ignax.dev Growth Engine — AEO Case Study

This is the meta case study — the one where ignax.dev runs its own SEO and AEO playbook on itself and publishes the results publicly. Bilingual SvelteKit content engine, FAQPage and Service schema on every page, llms.txt and llms-full.txt walked from the content registry, IndexNow ping on every publish, weekly GSC-fed loop, monthly update to this page. The playbook is the same one I sell as a service; this page is the live experiment. Day-0 today, updated monthly. Spain-born, Paraguay-based, running this in the open because the discipline of updating publicly is what makes the case study worth anything.

What is the playbook?

The playbook has three parts, all documented in the supporting service and article pages.

Content engine (§6 of the growth plan). Forty articles total — twenty EN plus twenty ES siblings — covering buyer-intent keywords across SaaS MVP work, AI automation, RAG chatbots, blockchain development, and SEO/AEO services. Each article is bilingual, schema-rich, with Quick Answer blocks in the first 200 pixels, H2-as-question structure, FAQPage schema attached, and explicit internal-link patterns to related services and case studies. The full article plan is in the public growth plan document; this page does not duplicate it.

Service page template (§8 of the growth plan). Nine service pages — four dev plus five SEO — following a strict template: H1 with buyer-intent keyword, Quick Answer, who-this-is-for, what's included, process, stack-or-methodology, pricing band, for-agencies block, FAQ, CTA, internal links. The SaaS MVP service and SEO + AEO setup service pages are the templates the rest follow. Every service page targets a specific buyer-intent keyword and links to at least two related case studies and two related articles.

Weekly growth loop (§13 of the growth plan). Every Monday: pull Google Search Console CSVs, identify keyword gaps and CTR opportunities, ship 2–3 new pages or rewrites, ping IndexNow, log AI-engine citations spotted across ChatGPT, Perplexity, Gemini, and Claude.ai for the target prompt set. The loop runs in 60–90 minutes when the content registry is healthy. Once a month, this case study updates with what shipped, what indexed, what ranked, what got cited.

The supporting articles fill in the methodology details. What is AEO? explains the underlying concept. How to get cited by ChatGPT is the practical checklist. llms.txt explained covers the specific file format. AEO citation checklist is the ship-before-you-expect-citations punch list.

What does Day-0 look like?

This is the baseline before the playbook has had time to work. Numbers below are placeholders for the first commit; Ignacio fills them in from the relevant dashboards. The honest move is to start empty.

• URLs indexed today (GSC + Bing): _______
• Total impressions last 28 days: _______
• Total clicks last 28 days: _______
• Top 10 queries (or "none yet"): _______
• Backlinks (Ahrefs WMT free): _______
• AI-engine citations (manual probe): _______

The probe methodology: ten EN and ten ES target prompts, each one a question a real buyer would ask before hiring someone in my service categories, run across ChatGPT, Perplexity, Gemini, and Claude.ai with citations turned on where the engine supports them. Results are logged with date, engine, prompt, whether ignax.dev appeared in the citation list, and which specific page was cited. The probe runs monthly; results feed the update below.

What did I ship this week?

First week — see the weekly updates log below.

What's the weekly update log?

TBD — first update on the first Monday of June 2026.

The marker comment above is the append point. A monthly cron run pulls the weekly summaries and appends them here in reverse-chronological order. Each entry is a short bullet list: articles published that week, services updated, case studies updated, queries discovered in GSC, AI-engine citations spotted in the manual probe. No analysis in the log — analysis goes in the next two sections, on a monthly cadence.

What's working and what isn't?

Day-0; TBD.

This section updates monthly with the honest read on the playbook. What is producing measurable lift on indexing, ranking, or citation count. What is not. What I have stopped doing and why. What I am about to try in the next month.

The point of publishing this section is not to look smart — it is to maintain the discipline of writing down what is and is not working, in public, on the same site that sells the methodology. A SEO consultant who cannot tell you what is currently failing in their own playbook is not a SEO consultant you should hire.

What did I learn?

TBD — will update after Day 30.

The full lessons section starts after the first month of running the playbook. Before that, anything I write here is speculation. The lessons that will actually generalize are the ones earned from the first month of indexing data, the first AI-engine citation hits, the first surprises in the GSC query report, and the first refactors of the playbook in response to what the data says.

Until then, the supporting articles cover what I already know going in. What is AEO?. How to get cited by ChatGPT. llms.txt explained. AEO citation checklist. The methodology in those articles is the methodology being run on this site; the differences between what I expected and what actually happened will land in this section over the coming months.

For the service this case study sells, see the SEO + AEO setup service page. The content engine service covers the related but different engagement of running the content side as a retainer rather than a setup. The SaaS MVP service covers the dev side of the same shop, for the readers who arrive here through the case study and realize they need the build, not the SEO.

External references for the technical pieces: Google Search Central documentation, Schema.org, Cloudflare Pages documentation, SvelteKit documentation.

Email hello@ignax.dev if you want this playbook run on your site. Repository style and commit cadence at github.com/ignaxdev. This case study updates monthly — bookmark it and check back, or read the supporting articles in the meantime.