Reddit Account Bans - Looking for Insights

V

Elite Member
Jr. Executive VIP
Jr. VIP
Joined
May 18, 2012
Messages
7,630
Reaction score
7,493
Hey guys!

So I finally decided to test the Reddit hype everyone’s been talking about for a while now.

About two weeks ago, I created roughly 100 Reddit accounts using separate proxies and an anti-detect browser. Everything looked fine initially - no issues, no instant bans. I didn’t really touch the accounts after creation, though, just let them sit.

Fast forward to a couple of days ago, and when I checked again yesterday, almost all of them were banned.

Now I’m wondering if not logging into or warming up the accounts after creation triggered the suspensions, or if Reddit is just getting way more aggressive with detection lately.

Has anyone here actively working with Reddit marketing experienced something similar? Would appreciate any insights on what might’ve gone wrong or how to avoid this next time.
 
Anti detect browsers are detected by most platforms. I personally switched to using a google pixel and installing grapheneOS on it. This allows for 32 sandboxed accounts on the phone that can’t be detected.

Register on the Reddit app using a Google account. Create a password in the Reddit app after registration in case G kills your account.

I have more than 300 accounts this way for more than 8 months now. I don’t have to use them daily. Some have sat for months and they are fine. I used one non rotating mobile proxy that matched the SIM card carrier for all of them.

The catch: The phone needs to have a real SIM card and have been activated within the last year by the company that made the SIM card. The SIM card does not have to remain activated but you need to keep the same SIM card in the phone at all times.

2nd catch, the phone has to be sold directly from Google so you can use OEM unlocking. That is very important. If you buy a used one on eBay like me verify with the seller that OEM unlocking isn’t greyed out. I learned that the hard way.

Reddit is easy. Google and discord are harder.
 
Last edited by a moderator:
There will be all sorts of automated heuristics.

I would not be surprised if, yes, creating accounts then just not making a first interaction with them most definitely would be raising some kind of suspicion score.

Plus also depending on what email domain you're registering with. If it's a no-name throwaway '10 minute mail' type domain then yes I would expect registering all of them with random string usernames 'at' SomeThrowawayFreemailDomain.tld that nobody has ever heard of except the countless spam accounts also using that domain as seen by Reddit then yes that won't be helping either.

Some things to try and change the outcome

1. Register some cheap domain with a real name not a random string and set up a mailserver for it with one address and catchall the rest. Register some accounts using that throwaway domain and see if they get banned in the same amount of time. make sure your registrar offers domain privacy to avoid reddit making connections about your registered domains if one gets marked as bad faith

2. When signing up add 'personality weights' to your code so your new accounts pick some consistent interests when they sign up and pick their starting subreddits. Instead of picking either nothing, or clicking them randomly with zero interest consistency

3. Probably most important. Make meaningful non-throwaway non-single-word comments, comment replies and posts in some of their selected subreddits to bootstrap the account as not "registered to be inactive". If you use chatgpt to write the comments and submissions for you they will get caught reported and banned by the users who look for these accounts (They got one of our iptv operations last month). Write the comments yourself just to see if they survive longer

4. Use residential proxies. Yes I know they are not cheap. But they are also not shared nor marked vpn public ip addresses. vpn ips and which are also going to be a contributing factor to untrustworthy account weight scoring done by reddit
 
Hey guys!

So I finally decided to test the Reddit hype everyone’s been talking about for a while now.

About two weeks ago, I created roughly 100 Reddit accounts using separate proxies and an anti-detect browser. Everything looked fine initially - no issues, no instant bans. I didn’t really touch the accounts after creation, though, just let them sit.

Fast forward to a couple of days ago, and when I checked again yesterday, almost all of them were banned.

Now I’m wondering if not logging into or warming up the accounts after creation triggered the suspensions, or if Reddit is just getting way more aggressive with detection lately.

Has anyone here actively working with Reddit marketing experienced something similar? Would appreciate any insights on what might’ve gone wrong or how to avoid this next time.
Reddit's detection got much more aggressive lately. Creating 100 accounts without activity triggered their automation detection.

You need gradual warming - subscribe to subreddits, upvote content, comment occasionally. Build karma slowly over weeks not days.

Use residential proxies not datacenter IPs. Spread account creation over weeks and vary creation patterns.

Mass creation without proper activity simulation almost always results in bans now.

What were you planning to use the accounts for?
 
  • Like
Reactions: V
Hey guys!

So I finally decided to test the Reddit hype everyone’s been talking about for a while now.

