I built it for people buying guest posts. It turned out to matter more for the people writing them.
An invitation arrives. Someone would like you to contribute a piece to their site. The audience sounds right, the editor sounds pleasant, and the subject is one you can write well.
How do you decide?
Most of us decide on feel. We open the homepage, read a couple of articles, look at the design, and say yes. That takes five minutes and it is mostly a judgement about whether the place looks serious.
There is one question that five minutes does not answer, and it is the one that matters most.
Is this site still publishing?
The site that had been quiet for fourteen months
I built a tool called Site Dossier to answer a different question. I sell guest post campaigns, and I wanted to check a site before paying for a placement on it. You paste in a website, choose your subject, and it reads the site’s most recent articles and tells you what it actually publishes.
The first real site I ran it on was a developer community. Fifty-seven recent articles about backend work, machine learning and React. Plenty of names in the author list. A homepage that looked completely alive.
Then I read the dates.
All fifty-seven of those articles went out in the same eight days of July 2025. After that, nothing. Fourteen months of silence.
Nothing on the homepage tells you this. The articles sit there looking current. The newsletter signup still works. If you had been invited to write for that site last week, you would have spent a day on a piece that nobody was coming to read, on a site where your name would sit at the top of a pile that stopped growing before you arrived.
That is the part I had not thought about when I built the thing. I was worried about wasted money. For a writer the loss is worse, because you cannot get the week back.
Three other things worth knowing before you say yes
The publishing date is the big one. There are three others that take seconds to check and that most of us never look at.
Who actually writes there. On that developer site, 43 different people had written the 57 articles, and 32 of them had written exactly one. That is a particular kind of place. It is not a publication with writers, it is a door that anyone can walk through once. Neither is automatically bad, but they are different things, and your name sits differently in each.
The opposite pattern is worth knowing too. I ran the tool on a small blog about birds and every article was credited to the site’s own brand name rather than to a person. No individual writer to look up, no editor to build a relationship with. Again, not a crime. Just something you want to know before you offer them a week of your work.
What the site links out to, and in what words. This one surprised me. On the bird blog, the outbound links went to veterinary references and conservation bodies, and the anchor text read like a person describing a source. Things like “abnormal droppings in birds” and “avian nutrition guidance published by Lafeber”.
That is what a site looks like when nobody is being paid for the links. Compare that to a site where the same few commercial destinations appear again and again with neat keyword phrases attached, and you can see what the place is for without anybody telling you.
What it actually publishes, rather than what it says it does. Every about page describes a mission. The article list describes a reality. Where those two disagree, the article list is the one to believe.
None of this is new thinking. It is the same set of questions I work through before paying for a placement, and I wrote the longer version of it up as how to vet guest post sites before you pay. The checks are identical whether money is changing hands or only your afternoon is.
The part where I tell you it got things wrong
I would rather you heard this from me.
That first verdict on the developer community was not “a site that has stopped publishing”. It was “this looks like a link shop, not a publication”. The tool printed that in large letters, with five reasons underneath.
I read the five reasons. Three of them did not hold up.
It said the subject matter was scattered. The breakdown was software 25 percent, machine learning 23, technology 12, gaming 5. Those four are one subject. My own category list had cut a focused site into pieces and then complained that the pieces were small.
It raised a risk flag over two technical articles about Ethereum, while the same line admitted that none of the site’s outbound links went anywhere near crypto.
It counted links to GitHub Pages and Vercel as commercial destinations. On a developer site those are people pointing at their own side projects.
So the verdict was harsh, the reasoning was partly wrong, and the one fact that mattered was buried near the bottom.
I spent two days fixing it. Related categories are now grouped before the scattered test runs. The risk flag now needs actual commercial links, not just a topic. Project hosting is out of the commercial count. And the dormant site warning now fires on the site’s own publishing dates and sits at the top of the report, where it belongs.
The same site now reads: this looks like a publication, but it has stopped publishing. Nothing for fourteen months. Most recent article 21 July 2025.
I only caught any of this because I tested a site where I already knew the answer. That is the part I would pass on to anyone, whatever tool they are using. Run it on three sites you know well before you trust it on one you do not.
What it will not do for you
It will not tell you whether the editor is good to work with, whether your piece will be read, or whether the audience is yours.
It does not show Domain Authority, Domain Rating or traffic estimates. Every other free tool shows those, and none of them tell you what a site publishes.
It reads recent articles rather than the full archive, so it describes where a site is now, not where it has been.
And it came out of two days of testing on two sites. There are almost certainly failure modes I have not found yet.
Before you say yes
You do not need a tool for most of this. You need a habit.
Open the site’s most recent article and look at the date. Then look at the date on the one before it. If there is a long gap, you have learned the most important thing in about fifteen seconds.
Then read three article titles and ask whether they are really about the same subject, or whether the site is publishing whatever walks in.
Then look at where those articles link out to, and in what words.
If you would like all of that in one place, Site Dossier does it in about twenty seconds. It is free and needs no signup. I built it, so weigh my opinion of it accordingly.
Your name on a piece of writing is not a small thing to hand over. It is worth twenty seconds to find out where it is going to sit.
Muhammad Kashif Khatri runs SEO26MASTER, builds Site Dossier, and is an editor at ILLUMINATION and four other Medium publications.
Leave a Reply