Between October 1 and October 6, 2026, 16 of the 18 .com sites we measured each gained between 294 and 399 new referring domains. For the median site in that group, six days brought about 24 times what it gained in an average month from January to August 2026, and September had already brought about 12 times. Most of it is one template: a page listing 1,000 .com domain names in alphabetical order, copied onto hundreds of throwaway domains. None of the 7 non-.com sites in the same sample gained more than 6.
As of October 6, 2026. Data from the DataForSEO Backlinks API, pulled October 6. Every number is recomputed from the saved raw files by two scripts, and the chart data is free to download.
Two of the sites are ours: sceptermarketing.com and ecomcircles.com. We found this while measuring where links really come from, and our own reports suddenly showed a growth spurt we had not earned.
The numbers at a glance
| What we checked | Result |
|---|---|
| .com sites that gained 294 to 399 new referring domains, October 1-6 | 16 of 18 |
| Non-.com sites that gained more than 6 in the same days | 0 of 7 |
| New referring domains, October 1-6, across those 16 sites | 5,474 (about 10.5x their combined average month, January to August 2026) |
| New referring domains in September across the same 16 | 2,821 (151 to 230 each; median about 12x each site's average month) |
| Links first seen on or after October 3, in each site's top 200 referring domains | up to 191 of 200 (calebulku.com) |
| Share of those new rows that are the one list template | 1,747 of 1,829 (95.5%) |
| Outbound links on each of those 1,747 pages | exactly 1,000 |
| Those pages hosted on .website, .store, .site, .shop, .online or .space domains | 1,688 of 1,747 (96.6%) |
What happened, in plain English
Starting in September and accelerating in its last week, DataForSEO's crawler found a network of cheap domains serving pages with addresses like example.website/dir/seo-link-building-experts-183786. Each page is one "page" of a giant numbered directory. Page 183,786 lists entries #183,785,001 to #183,786,000, and every entry is a .com domain name with a sales-style anchor wrapped around it, for example "Premium Authority Link Building for sceptermarketing.com Businesses and Websites."
The list is alphabetical. On the copy we saved, sceptermarketing.com is entry #183,785,181, between other domains that start with "scepter." ecomcircles.com sits on page 61,672 between look-alike names, and calebulku.com sits on page 32,515. Nobody picked those neighbors; they simply sort next to ours.
In our data, each hit site appears on exactly one page path, repeated on every host that carries it, so it gets one new referring domain per host. So every hit site gained a similar count regardless of size. A site with 534 referring domains and one with 5,617 each gained 300 to 400.
Each page also advertises an "SEO tool" with a sign-up form, and all 1,000 links are plain followed links (no nofollow).
It did not start in October. Our earlier pull, limited to links first seen before September 25, already held the template 129 times across 12 of these sites, first seen September 10 to 24. The monthly counts show the rise was well under way before October: the 16 hit sites gained 151 to 230 new referring domains each in September, a median of about 12 times their average month, while the other sites stayed near normal. For five hit sites where we hold every link first seen before September 25, roughly 135 to 148 referring domains each were first seen between September 25 and 30.
Our top-200 samples cannot show those days. Among the lowest-ranked links, DataForSEO returns the newest first, so a sample that reaches them fills with the latest days: the oldest template row in the calebulku.com, ecomcircles.com and sceptermarketing.com samples is dated October 4. We have not confirmed that the September 25-30 links are the same template, though they match it in every way we can see (.com only, the same 16 sites, similar counts).
It does not look aimed at anyone. A list that runs past 183 million .com entries does not target a particular agency. The 16 hit sites include SEO educators, agencies, link vendors, software companies and our own two. None of this says anything about how these sites build their own links.
Why no non-.com sites? Every page we saved lists only .com names (1,000 of 1,000 on all three copies). Our sample has only 7 non-.com sites, so treat that as an observation, not a rule. Two .com sites were not hit either: one with 2 referring domains and one local business with 194, so it is not every .com.
Chart 1: new referring domains, October 1-6, 2026
.com siteNot .comIts average month, January-August 2026
Our two sites and the five creators' sites are named; the rest are Site A to Site R, the same letter in every data file (Site Q, a .com with 2 referring domains, is below this chart's cut and appears only in chart-share.csv). Data: chart-oct-new-rd.csv.
