The five-minute test that would have saved me eleven hours of directory submissions
Eleven hours of directory submissions produced two followed links. Not because the directories were fake, but because I never asked the one question that decides the outcome, and it takes five minutes to answer: Pick a
Eleven hours of directory submissions produced two followed links. Not because the directories were fake, but because I never asked the one question that decides the outcome, and it takes five minutes to answer:
Pick a game that is already listed on that site, open its page, and look for an anchor to the game's own domain.
That is the test. A directory that links out to the games it lists will link out to mine. A directory that embeds, mirrors, or wraps them will not — and no amount of accurate metadata on my submission changes that.
What the test actually reads
Open any listing page and there are exactly three facts that matter. They are the same three facts I check on my own placement later, which is why the script is one script used twice.
import re, urllib.request
UA = ("Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 "
"(KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36")
def fetch(url, timeout=20):
req = urllib.request.Request(url, headers={"User-Agent": UA})
with urllib.request.urlopen(req, timeout=timeout) as r:
return r.status, r.geturl(), r.read().decode("utf-8", "replace")
def anchors(html, domain):
out = []
for m in re.finditer(r"<a\b([^>]*)>", html, re.I):
href = re.search(r'href=["\']([^"\']+)', m.group(1), re.I)
if href and domain in href.group(1):
rel = re.search(r'rel=["\']([^"\']*)["\']', m.group(1), re.I)
out.append({"href": href.group(1), "rel": rel.group(1) if rel else ""})
return out
Feed it the domain of the game that is already listed — itch.io, zgames.io, whatever its play link points at — and you get one of four answers:
-
An anchor, no
nofollow→ submit. This site passes links. -
An anchor with
nofollow→ submit only if you want the traffic. It will not pass equity. - No anchor at all, domain mentioned in text → do not submit. You are giving them a page about you.
-
An anchor to an internal
/go/…path → do not submit unless you accept a wrapped,nofollowclick-out. This is answer 3 wearing a link's clothes.
The fourth one is the trap that cost me the most. A listing page looked complete: 200 OK, index, follow, canonical to itself, my name and description all present. The play button was <a href="/go/copero" rel="nofollow noopener">. My domain appeared twelve times in that page — meta description, structured data, screen-reader text — and zero times in an href.
Check three listings, not one
One listing tells you what the template does. It does not tell you what the template does for you, because sites often treat "featured" and "default" differently.
So take three existing entries: the one on the front page, one from the middle, one you had to search for. If all three behave the same, that is the site's policy. If only the front-page one links out, the site links out for people who pay — and the pricing page will confirm it.
def test_site(listing_urls, game_domains):
for url in listing_urls[:3]:
status, final, html = fetch(url)
for domain in game_domains:
found = anchors(html, domain)
if found:
print(url, "->", found[0])
break
else:
print(url, "-> no outbound anchor")
Run it, read three lines, close the tab. Ninety seconds if the pages are fast; five minutes if you have to hunt for the listing URLs.
What it saves, in numbers
I ran it on 58 game and daily-game directories, each measured by fetching a real listing rather than by reading the site's pitch.
-
29 published outbound links to game sites with no
nofollow— a checkable property of the page. - 9 of those 29 had a submission entry I could fill in at all.
- 6 of the 9 did not require an account.
From 58 directories to 6, before writing a single word of submission copy. Every one of those rejections is a form I did not fill in, an account I did not create, and a mailbox round-trip I did not spend.
The same filter applied to a different question — "will the free tier actually pass a link?" — over 23 listings that demanded a reciprocal badge: 10 are checkable today, and 3 pass a followed link. Four are nofollow, two are pages with no anchor, one is a 404. The badge requirement alone does not predict the outcome; only reading an existing listing does.
Why the directory's own dashboard does not count
Every one of those failures reported "live" somewhere. The dashboard is not lying — it is answering "does a page exist for your submission", which is a different question from "does that page link to you in a way that counts".
The gap between those two questions was 147 records in my own ledger: 20 marked live, 8 actually followed. And of the 13 "submitted" records that even had a URL to check, zero carried a followed link.
The rule I use now
Before submitting anywhere:
- Find three existing listings on the site.
- Read their outbound anchors —
hrefandrel, not the visible text. - Check the listing page's
robotsmeta andcanonicalwhile you are there. - If the template does not link out, close the tab. There is no submission worth making to a page that will not carry a link.
- Record the answer in the ledger before you fill the form, so the next person does not repeat the five minutes.
The order matters. I used to do this after submitting, which is how you end up with a spreadsheet full of "live" records and a browser history full of forms.
I build Copero, a free browser football career game — no download, no account:
https://copero.games/?utm_source=devto&utm_medium=blog&utm_campaign=backlink-2026-11
The counts above come from 58 game directories measured on 2026-11-18 and a 147-record ledger audit; the /go/ example was fetched the same day, twelve mentions and zero anchors.
Originally published by Dev.to WebDev. Aggregated on AIWithGhost for educational purposes — full credit and traffic to the original publisher.