About two weeks ago, I created roughly 100 Reddit accounts using separate proxies and an anti-detect browser. Everything looked fine initially - no issues, no instant bans. I didn’t really touch the accounts after creation, though, just let them sit.

Fast forward to a couple of days ago, and when I checked again yesterday, almost all of them were banned.

Now I’m wondering if not logging into or warming up the accounts after creation triggered the suspensions, or if Reddit is just getting way more aggressive with detection lately.

Has anyone here actively working with Reddit marketing experienced something similar? Would appreciate any insights on what might’ve gone wrong or how to avoid this next time.


1. Anti-detect browsers really don't work.
2. VPNs and residential proxies really don't work.

What works is multiple phones and wifi hotspots.
 
anti detect browsers dont work you have to do your own fingerprint spoofing
 
Chameleon Mode, Adspower and Dolphin Anty should work, otherwise clean phone either GrapheneOS, iOS or stock android should work. I wouldn't recommend proxies as they are easily detected. You have to use these accounts a bit like scrolling. There is good warmup technique that involves creating AD profile, using the profile by using Google and normally browsing, visiting few sites beforehand, using Reddit whilst not logged in. Then test your reCAPTCHA result here: https://antcpt.com/score_detector/

If your score is above 0.7 you can create the Reddit account. If below and score doesn't increase make new AD profile.

For Reddit, you should preferably reply to someones comment or comment within one week in a subreddits that allow fresh new low karma accounts. If you want I can share additional tips.
 
What kind of proxy did you use?. you're skipping that part
 
  • Like
Reactions: V
Anti detect browsers are detected by most platforms. I personally switched to using a google pixel and installing grapheneOS on it. This allows for 32 sandboxed accounts on the phone that can’t be detected.

Register on the Reddit app using a Google account. Create a password in the Reddit app after registration in case G kills your account.

I have more than 300 accounts this way for more than 8 months now. I don’t have to use them daily. Some have sat for months and they are fine. I used one non rotating mobile proxy that matched the SIM card carrier for all of them.

The catch: The phone needs to have a real SIM card and have been activated within the last year by the company that made the SIM card. The SIM card does not have to remain activated but you need to keep the same SIM card in the phone at all times.

2nd catch, the phone has to be sold directly from Google so you can use OEM unlocking. That is very important. If you buy a used one on eBay like me verify with the seller that OEM unlocking isn’t greyed out. I learned that the hard way.

Reddit is easy. Google and discord are harder.
Yes, I've also heard that Graphene OS on a Pixel does a great job if you're into SMM. The only downside is that it will be quite difficult for me to automate tasks on a newer Android device. I have automated stuff until Android 9, and after that, they have tightened up the security and privacy.
There will be all sorts of automated heuristics.

I would not be surprised if, yes, creating accounts then just not making a first interaction with them most definitely would be raising some kind of suspicion score.

Plus also depending on what email domain you're registering with. If it's a no-name throwaway '10 minute mail' type domain then yes I would expect registering all of them with random string usernames 'at' SomeThrowawayFreemailDomain.tld that nobody has ever heard of except the countless spam accounts also using that domain as seen by Reddit then yes that won't be helping either.

Some things to try and change the outcome

1. Register some cheap domain with a real name not a random string and set up a mailserver for it with one address and catchall the rest. Register some accounts using that throwaway domain and see if they get banned in the same amount of time. make sure your registrar offers domain privacy to avoid reddit making connections about your registered domains if one gets marked as bad faith

2. When signing up add 'personality weights' to your code so your new accounts pick some consistent interests when they sign up and pick their starting subreddits. Instead of picking either nothing, or clicking them randomly with zero interest consistency

3. Probably most important. Make meaningful non-throwaway non-single-word comments, comment replies and posts in some of their selected subreddits to bootstrap the account as not "registered to be inactive". If you use chatgpt to write the comments and submissions for you they will get caught reported and banned by the users who look for these accounts (They got one of our iptv operations last month). Write the comments yourself just to see if they survive longer

4. Use residential proxies. Yes I know they are not cheap. But they are also not shared nor marked vpn public ip addresses. vpn ips and which are also going to be a contributing factor to untrustworthy account weight scoring done by reddit
Great points. I will keep in mind when registering the next batch of accounts.
Reddit's detection got much more aggressive lately. Creating 100 accounts without activity triggered their automation detection.

You need gradual warming - subscribe to subreddits, upvote content, comment occasionally. Build karma slowly over weeks not days.

Use residential proxies not datacenter IPs. Spread account creation over weeks and vary creation patterns.

Mass creation without proper activity simulation almost always results in bans now.

