A private blog network is a set of websites controlled by one party, existing primarily to sell links. The individual sites look independent. They have domains, designs, article archives and often respectable authority scores. They differ from real publications in one respect: nobody reads them, and nobody was ever meant to.
The modern version is more sophisticated than the 2014 version. Sites are hosted on different providers, use different themes, publish real-ish content on a schedule, and have plausible-looking author pages. Some are expired domains that genuinely were publications once, bought for their existing backlink profile and repurposed.
Why they are a bad trade
Three reasons, only one of which is about search engines.
The asset is single-purpose. A network link produces no referral traffic, no brand exposure, no credibility with a buying committee, and no chance of being the source an answer engine cites. Its only value is a ranking effect.
That effect can be revoked retroactively. When a network is identified, links from it stop counting — and if the profile is bad enough to trigger a manual action, you pay to remove what you paid to acquire.
It is visible to humans. In enterprise sales, a competitor forwarding your prospect a screenshot of your backlink profile is a thing that happens. Rare, and expensive when it does.
Eleven signals
Ordered from cheapest to fake to most expensive. Start at the bottom of this list if you want to be efficient.
Easy to fake — weak signals on their own
- Generic name. "BusinessInsightsDaily", "TechTrendsHub". Real publications usually have a history behind the name.
- Stock imagery throughout. Cheap, and plenty of legitimate small publications do it too.
- No social presence. Easily manufactured, and some genuine trade titles genuinely have none.
- Recent domain registration. Checkable in seconds, though expired-domain networks defeat it entirely.
Moderately expensive to fake
- Topical spread. A site publishing on SaaS, dentistry, personal injury law, crypto and home improvement is not a publication; it is inventory. This one is genuinely hard to disguise, because narrowing the topics reduces the addressable market of buyers.
- Author pages with no external footprint. A named byline with a headshot and a bio, and no presence anywhere else on the internet. Real writers have LinkedIn profiles, other bylines and opinions.
- A "write for us" page with pricing. Publications have contributor guidelines. Inventory has rate cards.
- Outbound links to unrelated commercial categories. Check where else the site links. Casinos, essay-writing services and crypto exchanges in the outbound profile mean it will link to anyone.
Expensive to fake — strong signals
- High authority, negligible organic traffic. DR 55+ with under 500 monthly organic sessions. This is the signature pattern, because authority can be manufactured through link acquisition while readership cannot be manufactured at all.
- No traffic to any individual article. Even a modest real publication has a handful of pages that rank for something. A network site's pages typically rank for nothing.
- Footprint clustering across the network. The hardest one to hide and the most conclusive. Similar site architecture, comparable outbound patterns, overlapping sets of companies linked to, sometimes shared hosting ranges or identical plugin sets.
The three-check version
1. Authority versus traffic. Pull DR and monthly organic sessions. High-and-low is the pattern.
2. Topical spread. Read the last twenty article titles. Do they belong to one field?
3. Cross-site comparison. Take three domains from the same supplier's list side by side. Do they link to overlapping sets of companies? Do the author bios read like they were written by the same person?
Ten minutes, and it catches most networks.
The expired-domain variant
The harder case. Someone buys a domain that genuinely was a publication — a regional newspaper, a defunct trade magazine, a university department — and rebuilds it as inventory. It arrives with real historical backlinks, real domain age and a genuine archive.
Two things still give it away. First, the archive stops: there is a gap of months or years between the last genuine article and the beginning of the new content. Second, the topic changes across that gap — a former local news site now publishing exclusively about business software is describing what happened to it.
Check the Wayback Machine. Compare what the site published three years ago with what it publishes now.
What to do if you already have them
Do not panic and do not mass-disavow immediately. Sequence:
- Classify. Export every referring domain and score each on the eleven signals. You need a count before you need a plan.
- Assess proportion. Twelve network links in a profile of nine hundred is different from three hundred in a profile of nine hundred.
- Request removal first. Some operators will comply, particularly if there is an ongoing commercial relationship with your former agency.
- Disavow what remains, at domain level, if the proportion is material. This is a real tool with real downside if used carelessly.
- Pause acquisition while the profile contracts. Adding new links on top of a shrinking profile creates its own distinctive shape.
How to avoid buying them
One question, asked of any supplier: "What is your minimum verified organic traffic on a referring domain, and how do you verify it?"
A number sourced from a third-party index is the answer you want. A domain-rating threshold instead tells you the filter being used — and a DR filter selects for exactly the inventory this article describes, because DR is the thing networks are built to optimise.
Every network site that has ever been sold to anyone passed a DR filter. That is what they are for.
The second question, which is blunter and works: "Do you own, part-own or have a standing arrangement with any of the sites you place on?" The answer should be an unambiguous no, without a distinction drawn between owning and arranging.