I open maybe a hundred unfamiliar sites a year for research. Most of them are fine. A small number are link farms dressed up as blogs, and they share enough fingerprints that you can sort them in under five minutes if you know what to look for. This post is the checklist I now use, walked through one real example so you can see each step in context. The example is yonochomp.com, but the shape applies to every site like it.
Why I bother at all is that the cost of mistaking a link farm for a real source is higher than people realize. A bad citation undermines an article. A bad backlink wastes budget. A bad guest post on a quiet domain can sit there for two years, contributing nothing to your SEO and quietly rotting your link profile. None of those failures feel dramatic on the day. They are the kind of slow drain you only notice when you look back at six months of work and wonder why nothing moved.
The five minute test is built from things you can see on the site itself. No SEO tool required. No third party audit. Just a browser and the discipline to look at what is actually there before you trust it.
The menu is the first thing to check
Every blog has a menu. The shape of the menu tells you whether the site is built around a beat or built around a sales pitch. A blog with one beat and three or four tabs in the menu is a real publication. A blog with sixteen tabs covering everything from Automotive to Travel is a template dressed up as a magazine.
Yonochomp.com runs the sixteen tab version. Eleven of those tabs hold no content at all. Click Automotive and you get a blank archive. Click Education and the same. The five tabs that do have posts jump between jade earrings, a teen patti strategy piece, a Lotto24 explainer, and a writeup about a small town in the Netherlands. There is no editorial throughline. There is no reason a single person would cover all of those subjects. There is a reason an operator running a template would.
A few quick ways to run this check yourself:
- Open the menu, count the tabs, click five of them. If three or more return nothing, the breadth is decoration.
- Skim the five or six posts that do exist. If the subjects feel unrelated to each other, you are looking at a content mill, not a magazine.
- Note the read time on each post. Real writers vary read time based on depth. A blog that pins every post to a fixed three or four minute window is running a word count target, not a subject.
- Look at the categories in the URL slugs. If every post has a different category in its slug, the template is being used to manufacture breadth. Real blogs have a few core categories that show up again and again.
The menu test takes about thirty seconds. It is the single highest yield check on the list.
The byline and the URL slug should match
Every post on yonochomp.com is signed by a single display author. The author URL slug, however, points to a different name. That mismatch is not a typo. It is the kind of detail that only shows up when nobody is paying attention to the byline, or when the byline is set to look like a person but the underlying account belongs to someone else.
This check is easy to run and almost no one runs it. Click on the byline of any post on an unfamiliar site. Look at the URL it sends you to. Compare the slug in the URL to the display name on the post. If they do not match, you are looking at a borrowed credential, a renamed account, or a deliberate mismatch designed to obscure who actually wrote the thing.
A few related tells worth pairing with this check:
- One byline across every post, with no author archive page. Real writers who blog for a year usually have a small archive of their own. A solo byline with no archive is a flag.
- No About page, no Contact page, no policy page anywhere. The site should at least tell you who it is and how to reach someone. If none of those exist, the operator is hiding.
- The only contact route is a scrambled email link. A scrambled address (the kind where you have to type the letters to assemble the address) is a tradeoff. It stops bots. It also stops readers. If the operator has chosen to stop readers, that is the operator’s choice and yours to note.
The byline test takes about fifteen seconds. Combined with the menu test, you are already inside the five minute window.
The publishing history is the third tell
A real publication has a publishing rhythm. Some weeks have two posts, some weeks have zero, and there is a slow buildup as the writers find their voice. A link farm has a different shape. Posts land in a single burst, then the site goes quiet, then a single piece shows up months later to make the site look alive.
Yonochomp.com follows that exact pattern. Dozens of posts landed in a six to eight week window in mid 2023, spread across every tab. Then nothing. Then a single piece in 2025, almost two years later, and that was the last update. That is the signature of a batch publication, not a real blog. Someone bought a package of articles, the site absorbed them all in a single window, and the operator walked away until the next sale.