What were you planning to use the accounts for?
Yeah, they've been getting stricter by each passing day. I was thinking of going the nsfw route, but before I could do that, I am working on some kind of warmup automation.
Did u used datacenter proxies?
Yes, but some of the accounts that I customized a little are still alive. Not all of them, though. So, I am still wondering if it's because of leaving the accounts sitting idly after registration.
Every time Reddit gets hyped again, they quietly tighten the noose a little more
Yeah, man, they're getting too strict. even for genuine users. :D
What kind of proxy did you use?. you're skipping that part
Datacenter proxies, hehe. Sorry, my bad. :D
 
Reddit expects normal activity, so unused or mass‑made accounts often get banned fast.
 
tbh you can forget about automation at this point if you’re not a genius. There’s a lot to know about automation now people didn’t need to know about in the past. It’s all detected for the most part when you buy a product meant for mass automation

These websites are billion dollar corporations. They have already bought and reverse engineered the products on the market for automation of their services.

Try to emulate a mouse movement over a button with a click in JavaScript. There is so much that goes into just that simple action. It’s hard.
 
Yes, I've also heard that Graphene OS on a Pixel does a great job if you're into SMM. The only downside is that it will be quite difficult for me to automate tasks on a newer Android device. I have automated stuff until Android 9, and after that, they have tightened up the security and privacy.

Great points. I will keep in mind when registering the next batch of accounts.

Yeah, they've been getting stricter by each passing day. I was thinking of going the nsfw route, but before I could do that, I am working on some kind of warmup automation.

Yes, but some of the accounts that I customized a little are still alive. Not all of them, though. So, I am still wondering if it's because of leaving the accounts sitting idly after registration.

Yeah, man, they're getting too strict. even for genuine users. :D

Datacenter proxies, hehe. Sorry, my bad. :D

It's because u use Datacenter proxies, u need to use residential.
 
  • Like
Reactions: V
I've also had recent Reddit bans.
It seems like they do ban people for a wide variety of reasons
 
Here's my strategy to growing Reddit accounts from scratch, using Dolphin + Chrome + residential proxies:

1. On a new Anti-detect profile, I would just browse random websites and accept various cookies. Like Facebook, Youtube, X, Google, etc.

2. Google up something + Reddit (ie Iphone 16 reddit). Then click on a Reddit page.

3. Scroll to any post and try to upvote or comment. This will trigger a registeration.

4. Register using even temp-mail works for some reason. Or Gmail to be safe but I'm not going to bother buying proxy phone numbers lol.

5. Subscribe and browse and upvote random posts on popular subreddits.

6. Make ONE and only 1 comment on any subreddit that you can gain karma off of. Make sure the subreddit you choose has no CQS (contribute quality score) requirement.

7. Keep browsing and upvoting random posts. Don't need to spend hours on this. Only a few mins will work.

8. Every 24 hours, make 1 comment, until the account is at least 5-7 days old (7 to be safe). Seems like commenting more than once per day triggers Reddit's algorithm early on. Posting is probably out of question because that may trigger their shadowban too. You can also browse/upvote random posts/comments but doesn't have to be that frequent.

9. On day 7, you can ramp this up to every 12 hours or even more frequent. You might be able to post too. I wouldn't mind taking the risk at this point because if the account is still on Reddit's anti-spam radar, chances are, it won't leave their radar anytime soon.

Now I'm not sure when your account is fully Reddit-proof of anti-spam filters so you'll have to experiment. I grew lots of accounts this way, though my accounts aren't used for OFM. My use case is mostly for upvoting personal posts/comments.
 
Anti detect browsers are detected by most platforms. I personally switched to using a google pixel and installing grapheneOS on it. This allows for 32 sandboxed accounts on the phone that can’t be detected.

Register on the Reddit app using a Google account. Create a password in the Reddit app after registration in case G kills your account.

I have more than 300 accounts this way for more than 8 months now. I don’t have to use them daily. Some have sat for months and they are fine. I used one non rotating mobile proxy that matched the SIM card carrier for all of them.

The catch: The phone needs to have a real SIM card and have been activated within the last year by the company that made the SIM card. The SIM card does not have to remain activated but you need to keep the same SIM card in the phone at all times.

2nd catch, the phone has to be sold directly from Google so you can use OEM unlocking. That is very important. If you buy a used one on eBay like me verify with the seller that OEM unlocking isn’t greyed out. I learned that the hard way.

Reddit is easy. Google and discord are harder.

Are you 100% sure about the SIM thing ?
If they can detect that shit can’t they also link all of your accounts to same SIM card ?
 
Back
Top