Back to Home
Case Study

Optimizing Our Own Documentation

We ran Viclaro on our own quick-start guide to prove the playbooks. Here's how we audited citation risk, fixed the structure and evidence gaps, and reran the same prompts to confirm assistants finally cited it.

The Challenge

Our quick-start guide had grown organically. The info was solid, but assistants had nothing clean to quote. We wanted to:

  • Validate that our own recommendations actually drive citations.
  • Identify the structural gaps blocking retrieval.
  • Create a public example of the Viclaro workflow.
  • Make sure our docs follow the playbooks we pitch.

The Process

1

Initial scan

We ran the quick-start page through Viclaro to see how each axis scored and where assistants hesitated.

2

Identified issues

We found missing heading anchors, weak quote targets, and schema gaps that left assistants without proof.

3

Applied fixes

We shipped the recommended edits: heading anchors, cleaner H2/H3 hierarchy, and FAQ schema with actual buyer questions.

4

Rescanned & validated

We rescanned the page to confirm the before/after deltas and capture new transcripts.

Key Improvements Made

1. Added Heading Anchors

We added ID anchors to key sections like "Run a Scan" and "Share Results" so assistants can quote exact passages instead of paraphrasing us.

<h2 id="run-a-scan">Run a Scan</h2>

2. Improved Content Structure

We broke down long, continuous sections into smaller, task-oriented subsections with clear H2 and H3 headings. That gave assistants quote-sized chunks and reduced hallucinated summaries.

Before: One 680-word section

After: Four focused sections with clear headings

3. Added FAQ Schema

We identified common questions in our content and structured them as FAQ schema markup. This gives assistants authoritative language to cite verbatim.

<script type="application/ld+json">
  { "@context": "https://schema.org", "@type": "FAQPage", ... }
</script>

4. Enhanced Hero Summary

We rewrote the page introduction to be answer-first and explicit about outcomes, giving crawlers and assistants a concise thesis to reuse.

Results & Learnings

What we learned

The process works. Running our own content through Viclaro gave us concrete recommendations with citations attached — catching issues we missed even after living in the doc.

Small changes, big impact. Simple edits — heading IDs, tighter sections, FAQ schema — materially improved how AI assistants interpreted and cited the page.

Validation matters. The rerun confirmed assistants cited the page more often. That proof matters more than a prettier score.

Practice what you preach. Running Viclaro on our own content keeps us honest about the experience we expect customers to follow.

The Workflow

We shipped everything through a normal Git branch, merged only after the rerun validated the lift. That is the workflow we recommend to customers:

  1. Run an initial scan to establish a baseline
  2. Review findings and prioritize high-impact fixes
  3. Implement changes in a controlled environment
  4. Rescan to validate improvements
  5. Deploy and monitor

Key Takeaways

Our own content now follows the same best practices. We can point to transcripts when prospects ask if Viclaro actually drives citations.

Structured data matters. Adding FAQ schema and real heading structures gave assistants clean chunks to cite.

The workflow is reproducible. Anyone can follow the same loop to improve interpretability, recall, and citation readiness.

Ready to see how AI talks about you?

Start with a free scan, get the same style of guidance, and rerun until the transcripts prove it.