Yet Another IG Journal

All right let's try IG. Every other journal seems to be an IG one so this is just peer pressure. All the cool kids are botting IG.

This isn't my main project but it looks like it's mostly just waiting around so I'm going to set it up and leave it to run. Whatever happens, happens.

The goal:

Picking a few niches and building up accounts in that niche. I think 10k seems to be everyone's goal so let's get to 10k followers and from there drop some affiliate links in the bio, the posts and sell shoutouts I guess.

Doesn't seem like it will take that long although it's obviously niche dependent.

Setup so far:

Custom bot to create accounts.

I have what I *think* will be a good proxy setup. I've not used them on IG before but they work well on other projects and it's been fine so far. I'll find out for sure when the scaling starts.

Custom bot to like/post/follow/unfollow/comment. No, there's nothing wrong with the public ones and they're cheap. But I like to know what's going on under the hood and there are some bits I wanted to do differently.

Custom content setup. It's pretty basic at the moment I'm going to try and find the time to improve this.

Account settings:

I haven't a frigging clue. I've skim read a bit about what others are doing and there's a degree of randomisation to it of course.

I'm just getting started and all the accounts are new so I'll limit it to like 300 actions a day for now and see what happens. I'll tinker with it from there and I'll track my actual numbers as it goes.

Obviously the faster the growth, the better. But as a little side hustle if it's growing at a decent rate without me having to do shit then I'm happy to just let it sit. If I find the time or the project proves worthwhile I'll start experimenting a bit more.

Nothing to report yet. I've just finished setting things up. I'll start reporting stats and growing pains as they happen.

Ah crap. I titled it SEO rather than IG.

Did you build the bot to like/follow/comment yourself?

ddlmanager said:

Did you build the bot to like/follow/comment yourself?

Click to expand...

Yes. The public ones are fine I'm just a control freak.

If a mod happens to see this would you mind renaming my title to IG instead of SEO? I would also accept 'hapless moron can't title a thread properly'.

Following, I have read your other threads and would like to see how you do this one, I am sure it will be interesting

SEOMadHatter said:

Ah crap. I titled it SEO rather than IG.

Click to expand...

I've fixed that for you. Good luck with this one.

SEOMadHatter said:

Yes. The public ones are fine I'm just a control freak.

If a mod happens to see this would you mind renaming my title to IG instead of SEO? I would also accept 'hapless moron can't title a thread properly'.

Click to expand...

I like your second title suggestion

Good luck on your journey!

Naulex said:

Following, I have read your other threads and would like to see how you do this one, I am sure it will be interesting

Click to expand...

Cheers. I like the idea of doing a journal where I can talk about the actual details. Plus it's another project to mix it up with.

MisterF said:

I've fixed that for you. Good luck with this one.

Click to expand...

I appreciate that, thanks. My subconscious while I was writing the title: "Oh you're going to sit around on your hands waiting to see what happens? Must be SEO."

Alright, let's get this started.

Started a couple of niches up. Turns out one of the niches I already do some work in is a thing on IG and I never knew. I put a link in a bio and it got some traffic. There are people on IG. Who knew. I could dive straight for the 100 accounts running (I think that's the first milestone everyone uses, right?) but no doubt I'm going to do something wrong so I might as well find out what and burn the first 20 testing. Whether they're burned or not I'll probably wait until next week to do the first 100 I've got some stuff to do in the meantime.

I've set them to do 2-300 actions a day, taking breaks and sleeping normal hours for the account's country. No idea if this is too much or too little, let's find out.

So while the first accounts grow themselves or get themselves banned (whichever) I want to look at the posting strategy. Leaving accounts to promote themselves requires nothing but the follow count doesn't mean anything if the accounts don't post anything. I haven't decided on a posting rate yet (I assume it's going to have to depend on the niche) I'll have to look at decent accounts and see how often they post. But I don't want to be posting the same stuff as everyone else in the niche and I really don't want to actually have to work on any of this. A couple of hours, maybe a day tops and I'll have the content side of things set up so they'll get a half decent feed.

So the goal is to finish that up and then leave it a week and see how the accounts do.

Just noticed SocialBlade doesn't do IG personal accounts so I'm going to have to add some tracking in tomorrow so I can properly test what works.

I had a lot of other stuff I needed to do today for the SEO project but this was playing in the back of my head all day so I'm working a little on it tonight and I'm planning to give it most of the day tomorrow.

I tweaked the strategy a little bit trying to focus more on active users rather than just followers. I improved the speed of the bot and made it look a little more human in the process. I've decided to go with what I think is a common strategy - split the accounts up into main accounts and feeder accounts - or master and slave as I'm calling them.

The master accounts are the ones I want to actually grow with people who are interested in that niche. My theory (based on absolutely no experience) is the f/uf will draw more attention to the account and the account will have a lower engagement from people who are following back without really being interested. The bot will still operate the accounts and keep them posting and engaging somewhat but they should look reasonably white hat.