Chart data as a table
| Site | .com | Referring domains now | Average new per month, Jan-Aug 2026 | New, Oct 1-6, 2026 |
|---|---|---|---|---|
| Site A (.com) | yes | 1291 | 36.5 | 399 |
| Site B (.com) | yes | 857 | 7.4 | 396 |
| Site C (.com) | yes | 886 | 15.8 | 379 |
| juliangoldie.com | yes | 1402 | 55.5 | 377 |
| Site D (.com) | yes | 1580 | 43.5 | 362 |
| diggitymarketing.com | yes | 5617 | 168.4 | 361 |
| Site E (.com) | yes | 636 | 14.0 | 347 |
| Site F (.com) | yes | 987 | 50.4 | 342 |
| jackychou.com | yes | 554 | 5.5 | 339 |
| ecomcircles.com | yes | 652 | 8.3 | 327 |
| calebulku.com | yes | 534 | 5.4 | 318 |
| Site G (.com) | yes | 2355 | 61.1 | 317 |
| Site H (.com) | yes | 639 | 7.5 | 315 |
| charlesfloate.com | yes | 722 | 10.4 | 305 |
| sceptermarketing.com | yes | 977 | 12.6 | 296 |
| Site I (.com) | yes | 729 | 20.8 | 294 |
| Site J (.com) | yes | 194 | 11.1 | 7 |
| Site K (not .com) | no | 1163 | 48.6 | 6 |
| Site L (not .com) | no | 986 | 53.1 | 6 |
| Site M (not .com) | no | 124 | 12.9 | 4 |
| Site N (not .com) | no | 43 | 2.5 | 3 |
| Site O (not .com) | no | 61 | 4.9 | 2 |
| Site P (not .com) | no | 44 | 3.8 | 1 |
| Site R (not .com) | no | 40 | 3.4 | 0 |
Chart 2: September and early October against a normal month
Average month, January-August 2026September 2026October 1-6, 2026
Same 24 sites and labels as Chart 1. Data: chart-monthly.csv.
Chart data as a table
| Site | .com | Average month, Jan-Aug 2026 | September 2026 | October 1-6, 2026 |
|---|---|---|---|---|
| Site A (.com) | yes | 36.5 | 173 | 399 |
| Site B (.com) | yes | 7.4 | 164 | 396 |
| Site C (.com) | yes | 15.8 | 178 | 379 |
| juliangoldie.com | yes | 55.5 | 175 | 377 |
| Site D (.com) | yes | 43.5 | 185 | 362 |
| diggitymarketing.com | yes | 168.4 | 230 | 361 |
| Site E (.com) | yes | 14.0 | 179 | 347 |
| Site F (.com) | yes | 50.4 | 185 | 342 |
| jackychou.com | yes | 5.5 | 151 | 339 |
| ecomcircles.com | yes | 8.3 | 160 | 327 |
| calebulku.com | yes | 5.4 | 159 | 318 |
| Site G (.com) | yes | 61.1 | 203 | 317 |
| Site H (.com) | yes | 7.5 | 166 | 315 |
| charlesfloate.com | yes | 10.4 | 178 | 305 |
| sceptermarketing.com | yes | 12.6 | 165 | 296 |
| Site I (.com) | yes | 20.8 | 170 | 294 |
| Site J (.com) | yes | 11.1 | 14 | 7 |
| Site K (not .com) | no | 48.6 | 33 | 6 |
| Site L (not .com) | no | 53.1 | 39 | 6 |
| Site M (not .com) | no | 12.9 | 15 | 4 |
| Site N (not .com) | no | 2.5 | 14 | 3 |
| Site O (not .com) | no | 4.9 | 8 | 2 |
| Site P (not .com) | no | 3.8 | 6 | 1 |
| Site R (not .com) | no | 3.4 | 9 | 0 |
The template families, old and new
The October list is the loudest machine-made page in these profiles, not the only one. Grouped by URL shape, title and outbound-link count, with the earliest date DataForSEO saw each in our data:
| Family | What it looks like | First seen in our data | Outbound links per page |
|---|---|---|---|
| Alphabetical keyword list (the October blast) | /dir/<keyword-slug>-<page number>, "Page 183,786" in the title | 2026-09-10 (surge from October 4) | 1,000 |
| PBN-seller ad pages | /all/<n>/<n>.html, emoji title selling "Premium PBN & Link Building" | 2026-07-03 | 3,000 |
| "Website list" directories | website-list-<n> or most-visited-website-list-<n> | 2023-07-26 | about 5,000 |
| URL-stats scrapers | /share/<n> and /report/<n>, titles like "URL Shared" | 2024-10-20 | about 3,500 |
| Aged-domain sales lists | /<32-character hex>-l/ or /list-<hex> | 2025-02-06 | 9,000 or more |
Live examples, fetched October 6, 2026 (we saved the HTML of each):
- Our own listing:
<host>.store/dir/seo-link-building-experts-183786: 1,000 followed links, ours at position 181. - ecomcircles.com:
<host>.site/dir/relevant-domain-backlinks-61672: 1,000 followed links, ours at position 329. - calebulku.com:
<host>.info/dir/manual-seo-backlinks-32515: 1,000 followed links, position 140. - A PBN-seller ad page (
/all/323/22.htmlon an unrelated domain) now redirects to a link seller's landing page with Telegram and WhatsApp contact links. - A "website list" page: 5,000 links, all marked
noopener, headed "30 GUEST POSTS & Parasite SEO Package."
The older families were already there. When we classified, link by link, the links first seen before September 25 for nine of these sites, 12% to 78% of each slice was already this kind of machine-made page. Across all 15 slices we classified, the range was 0.5% to 92.3%. The October blast is a spike on top of a steady background.
