Watching Reddit kill aged accounts over time

[trueplayer]

Been watching Reddit closely the last year because I kept seeing older accounts randomly lose reach or get shadowed without doing anything wild.

What stood out is that age alone doesn’t protect anything anymore, fr. Some accounts were years old, light posting history, no obvious spam, and still slowly stopped getting visibility.

What broke first wasn’t bans, it was engagement. Posts would submit fine, show on profile, but never hit feeds or got stuck at 0 views for hours.

Another pattern was delayed action. Stuff would work for weeks, then one day the account just felt “cold” overnight, like trust got re-evaluated in batch.

The biggest flag I noticed was behavioral mismatch. Posting style, timing, and subreddit hopping didn’t line up with how a normal user drifts over time, and Reddit seems way better at spotting that now.

IP or device stuff didn’t instantly nuke anything, but over time the system seemed to connect dots anyway. No drama, no warnings, just quiet throttling.

What surprised me is how much weight they seem to put on long gaps + sudden activity. Old accounts waking up and acting purpose-driven just don’t age well anymore.

Basically multiaccounting doesn’t usually fail loudly, it just decays. You think it’s working until you realize nothing is actually landing.

Curious if others are seeing the same slow death pattern or if anyone still sees old accounts holding trust long-term.

 

[ScribScribScrib]

I've been saying this for 2 years now: You need accounts that are both recently active and that have a history of activity.

I spend more time managing Reddit accounts than anything else.

I would love to outsource it, but I'm too afraid and kinda have trust issues.

 

[moonshine7000]

Reddit flags based on behaviour

 

[washingtondc]

It seems Reddit quietly lowers reach on older accounts that suddenly act differently, rather than banning them outright.

 

[stin99]

I agree guys, dont use reddit if you dont how to use it

 

[moonshine7000]

Reddit detect user's behavior and pattern

Yep it comes down to how people post.Proxies are only one part of the question

 

[RedditHelper]

Yep it comes down to how people post.Proxies are only one part of the question

Yeah, that true. People always talk about the proxy while discussing about Reddit but no one talk about the user behavior and actions.

 

[moonshine7000]

Yeah, that true. People always talk about the proxy while discussing about Reddit but no one talk about the user behavior and actions.

Yes,going to make a separate thread on that later

 

[updownleftright]

Yeah, that true. People always talk about the proxy while discussing about Reddit but no one talk about the user behavior and actions.

Agreed, proxies help but behavior matters so much more now. old accounts acting different after being idle seem to get flagged/lose reach fast.

 

[BannerBandit]

Yep same here, Reddit slowly kills old accounts with sudden bursts or weird behavior, engagement dies before anything else.

 

[quynhuyen]

Please submit a support request; I just submitted one and the account return rate is very high.

 

[megakim]

Yep seeing same thing. Old accounts don't die they just get quietly muted after a trust refresh. Long gaps plus sudden focused posting kills reach over time.

 

[Nut-Nights]

Use reddit like a regular user, i mean dont use it with an intent of growing it. Just be natural and watch your account grow and sustain even if you go wild with posting comments and all other things. I have tested this theory with mainstream and adult niches.

 

[theusualkeysersoze]

Been watching Reddit closely the last year because I kept seeing older accounts randomly lose reach or get shadowed without doing anything wild.

What stood out is that age alone doesn’t protect anything anymore, fr. Some accounts were years old, light posting history, no obvious spam, and still slowly stopped getting visibility.

What broke first wasn’t bans, it was engagement. Posts would submit fine, show on profile, but never hit feeds or got stuck at 0 views for hours.

Another pattern was delayed action. Stuff would work for weeks, then one day the account just felt “cold” overnight, like trust got re-evaluated in batch.

The biggest flag I noticed was behavioral mismatch. Posting style, timing, and subreddit hopping didn’t line up with how a normal user drifts over time, and Reddit seems way better at spotting that now.

IP or device stuff didn’t instantly nuke anything, but over time the system seemed to connect dots anyway. No drama, no warnings, just quiet throttling.

What surprised me is how much weight they seem to put on long gaps + sudden activity. Old accounts waking up and acting purpose-driven just don’t age well anymore.

Basically multiaccounting doesn’t usually fail loudly, it just decays. You think it’s working until you realize nothing is actually landing.

Curious if others are seeing the same slow death pattern or if anyone still sees old accounts holding trust long-term.

You are one of the very few people who have analytical thinking and actually apply it, greatly value reading your stuff.