The slave accounts will be promoting themselves and then the idea is to link them back to the master account. How to safely do that I have no idea. I've seen people put links in bios, in posts and I've seen mentions of DMing people though I don't even know where the DM button is on IG.

If anyone has any insights on how to safely send followers from the slaves to the masters I'm all ears. In theory, IG shouldn't ban the master accounts, right? I mean otherwise, we'd have the IG equivalent of negative SEO. So if my master accounts get banned presumably I could just get my competitors banned.

The goal for tomorrow is going to be finishing up the content posting and making sure the masters have a good quality content feed without me having to you know... work. If I manage to get that done early enough I'll get a new server and stick 50~ accounts live. No doubt I'll screw them up without knowing what I'm doing yet but if I'm going to get it wrong I might as well get it wrong on a decent enough scale to find out quick.

I'll start posting stats in a few days.

Alright, I hit the first roadblock. Looks like maybe I need to tweak my warmup timings the slaves have been hit by what I'm assuming is a follow block. Cool, there's only one way I'm really going to learn this stuff. I'll give that account a rest and tweak the settings a bit.

I finished with the content stuff for the slave accounts. Still not entirely sure on the strategy on diverting their traffic to the masters I guess I'll split test bios/tags/DMs and see what works. Now I can give them a niche and just let them run (after tweaking these timings I guess).

The master account data and setup doesn't need to be as automatic. Since I'll be doing them on a smaller scale their content goes through a web backend first so I can moderate and set a decent description. It's' manual work but I can line up a months work of posts for an account within a half hour. If that's my workload I can deal with it.

I wanted to use this as an excuse to work on my reverse engineering so I'd at least feel more productive working on it rather than my normal projects. The problem with that plan is I don't like giving up. Rather than take the quick and easy way of doing something so I could move on I blew 8 hours working on something which from the project's point of view doesn't actually need to be done at the moment.

There's a crap ton of things I'd like to add in but, in reality, I just need to finish up the easy part of the master accounts content posting and I can let it scale to 100 accounts by tomorrow morning.

I just need to stop getting distracted.

Starting to like this project now.

I've been reading a little from other journey's here. Not as much as I probably should but I've been taking the same approach as I take with SEO. Look at what people are actually doing in the wild. Kind of wishing I'd got into this a while ago but that's how it goes.

Account types are now split into three different types.

Master accounts are as whitehat as I'm ever going to get. Content is filtered, the good stuff is shown to the web front end and then it's posted over time. 10 minutes work will keep it fed for at least a week and it's easy to outsource if I ever went to scale.

Slave accounts I fully expect to see banned. They're entirely automated. I don't quite know how much traffic they'll get to the master account but there's only one way to find out.

The third I've called Blackhat Master accounts. Their content is still manually approved through the back end and for all intents and purposes, it'll create a good quality feed the niche will want to see. But I'll be boosting their posts engagement and running a slightly modified f/uf plan for these accounts.

Testing a bunch of proxies out it'll likely be a few weeks before I see how well these ones are doing. Burned a couple of accounts during development and I'll no doubt burn this entire set testing but that's how to learn.

I think the code is pretty much done for now. The more I look at it the more I see things I would like to add but I'm going to give at least a few dozen accounts and leave it running for a few weeks while I get back to my main project. Ordered some new proxies and waiting for a server to be provisioned so I can give this some decent hardware.

No stats worth speaking of yet accounts are still slowly warming up. Surprised how high the follow back rate is though I'd expected it to be a lot lower.

cloaking is helpful for some links to put in instagram or fb for example

Rough start to the scaling.

Ah bunch of the slave accounts were phone verified. Not sure if it's to do with the proxy range they are on or the spammier following method they were using. To find out I'm going to switch some of the blackhat masters to the same range. If they get burned then I'm pretty sure it's the proxies.

My host refunded the money for the server. No reason given so I'm assuming they're not setting that one up. Ordered another with a different host. For now I have a video project which isn't doing much so it can borrow room on that server.

I was reading some stuff on the IG algo. Some cool ideas on stuff I could add to the code but for the moment I think I'm going to just shift things up a bit.

One of my niches has a massive account. Low engagement rate but we're still talking thousands of comments and likes even with dozens of posts a day. I'd need to run a bunch of accounts in that niche and watch them to see if he's using slaves but it looks like he's just been getting away with running f/uf for an age and it's clearly working.

I already work in a bunch of different niches so I have a bunch of landers. I can leave these accounts to run for weeks with just a few minutes moderating content and I don't really have to worry if they eventually burn out because it's not like it takes a bunch of work to maintain. Plus, I can always use them to promote a more white hat account later on.

Still running all three types but I think I'm planning to focus on the blackhat ones. They'll require the least effort and I'm relatively sure if I get them on a half decent proxy they'll be the best short-medium term earners.

amandaedson said:

cloaking is helpful for some links to put in instagram or fb for example

Click to expand...

