clear by month 3, well before the formal checkpoint.

The most replicable lesson from this engagement is that content quality is no longer enough. The same 22 articles, written exactly the same way, would have failed to move the needle without structural changes — TL;DR boxes, FAQ schema, citable statistics, and entity-strength reinforcement through digital PR.

The second lesson is on timing. AI citation lagged 4-5 months behind on-page changes — Perplexity caught up first, Google AI Overviews second, ChatGPT slowest. Setting realistic expectations on this lag is part of the engagement scoping conversation we now have with every new AI SEO client.