What it does to your reports
If you report referring-domain growth to a client or a boss this month, you are probably reporting the blast.
Our own numbers show it. sceptermarketing.com gained 101 referring domains and lost 378 from January to August 2026, a net loss of 277. Then it gained 165 in September and 296 in the first six days of October. ecomcircles.com gained 327 in those six days, which is half of its 652 referring domains today. Neither number is link building we did: the October jump is mostly this list, and the September jump has the same .com-only pattern.
The usual top-200 view can hide it, too. Backlink tools often sort by link strength, and all 1,747 template links in our samples carry a DataForSEO link rank of 0. diggitymarketing.com gained 361 new referring domains from October 1 to 6, but none of them appear in its top 200, because it has more than 200 stronger links. Smaller sites show the opposite: links first seen on or after October 3 took 191 of calebulku.com's top 200, 189 of jackychou.com's and 169 of charlesfloate.com's, nearly all of them this template. So the same event can look like nothing or like a takeover, depending only on the size of the profile and the sort order.
What a site owner should do
- Mark the dates. Put a note on late September through early October 2026 in every link report and dashboard, so nobody reads the jump as growth, or misreads a drop later if these hosts disappear.
- Filter it out of your counts. The fingerprint is easy: a
/dir/URL ending in a number, exactly 1,000 outbound links, a title ending in "Page <number>," and a host on .website, .store, .site, .shop, .online or .space. Report link growth with and without it. - Check the Manual actions report in Search Console. Google reports manual actions there, and a manual action is the condition its disavow page names. If the report is empty, you have none.
- Keep measuring rankings and traffic for the pages you care about. If they hold steady through October, that is your answer.
What a site owner should not do
- Do not disavow these by reflex. Google's own disavow page says "In most cases, Google can assess which links to trust without additional guidance, so most sites will not need to use this tool." It says to disavow only if you have "a considerable number of spammy, artificial, or low-quality links pointing to your site, AND the links have caused a manual action, or likely will cause a manual action, on your site." The same page warns the tool "can potentially harm your site's performance" if used incorrectly, and says "Google works very hard to make sure that actions on third-party sites do not negatively affect a website."
- Do not buy a "toxic link cleanup." Charles Floate, who runs link-selling businesses himself, was asked on a June 30, 2025 webinar whether to disavow the spam everyone gets. His answer: "No... Google ignores those links." Earlier in the same webinar he said disavow does work against real negative SEO, which he says can still hurt a page, but that is a different thing from an alphabetical list numbered past 183 million entries.
- Do not copy it. Google's spam policies (last updated August 28, 2026) define link spam as "creating links to or from a site primarily for the purpose of manipulating search rankings," and list "using automated programs or services to create links to your site" and "low-quality directory or bookmark site links" as examples. These pages fit both. Paying for links like this puts you on the wrong side of that policy; merely being listed is what the disavow guidance above covers.
Method and limits
- Sites. We pulled 26 domains on October 6, 2026; 25 returned data (one was a mistyped domain). They are not a random sample: they are the sites of SEO educators we were studying, related businesses, and our own two sites. Sites other than ours and the creators' are described in aggregate.
- Data. DataForSEO Backlinks API: a site summary, monthly new and lost referring domains since January 2025, and the top 200 referring domains by DataForSEO rank with one backlink each. "First seen" is when DataForSEO's crawler found the link, not when the page was made. When a top-200 sample reaches a site's lowest-ranked links, DataForSEO returns those newest first, so the sample shows the latest days and undercounts earlier ones; that is why the timing above comes from the monthly counts.
- Partial days. Pulls ran from 00:01 to 16:54 UTC on October 6, so later pulls saw more of October 5 and 6. For 11 of the 16 hit sites the monthly counts were pulled before 02:00 UTC on October 6, so their "October 1-6" is effectively October 1-5. The October numbers are a snapshot, not a final count.
- Ordinary growth is mixed in. October's new referring domains include normal links. The median hit site gained about 15 referring domains in an average month this year, so about 3 in six days; the largest, diggitymarketing.com, about 168 a month.
- Template attribution. For 12 of the 16 hit sites, the template is at least a quarter of the top-200 sample, and juliangoldie.com shows it in 17 rows. For the other 3, the top-200 sample does not reach these rank-0 links, so the attribution rests on the matching volume and timing.
- The .com pattern is 16 of 18 against 0 of 7. That is a small sample.
- Not measured. We did not measure any ranking or traffic effect, and we did not identify who runs the network.
Download the data
- chart-oct-new-rd.csv: new referring domains October 1-6 and the January-August monthly average, per site.
- chart-monthly.csv: new referring domains per site: January-August average month, September, and October 1-6.
- chart-daily.csv: rows first seen per day in our top-200 samples (newest-first, so read it as the shape of the sample, not of the event).
- chart-share.csv: share of each site's top 200 first seen on or after October 3.
