Is PubGalaxy legit? What its own published records show
PubGalaxy is a substantial network with a genuine specialism, and it also has one of the lowest confirmation rates we measure. Both facts are worth understanding, because the second one is more often misread than it is understood.
1 August 2026 · 6 min read
The question gets asked because PubGalaxy is large enough that publishers encounter it and obscure enough that there is little written about it. So it arrives without the reputational context that surrounds the household names.
We can answer it from the public record — with a caveat about what the record can and cannot establish.
First: it is a real network with a real specialism
Nothing in our data suggests otherwise, and the size alone rules out the more cynical readings. PubGalaxy leads the gaming category in our index — a genuinely difficult category that the large consumer networks have largely ceded — and holds meaningful positions in technology and news.
| Category | Sites |
|---|---|
| Gaming | 1,162 |
| Other | 269 |
| News & Politics | 240 |
| Technology | 148 |
| Entertainment | 146 |
| Sports | 96 |
That profile — gaming, technology, news, international, programmatic — is coherent. It is a network selling the kind of inventory that the domestic lifestyle networks price poorly, and it has built a substantial book doing it. The gaming category breakdown covers why that inventory needs different partners.
The confirmation rate, and how to read it
Every relationship we index is tested against both sides: the network’s sellers.json and the publisher’s own ads.txt and homepage. 1,271 of PubGalaxy’s 2,599 published relationships are one-sided — the network lists the site, the site does not point back.
There are four explanations for that, and only one of them is a problem:
| Explanation | Is it a concern? |
|---|---|
| Reselling | No. A network selling through upstream partners may never appear under its own name in a publisher’s file. The relationship is real and structurally invisible from that side. |
| Shared seats | No. Several publishers placed under one account, so the id does not match the one published for that specific domain. |
| Unreadable publisher file | No. Some sites serve no ads.txt, or serve it in a way crawlers cannot read. Absence of evidence rather than evidence of absence. |
| Stale entries | Yes. The publisher left and the network never pruned the file. This is the common one, and it is what makes a published roster misleading. |
Our data cannot tell you which applies to any individual entry, and it would be dishonest to present the rate as a verdict. What it does tell you is that PubGalaxy’s published roster should not be taken at face value — roughly half of it is unverifiable from outside, and if you are evaluating the network on the strength of its publisher list, half that list is a claim rather than a demonstrated relationship.
What to do with that
Ask directly. Three questions, all reasonable, and the precision of the answers is informative regardless of the numbers: are the publishers you have shown me currently live, and as of what date? Will my site appear in your sellers.json under my own domain? How quickly do you remove publishers who leave?
The last question is the one this data actually raises. A network that prunes promptly has a high confirmation rate almost automatically. The method for checking any network yourself is in how to verify an ad network’s publisher claims, which also has the full confirmation table for every network we track.
Publishers it carries
These are the ones where both parties agree, which is the standard worth applying to any network’s claims. Its full indexed list, including the unconfirmed entries, is on PubGalaxy’s network page.
Who it suits
On the evidence, publishers whose inventory the domestic lifestyle networks handle poorly: gaming, technology, news, and sites with heavily international audiences. That is a real gap in the market and PubGalaxy is genuinely in it.
The caution is not about legitimacy but about diligence. This is a network where checking the specific publishers it cites to you — rather than trusting the roster — is worth the twenty minutes, and where asking about its pruning practices is a fair question rather than a hostile one.
Common questions about PubGalaxy
Is PubGalaxy a legitimate ad network?
Nothing in the public record suggests otherwise. It carries 2,599 indexed publishers, leads the gaming category in our index, and holds real positions in technology and news — a coherent specialism in inventory the large consumer networks price poorly. The one thing worth scrutiny is its published roster, roughly half of which is not confirmed by the publisher side.
Why is PubGalaxy's confirmation rate so low?
We cannot tell from outside which of four explanations applies to any given entry: reselling through upstream partners, shared account seats, publishers who serve no readable ads.txt, or stale listings for publishers who have left. Only the last is a genuine problem, and it is also the most common across the industry. The rate is a reason to ask the network directly rather than a verdict.
What kind of sites does PubGalaxy work with?
Its largest categories are Gaming, Other, News & Politics. That is the inventory profile the domestic lifestyle networks handle least well — internationally distributed, heavily ad-blocked, and outside the household-purchase categories that command the strongest display demand.
Should I check a network's sellers.json before signing?
Yes, and it takes about twenty minutes. Find the publishers the network cited to you in its own sellers.json, note the account ids, then check those publishers' ads.txt files for the same ids. Relationships evidenced from both sides are real; one-sided entries are claims. Checking five or six names tells you whether the roster is broadly reliable.
“Confirmed” means the publisher’s own records evidence the network. Unconfirmed relationships may still be real — see the four explanations above. Every figure on this page is read from our own index when the page is built, and refreshes on a weekly cycle — so the numbers move as the data does. Method and evidence rules: coverage & method.