30 Days of Search Visibility After a Domain Move — GSC, Naver and Cloudflare
Correction, 2026-10-07: The earlier version predicted automatic recovery after a fixed waiting period. These observations do not establish that explanation. Historical figures are retained; claims about recovery deadlines and hidden algorithms have been corrected.
Background and scope
This is an operating record following the move from taystudio.github.io to taystudios.com on May 9, 2026. The figures below were recorded in the original article and do not represent current Search Console status. Reporting windows and metric definitions differ between services; their figures should not simply be added or their ratios generalized.
30-day data
GSC (Google Search Console)
| Date | Clicks | Impressions | Avg position | CTR |
|---|---|---|---|---|
| 5/9 (migration) | 0 | 0 | — | — |
| 5/14 | 1 | 450 (peak) | 14 | 0.2% |
| 5/17 | 0 | 300 | 13 | 0% |
| 5/20 | 1 | 280 | 12 | 0.4% |
| 5/24 | 0 | 10 | 18 | 0% |
| 5/28 | 0 | 5 | 22 | 0% |
| 6/2 | 0 | ~0 | — | — |
It peaked on 5/14, dipped by 5/24, and then sat near zero. From the operator's seat, watching that graph, you can't help but wonder whether the site just died.
Bing Webmaster
Bing showed a similar pattern. There was residual data from the old domain through 5/12, and from 5/22 the sitemap was discovered with 124 URLs. Search impressions stayed at zero. This record does not establish how Bing evaluated the site or a waiting period.
Naver SearchAdvisor (registered 5/22)
This is where it flipped. The spike started right after I registered on 5/22, peaked on 5/27 at 7,600 impressions, and after 5/28 settled into a stable 2,000–3,000 per day. Over the 30 days the totals came to 26,000 impressions and 230 clicks.
Naver referrals were observed, but these tables do not establish a like-for-like ratio against Google or explain how either engine evaluated the site.
Cloudflare Web Analytics
I used Cloudflare to see total traffic across all channels. Visits averaged around 55 a day, and 79% of that was Naver/Daum referrers. Some chatgpt.com referrers also appeared. Visit statistics alone do not establish that every page was accessible or indexed by Google.
What these observations establish
A rise and fall in impressions does not establish trial exposure, a CTR-based penalty or a new-domain waiting period. Search impressions, ranking and indexing measure different things. Discovery through a sitemap is not the same as indexing.
“Sandbox” is an informal explanation sometimes used for weak visibility on new sites. This record does not show that a specific restriction applied or would expire on a schedule. Early Naver referrals also do not guarantee immediate indexing for other sites.
A chatgpt.com referrer may indicate a visit from that service. It does not establish model training or citations by Claude or Perplexity.
Checks instead of a recovery deadline
- Check that old URLs lead to their matching new URLs.
- Inspect response status, robots/noindex directives and canonical URLs.
- Distinguish the aggregate report date from the individual URL inspection date.
- Compare deployment time with Google's latest crawl time.
- If a recrawled page remains unindexed, review its functionality, accuracy and links from related pages.
Repeated crawl requests do not guarantee faster crawling, and a request does not guarantee indexing. Do not defer investigating errors until an assumed waiting period ends.
Comments