A few details that make the pattern easier to spot:
- The archive widget on the sidebar usually lists only a few months. If the visible archive is a six to eight week window in one year and a single post in another, you are looking at a burst.
- The subjects in the burst are unrelated. A real writer on a single beat does not jump from fashion to gaming to finance in eight weeks. A content package does.
- A long silence followed by a single new post is a strong tell. Real blogs that go quiet usually do not come back. Sites that come back with one piece are showing the operator is still around and the package model is still in use.
- The newest post and the burst posts share the same voice. When the single 2025 piece reads exactly like the 2023 pieces, the same author template is still in use. That is a hint the model is alive, not a one off.
The publishing history test takes about a minute. You are now at two minutes total.
The numbers and the content rarely agree
The metrics on link farms are inflated. That is the whole reason people buy links on them. A domain with a Moz Domain Authority (a score Moz assigns to predict how well a site will rank in search, on a 0-100 scale) of 54 looks like a strong placement. A domain with an Ahrefs Domain Rating (a similar 0-100 score from Ahrefs, weighted by backlink quality) of 25 is mid market. When both numbers exist on the same site and they are far apart, one of them is more honest than the other. The gap is the signal.
Yonochomp.com has a DA of 54 and a DR of 25. The 29 point spread is wider than you usually see on a real publication. It suggests the site has historical link equity from older placements that aged well, while the current backlink profile is much thinner. One number is real, one is residual. The operator is selling the residual number to anyone who does not check the second metric.
A few other small signals that often show up on the same sites:
- Images load from a separate CDN (content delivery network), not the WordPress media library. A site that hosts assets on cdn.somethingelse.com is running infrastructure for a network of blogs, not a single site. It is not proof of anything on its own, but it is a data point.
- The site runs a common theme on the default settings. A site using a free or low cost theme with the demo content still half in place is running a template, not curating a design.
The metrics test takes about a minute. You are now at three minutes.
The guest post banner is the business model
If you scroll past the posts on yonochomp.com, there is a guest post banner in the sidebar. That banner is the clearest tell of what the site is for. The site exists to sell links. The 2023 burst of content is the proof of concept. The 2025 piece is evidence the model is still occasionally active. Buying a guest post here gets you a backlink on a domain with middling authority and no real readership, from an author whose name does not match the byline, on a site that does not publish consistently.
If you are reading this and considering a link buy, the same arithmetic applies across the market. Cheap sites have worse metrics. Expensive sites have real traffic and editorial standards. The middle tier, where sites like yonochomp.com sit, is where most link buyers end up and where most of them also get disappointed. You are paying for a number on a dashboard, not an audience that will ever click through.
The business model check is the last step. At this point you have either decided the site is fine for what you needed, or you have decided to walk away. The five minute budget is up.
Trade-offs
The checklist above trades depth for speed. You will catch obvious link farms in under five minutes, and you will miss the well-run ones that have invested in design, varied bylines, and a slower publishing rhythm. A five minute test is a screening tool, not a guarantee. For a high stakes decision (a link buy on a $1,000 budget, a citation in a published article, a partnership mention), the same checklist should take you fifteen minutes and include a real backlink profile check on a paid SEO tool.
Running the checklist on every site you visit is also tedious. I do not run it on sites I already trust. I run it on sites that show up in search results, sites that pitch me on guest posts, and sites that someone I trust has linked to without explanation. Discipline is to run it on the unfamiliar, not on everything.
Finally, the checklist is biased against new sites. A real blog that launched last month will not have a long publishing history. The byline might still be settling. Metrics will be low. Some of those new sites are link farms in waiting and some are real publications that will grow into something worth citing. The checklist cannot tell the difference on a five minute budget. Run it as a screen, not as a final verdict.
Bottom line
If a site fails the menu test, the byline test, the publishing history test, the metrics test, or the business model test, walk away. If it fails two of them, you can be sure. If it fails three, you can cite this post when you explain why. The five minute test is not fancy, it is not deep, and it will not catch every bad site on the web. It will catch the link farms that pretend to be blogs, and that is enough for most of the work I do.