magsrv.com: 61 sites and no banner on any of them

Zero of 61 carried a consent tool

The count is not low, it is nil. The list of consent tools seen alongside magsrv.com is empty, and the number of its sites with a detectable banner is 0.

That matters for how the rest of these numbers read. Elsewhere in this index, a company contacted before the banner could be answered raises a question about sequencing — did something run ahead of a choice a visitor was being offered. Here there is no choice on offer. All 61 sites had at least one company contacted before a consent choice could have been recorded, which is 100 percent — no company in this index has a higher share, and two others, Osano and bkcdn.net, also reach 100 — but the mechanism is the plainest one available: nothing was ever asked.

The pages are small and quick

The median site here contacted 7 separate companies across 33 requests. Among the 88 companies in this index, 20 sit lower on companies contacted and 23 lower on requests — the lower half of the table on both, and a shared position rather than a distinct one: thirteen other companies report the same median of 7 companies, and one other the same 33 requests. Median first contact was 398 ms, rank 86 of 88 where rank 1 is the longest wait; only two entries in this index sit on pages whose median first contact came sooner. The earliest single first contact recorded was 64 ms.

The median count of companies arriving before the window closes is 6, and the median arriving after it closes is 0. These pages open their connections at the start and then get on with it.

Why the tracker count reads 1

The median site here has 1 company classified as tracking, against 7 companies contacted — and a median of 1 is what 37 other companies in this index report as well, with only two of the 88 sitting lower. That gap is a property of the catalogue, not of the page.

A hostname can only be classified once someone has catalogued it, and this neighbourhood is largely uncatalogued — magsrv.com itself included, which is why this page has no category to report. On a page assembled from unfamiliar advertising hosts, the tracking count is a floor. Reading it as "only one tracker here" would be reading the limits of a vendor list as a fact about the site.

The neighbourhood

Google was present on 47 of the 61 sites, 77 percent. bkcdn.net on 38, 62.3 percent. pemsrv.com on 22, 36.1 percent. Cloudflare on 18 and Cloudflare Analytics on 8. Then a tail of hostnames most readers will not recognise: tsyndicate.com on 7, tcdwm.com on 6, orbsrv.com on 6 and agego.com on 6. Google Fonts and APIs appear on 7.

bkcdn.net at 62.3 percent is the figure worth noting. Most entries in this index have a co-presence list headed by a household name and then a long thin tail. This one has a second hostname on nearly two thirds of the same pages. Rates that high usually mean components delivered as a unit rather than chosen separately, but the crawl records only that they were on the same pages.

The sites themselves

sxyprn.com is the heaviest example: 20 companies contacted, all 20 inside the window, first contact 234 ms. tokyomotion.net contacted 18 with 7 early, first contact 1450 ms — the one slow starter in the set. pimpbunny.com: 12 companies, 11 early, at 187 ms. pornone.com: 11 companies, 6 early, at 601 ms. pornhat.com: 8 and 8, at 262 ms. imagefap.com and youjizz.com: 6 and 6 each, at 254 ms and 193 ms. jable.tv: 5 and 5, at 277 ms. erome.com is the lightest at 3 companies, all 3 early.

With one exception the pattern holds across the whole set: a handful of companies, all of them contacted in the first half-second, and no interface anywhere that asks about it.

Seen on the same pages

Companies most often contacted by the same sites. Co-presence only — the crawl records what arrived, not who called whom.

• GoogleTogether on 47 of these sites
• bkcdn.netTogether on 38 of these sites
• CloudflareTogether on 18 of these sites
• Cloudflare AnalyticsTogether on 8 of these sites
• Google Fonts / APIsTogether on 7 of these sites