Experiment
The AS400System Hold-Steady Experiment
A live test of whether a legitimate family of specialist subdomains can keep expanding through an algorithmic visibility reset and regain strength as the network earns history, use, links, citations, and AI system attention. The AS400System.com family lost 85.8 percent of its Google impressions immediately after the August 2026 spam update. The architecture is staying.
If the AS400System.com subdomains have real, distinct jobs and keep adding useful, differentiated evidence for people and AI systems, the family can mature through Google's August 2026 ranking reclassification without dismantling the network. Google visibility may recover, but recovery is not the only success condition. The stronger test is whether total discovery across search, verified crawlers, human requests, AI citations, and commercial use continues to grow while each new subdomain remains specific enough to justify its existence.
What this test is watching.
This test watches the AS400System family after it almost disappeared from Google. We are checking whether the sites begin appearing again and whether clicks, links, AI mentions, and crawler visits return while the site structure stays in place.
How the test is structured.
AS400System.com launched as the central hardware property on July 20, 2026. By early September, its branch included twelve hosts and 781 URLs in public sitemaps: the root site, AI, FlashSystem, Pricing, and Power 5 through Power 12 generation properties. Each host was built around a different IBM Power or AS400 decision rather than a copied keyword variant.
Google rolled out a global spam update from August 18 through August 21. On August 22, every established property under as400system.com fell together. The family dropped from 9,278 weekly impressions to 1,322. Clicks fell from 60 to 1. The root site's average position moved from 50.5 to 66.4, and the number of root pages receiving impressions fell from 117 to 48.
The technical system remained intact. Search Console showed 683 indexed pages after the drop, no manual action, and no security issue. Googlebot requests increased from 938 to 984 across the comparison weeks, with no 5xx responses. AS400Software.com and AS400IBMSystems.com both grew during the same window, so the subject did not disappear. The reclassification stayed inside the AS400System.com family.
The intervention is deliberately simple: no rollback. Existing subdomains stay live. Existing network relationships stay visible. Planned subdomains may continue when they have a real specialist role. The campaign will not manufacture new hosts merely to force the test, but it will not let a ranking update veto a legitimate build either.
What has happened so far.
Baseline recorded September 3, 2026. The original Google comparison still stands: 9,278 impressions and 60 clicks from August 15 through 21, followed by 1,322 impressions and 1 click from August 22 through 28.
A later warehouse classification repair changed the crawler counts slightly. Recognized Googlebot requests were 935 before and 979 after, up 4.7 percent. Successful Googlebot content requests were 819 before and 847 after, up 3.4 percent. The old 938-to-984 receipt stays in the linked Field Note as the number available that day, with the correction attached.
September 11 interim ... still no recovery. The latest complete Search Console week, September 2 through 8, produced 705 impressions, 1 click, a weighted average position of 61.6, and 73 pages with impressions. That is 92.4 percent below the baseline impression count. Across the family, recognized crawler requests were almost flat against baseline, 3,430 versus 3,501, while successful crawler content requests fell from 2,691 to 2,133. Successful Googlebot content requests fell from 819 to 460.
The separation is still real, but it is more specific than the first read. Bot requests did not disappear with rankings. Successful content crawling also did not remain completely untouched.
The first formal 30-day checkpoint remains September 21. This interim does not replace it.
What would support or challenge it.
Status: active. This is not a recovery story yet. It is the point where the easy launch curve ended and the more useful experiment began.
The hypothesis weakens if the family remains below half of its baseline Google visibility after 90 days, new specialist hosts fail to earn independent crawl or index coverage, and non-Google discovery remains flat. It strengthens if search visibility returns while crawler and human activity keep expanding, or if the network creates measurable AI and commercial value even while Google remains constrained.
Google gets a vote. It does not get the only vote.
Keep Following the Tests
Move from this open thread back into the full experiment library.