How a 5-Week-Old Website Got Cited by Gemini: The Auroxa-on-Auroxa Case Study

Every SEO tool promises results. Almost none of them can show you what happens when they run their own playbook on themselves. So we did exactly that — and we're publishing the receipts, including the parts that haven't worked yet.

This is an interim report. Auroxa's own domain went live at the end of May 2026. Everything below happened in roughly five weeks, and every claim is dated so you can check the timeline yourself. We'll publish the full 90-day follow-up with rank charts and traffic data at the end of August.

What Exactly Did We Do?

We pointed Auroxa at auroxa.app — the same product our customers use, no special treatment — and ran the standard workflow:

1. Planned a 24-article content map across four topic pillars (answer engine optimization, AI search, SEO for solopreneurs, and content operations): one cornerstone guide per pillar plus five supporting articles that link up to it.
2. Generated every article with the platform, reviewed each one by a human before publishing — the same review-and-publish mode our customers use — and ran the built-in fact checker on drafts that made statistical or third-party claims.
3. Published all 24 articles to our own blog between May 30 and June 8, 2026. Average human time per article, including review and internal linking: under an hour.
4. Fixed what the audit told us to fix — including, as you'll see below, an embarrassing gap on our own homepage.

Can a Brand-New Domain Show Up in AI Answers?

That was the real question. Rankings on a five-week-old domain were always going to be slow — Google takes months to trust a new site, and no honest tool tells you otherwise. But AI answer engines re-fetch and re-evaluate sources much faster, which makes them the first place a new site can prove itself.

Here's the timeline of what happened:

• June 3: Sitemap submitted to Google. 58 URLs.
• Early July: Google Search Console showed 24 pages indexed, with the rest sitting in the normal new-site crawl queue — no penalties, no exclusions, just the standard waiting room every new domain sits in.
• July 4: We tested Gemini with grounding (its live Google-search mode). It found auroxa.app — but it blended us with an unrelated company that has a similar name, presenting the two businesses as one, and it dropped part of our pricing.
• July 9 — five days later: the same test returned an accurate, standalone description of Auroxa — what the product does, how the workflow runs, even our human-approval publishing mode — citing auroxa.app itself as the source, three times. The entity confusion was gone.

How Did We Fix the Entity Confusion in Five Days?

With the least glamorous tool in SEO: structured data.

When Gemini mixed us up with the similarly-named company, we ran our own audit logic on our own homepage and found something genuinely embarrassing for an answer-engine-optimization company: the homepage had zero structured data. Our blog articles carried full schema markup (the platform adds it automatically at publish), but the hand-built marketing pages never got the same treatment.

On July 4 we shipped an Organization schema block — a machine-readable statement of who we are, what we do, our logo, and links to our verified social profiles — plus a WebSite schema block, on the homepage and about page.

Five days later, the confusion was gone from grounded Gemini answers. We can't prove causation with a sample size of one — Google's index was also still catching up on a young domain, and that alone helps. But the sequence is exactly what structured-data theory predicts: give the machine an unambiguous statement of your identity, and it stops guessing.

The takeaway for your business: if an AI engine describes you wrong — or blends you with someone else — that's usually not bad luck. It's missing structured data and thin corroborating signals. It's fixable, and quickly, because AI engines re-evaluate sources on a much faster cycle than traditional rankings move.

What Do the Content Scores Say?

Every article was scored by the platform's SEO and AEO analyzers before publishing (AEO measures citation-readiness for AI engines: question-and-answer structure, fact density, schema completeness, citable formatting).

Across the 24-article run:

• Every article reached an AEO score of 90 or higher before publishing — most needed either zero or one AI improvement pass to get there.
• Several hit perfect or near-perfect AEO scores (two articles scored 100).
• SEO scores at publish ranged from 81 on the earliest cornerstone to 97 on later articles — the pipeline itself improved mid-run, because every weak score got treated as a bug and fixed.

That last point is the part we'd emphasize to any skeptic: we used the run as quality control on ourselves. When an article scored low, we didn't shrug — we found out why, fixed the underlying rule, and the fix shipped to every customer.

What Hasn't Worked Yet? (The Honest Part)

A case study without a failure section is an ad. Here's ours:

• Non-branded AI citations: not yet. Ask an AI engine a broad question like "what is answer engine optimization" and it cites the established players — the incumbents with years of authority. Our articles aren't in those answers yet. Branded and product-specific queries: yes. Head terms: no. That's the accurate picture of what five weeks gets you.
• Google rankings: early. Our first top-100 positions appeared in July on long-tail keywords. Nothing on page one for competitive terms. Anyone who promises page-one Google rankings on a five-week-old domain is lying to you.
• Traffic: small. Indexing and citations lead; clicks lag. That's the normal order.

What Should You Take From This?

Three things, whether or not you ever use our product:

1. AI answer engines are the fastest feedback loop a new or small site has. Google rankings take months. Gemini corrected its understanding of our business in days once we gave it structured data.
2. Being described wrong by AI is a bigger risk than being invisible — and it has a concrete fix: schema markup, consistent entity signals, and text on your site that says plainly what you do.
3. Volume without review is not a strategy. We published 24 articles fast, but a human approved every single one, and the fact checker flagged claims before they went live. That workflow — AI speed, human judgment — is the entire reason we're comfortable publishing this with our name on it.

We'll update this case study at the 90-day mark with rank movement, traffic charts, and backlink growth — including anything that flopped.

Want to see what an AI engine currently says about your business? Run the free 60-second audit and check.