Organic and AI search visibility for a training institute

The context

The Business Profile work covered the Maps side of this client. The website had to carry the rest: the long-tail queries about specific certifications, course durations and accreditation that people research before they pick up the phone.

This is also a bilingual market. The same person may search in Arabic on a phone and in English on a laptop, and the site has to be legible to both.

What I was solving

1. 01Course pages existed but were not mapped to the way people actually phrase certification searches.
2. 02Site structure made it hard for crawlers — and for readers — to see the relationship between a course, its accreditation and its outcomes.
3. 03Nothing was in place to earn visibility in AI-generated answers, which increasingly sit above the organic results for exactly this kind of informational query.

How I did it

1. 01Mapped keywords to intent rather than volume, so each page answers one question properly instead of several badly.
2. 02Rebuilt the content architecture around topical clusters — course, accreditation, outcome — with internal linking that reflects it.
3. 03Added structured data so the entities on the page are unambiguous to a machine.
4. 04Wrote for answer-engine extraction: direct answers near the top, clear headings, no burying the point.

What the data shows

01 / 02

02 / 02

Search Console performance over 28 days, and the generative-AI breakdown showing impressions inside AI answers.

On the AI numbers

The 9.76K AI-answer impressions are the part I would watch. That surface barely existed a year ago, and most local service sites are still invisible in it.

An average position of 8.4 across 51.4K impressions also says something useful: there is real headroom. Moving a page from position eight to position four is usually a content and internal-linking problem, not a link-building one.