Genuine question: Do you know why you'd cloak links or is this just something you've read somewhere? Doesn't really apply at the moment. I'm not putting links on mass profiles or even at the monetisation stage yet.

I spent some time reading up on the IG ranking algorithm specifically how to get your stuff showing up in the explore and top images feed.

IG seems weirdly similar to SEO in that everyone seems to parrot the same stuff but few seem to be actually testing any of it. Conventional wisdom is that f/uf hurts your accounts reach but split tests seem to say otherwise and I'm watching a bunch of big accounts do it just fine in the wild.

I am tempted to make a branch of slave accounts as an engagement group. They'd need to have a much smaller footprint to make it less obvious but they can certainly build their own follow numbers and like/comment the master accounts post as soon as it goes down.

Looks like the PV's only hit 3 accounts so my guess is it's the proxy not the follow method. Adding some new blackhat master accounts in some new niches and I'm going to put 20~ more slaves into an existing niche with existing slaves so I can use it to boost some posts in the first hour see if they bring in some organic followers.

Gor a bit caught up with travel, injury and the SEO project but I've been keeping this at the back of my head and I'm putting a day or two in this week.

Disclaimer: Don't take anything you read here as gospel. I don't know what I'm talking about.

Chuck a stone on FB groups, IM forums or YouTube and you hear the same stuff being spouted about the IG algo. But I've yet to see a decent case study to back half of the common knowledge shit up. Just because something seems logical does not mean that's how an algorithim actually works. And I really do suspect a lot of these theories are just getting parroted without any real testing going into them.

Parts of my system so far I like. My follow/unfollow rates are still incredibly low as I warm accounts up but the content feed system is working well and I can manage a lot of good quality feeds with relatively little work.

I am shifting gears a bit though. Rather than run a bunch of different niches with slave accounts for each I'm going to try and focus on one account for at least a month as a proof of concept.

The goal is to run a main master account with at least one post a day, a couple of slave accounts using follow/unfollow to feed into it and a couple of dozen in an automated pod within that niche. It's not a scientific test to see if niche engagement matters - it's a kitchen sink test.

I'm going to target a few dozen hashtags in my niche and smack the posts with spammy powerlikes and my automated niche pod. See if the posts can rank and bring in some real niche followers.

I'm also looking at a spider to grab my competitors hashtags and monitor them. See what kind of accounts are ranking for them with what kind of engagement. Might help me judge where I can rank.

A day or two to tweak the code a bit and we'll get started including stats.

Finally up and running.

Stats
Followers to master accounts: 340
Active slave accounts: 30 (Still warming up)
Earnings: -$190 (proxies, hosting and a bunch of spammy power likes for testing).

Decided to run two niches to start with rather than just one. It means I can give numbers, helps with testing and I wanted to try two different types of niches.

Revamped the system to do what I need it to do now. The focus is more on building accounts within a niche and using them as a pod to boost the master account rather than just f/uf. I'm curious to see how effective it is.

I targetted some tags which had accounts worse than mine ranking for the top spot with pretty low metrics. After hitting my post with a bunch of overkill metrics in the first hour I didn't take the spot. Not sure if there's a delay on this kind of thing or (more likely) a bunch of spam blasts just don't cut it. Didn't fully expect it to but I'm not one to trust conventional logic without seeing a study - especially when it's cheap and easy to test.

I revamped a lot of the system to do the engagement pod stuff and cut down a lot of my workflow. At this point there isn't really too much to do but sit and wait. Since account creation is automated I'm going to push that closer to 100~ accounts but I'm mostly just sitting around waiting for the accounts to warm up and see if their engagement makes any kind of difference.

Stats
Followers to master accounts: 346 (+6)
Active/Warming slave accounts: 32

Doesn't look like a huge change but actually, there were a bunch of new accounts added and a bunch of others hit verification

A couple of potential reasons for this. My chief assumption is it's the shitty proxies but it costs me next to nothing to add new ones so I'm adding more accounts and tweaking the settings to make it look more human-like to begin with. I'm also going to give the accounts a gap before they start doing any actions like follow/unfollow.

I'll be following your journey, good luck!

Im interested in something. How many followers do you get/day on child accounts? And what percentage are you sending to main account?

The bad news: The latest PV wave hit the majority of my slave accounts. Masters are fine and some slaves survived. Doing some split testing to see if I can figure out what the trigger was and avoid it happening again.

The good news: I did (briefly) see a master account ranking in the top posts section for one of the smaller niche hashtags. Didn't get much from it but it was a good proof of concept.

RoyalCobra777 said:

Im interested in something. How many followers do you get/day on child accounts? And what percentage are you sending to main account?

Click to expand...

Can't really tell you because I've disabled tracking on the slaves. I can tell you it's a very low % though. I could be a lot more aggressive with it but my focus at the moment is trying out automated niche pods to promote my posts.