CASE STUDY 02 · SOLO BUILD
Build, own
and operate.
I built kayjays.de for a Munich dance school in January 2025 and have operated it since. I own the information architecture, content, technical setup, publishing and ongoing search work.
LIMITATION, STATED FIRST
This is not a rebuild uplift claim.
GA4 records zero sessions in 2024 and effectively begins with the January 2025 relaunch. Search Console also has no pre-launch baseline. There is no defensible “before versus after the rebuild” comparison. The growth below is a like-for-like year-over-year comparison after launch.
LIKE-FOR-LIKE GA4
Organic sessions grew faster than the site overall.
From 1 January to 24 August, total sessions grew from 4,087 in 2025 to 9,203 in 2026, an increase of 125%. Organic Search sessions grew from 1,944 to 5,221, an increase of 169%.
Organic Search increased from 47.6% to 56.7% of all sessions. Its 70.5% engagement rate also exceeded Direct traffic’s 39.9%, indicating that the growth brought engaged visits rather than volume alone.
By 24 August, 2026 had already exceeded all of 2025 for total sessions: 9,203 compared with 7,547. This is useful operating evidence, but it is not a causal experiment and does not isolate the contribution of any one website change.
Eight course types needed one understandable system.
The school offers shuffle, K-pop, breakdance, dancehall, Bauchtanz, Kindertanz, hip hop and ballet. Each category needs its own search relevance without turning the website into a collection of near-duplicate local landing pages.
I structured course pages around distinct intent, audience, schedule and practical information. Shared location and school-level information stays central. This keeps the journey navigable for students and parents while giving search engines a clear relationship between the school, its Munich location and each course.
Local-intent terms reached the first page.
The Search Console export records “shuffle dance münchen” at average position 6.65 with 110 clicks, “kpop tanzen münchen” at 5.47, “breakdance kinder münchen” at 8.39 and “dancehall münchen” at 7.08.
Performance is uneven, which is useful planning evidence. “Bauchtanz münchen” averaged position 16.82, “kindertanz münchen” 20.71 and “kinderballett münchen” 29.06. Those categories are opportunities, not successes disguised by a site-wide average.
German local intent shaped the language and structure.
Course names mix English, German and genre-specific terminology. The architecture accommodates those real search habits instead of forcing a single vocabulary. Physical-business and class information is expressed in structured, consistent page elements designed to support LocalBusiness and scheduled-course interpretation.
The query export contains the maximum 1,000 rows. That is evidence of at least 1,000 surfaced query rows, not a complete ranked-keyword count: Search Console exports are capped and anonymise some queries.
What I would change now
- Instrument analytics and Search Console before launch, not with the launch.
- Strengthen course categories with visibility but weak positions, especially Bauchtanz, Kindertanz and ballet.
- Improve the 50,531-impression homepage’s 2.12% CTR without conflating page-table impressions with property totals.
- Audit GA4 event definitions and consent behaviour before interpreting key events or user counts.
See the governance model behind the operating workflow on the approach page.
Visit kayjays.de ↗Sources and limits
Sources: GA4 standard Reports for 1 January–24 August 2025 compared with the same dates in 2026; Google Search Console Web Performance export for 25 April 2025–24 August 2026.
Limits: No pre-launch baseline exists. GA4 Explore only retains event-level data from 6 September 2025 in this property, so year-over-year figures come from standard Reports. The Search Console query table is capped at 1,000 rows. Key-event claims are withheld because definitions may have changed; user counts are withheld because new users exceed total users in the retained period.