Now that it has dawned on people that SMM panels are total rubbish traffic and that cloud phones are not a good choice for automations or anything, really, they keep asking how to do multiple accounts on real phones and automate phone farms.

I understand that platforms give more trust to mobile traffic, especially if it is via app, but you cannot really have a more stable solution that an anti detect hosted on a good windows server with uncontaminated statis isp proxies, it still sucks, but the website today do absolutely graph behaviour. At least a windows profile looks like a user at home browsing from laptop or PC. This is important for later.

The entry point via mobile proxy and all is frankly a joke these days, since if you have a real phone, let us leave running multi accounts alone, since the signature is the same on all of them, but let us talk about the difficulties of mobile automation.

Say you have the real app isntalled, not apk side loading, nownow, wheres to process to emulate 40 minutes of jitter, more or less realistic browsing, scrolling etc and since you wanna run a proxy, youll have to root that thing which breaks many other things and leaves traces/noice.
So how to manipulate trianglation data which needs to be in sync with the proxy? And oh, one needs to consider gyro activity, does the perosn randomly walk etc, if your automation does not cover all of that, your accounts will not live. This cannot be done with a bit of JS manipulation at layer 7.

The websites and apps are getting better because they must, all these detection methods have been available all along, but now they are facing pressure from investors and such, cannot show these ads to 50% bots.

 

[cooooookies]

Thanks OP, this is very interesting observation. I started with Reddit automation recently, my setup tries to mimic human behaviour as far as possible. But there is a lot to consider and to research. Next thing I will consider is normal distribution vs equal distribution for waits.

 

[LucaBernadi]

Multi-accounting doesn't get punished fast, it just slowly stops working.

 

[trueplayer]

@ScribScribScrib yeah exactly - “recently active + historically active” is the combo people miss. Accounts need a continuous low-noise heartbeat, not phases of dead + grind. Outsourcing breaks because you lose behavioral continuity, not because helpers are dumb.

@moonshine7000 + @RedditHelper you’re both right and this is where most people still cope with proxies. Behavior is the graph, IP is just a node. Same posting cadence, same dwell times, same subreddit clusters = cooked eventually no matter how clean the proxy looks.

@washingtondc + @BannerBandit + @megakim that quiet muting before any ban is the key signal. Reddit doesn’t need to nuke you if they can just remove you from feeds. Most people think “account still posts = safe” while it’s already dead.

@updownleftright old idle accounts waking up with intent is probably the worst pattern right now. Even spacing posts doesn’t save it if topic focus suddenly snaps into place like a campaign.

@Nut-Nights “use it like a regular user” is correct but incomplete - regular users drift. Interests decay, posting hours change, comment vs post ratio shifts over months. Static “natural looking” behavior is still unnatural.

@quynhuyen support requests can work short term but they don’t fix the underlying trust score. You might get visibility back then lose it again next refresh if behavior stays off.

@cooooookies normal vs equal distribution matters more than people think. Equal waits scream automation. Humans are bursty, distracted, inconsistent. Long pauses + fast streaks look way more real than smooth curves.

@LucaBernadi that line sums it up perfectly. Reddit doesn’t punish, it devalues. People only notice when ROI hits zero.

@theusualkeysersoze appreciate that. I agree on cloud phones being mostly cope. App traffic isn’t magic if the behavior graph collapses. Desktop profiles with long-lived noise still blend better than people want to admit, especially when accounts age into “boring user” territory.

Net takeaway for anyone reading: aged accounts aren’t assets anymore, they’re liabilities if you don’t maintain behavioral continuity. Reddit is scoring trajectories, not snapshots. No action buttons, no warnings, just math running in the background.

 

[FirstLight]

I would love to outsource it, but I'm too afraid and kinda have trust issues.

You said it. There is a real opportunity for someone to start a business just selling accounts with current, on-going Reddit activity that you can just instantly start working on. The rate that old accounts get killed is insane.

 

[Gulcan_kamps]

exactly. discussions about Reddit usually center on proxies, but few consider how users behave and participate.

 

[dmitriyeva9]

I completely agree with trueplayer's observation on Reddit's algorithm changes, it's indeed getting more sophisticated in detecting behavioral mismatch and account trust decay. The part about long gaps in activity followed by sudden spikes in engagement is particularly interesting, as it highlights the importance of maintaining a consistent posting schedule and avoiding sudden changes in behavior. Have you guys noticed any correlation between account trust and the type of content being posted, such as links vs text-only posts or engagement-driven content? Additionally, what are your thoughts on using Reddit's built-in features like Reddit Ads to boost post visibility and potentially mitigate the effects of algorithmic throttling?