OnlyFans Agency TikTok Automation to manage creators

I spent a few days building tools I expected to need for large-scale content in an OnlyFans content house. That plan did not land. The Airtable base still did. It is built to run hundreds of social accounts, and the schema does not care if that becomes thousands.

This version is TikTok. The same tables work on Instagram. They also work as a Reddit tracker — upvotes against clicks over time if you grow organic. Add a model with a new row, her name, and the TikToks she runs.

One row per model: posts today, yesterday, and seven days

The first table is a rollup. Three models in the recording. You could drop in hundreds.

Posts today. I recorded at 11:44. They start at noon, so today was still zero. Posts yesterday: 34 on one model, 45 on another. Last seven days: 19 and 34. Account counts: six and five. The 28-day and 60-day lookups are the same formula. I want to see, without opening her accounts, why she missed the target.

Profile pictures, bios, and an hourly scrape

The second table is one row per TikTok. Username, profile picture, bio, posts per day, followers, views.

The pictures are not decoration. The team in Georgia creates a lot of these accounts and sometimes ships a weak photo. A picture that looks a bit hot from a distance — not everything visible — gets more clicks than a clean full-face shot. Click-through is the job: they found the profile, now they open it. That is underestimated on OnlyFans.

The scrape runs every hour. Faster is overkill and burns API credits. Hourly is enough to see an account move.

The same run checks the bio. We have sent traffic to the wrong Instagram over a spelling mistake. The bot confirms there is an Instagram in the bio, and that it is the right one.

Posts yesterday on one model: 11, 10, 11, 10 — and one account on three. That three is the row I fire to the account manager. Another model was simply posting more. You can see it without asking.

I usually leave the TikTok link field empty. Links increase shadowbans. We were testing the Instagram side the same week: do not even log into the Instagram. Leave it connected in the TikTok bio and see if reach goes up.

Followers, views, and the 12-versus-9 post check

Followers sit next to yesterday's count. One account: 123 yesterday, 651 now. Another: 1,062 to 1,078. Another: 145 to 779 — almost 600 in a day. You will not hold that in your head across thousands of accounts.

Views are a sum of every TikTok on that account, also hourly. One example: 459 total views yesterday, 24,715 now. There are 7-day, 14-day, and 28-day columns for views and for followers. 60, 90, 180 is the same formula. Those were empty because I built the base the day before.

Two post counts sit on the row: what TikTok says the account has, and how many videos we have linked. Match is good. TikTok said 12, we had 9. Either she posted three since the last run, or the scraper failed. That gap is the check.

TikToks roll up into accounts, accounts into models

It is a relational base. The TikToks table is the raw dump. That rolls into TikTok accounts — today's activity. That rolls into models.

Colors on the account: no posts is red, two is orange, five is yellow. Airtable changes the color from the post count.

The miss alert is Make.com — I still called it Integromat. If nothing is up in the first hour and a half I get an iOS notification at 1:30, and the account manager gets the same ping. That is how you run thousands of accounts without watching the table.

A second tool, not in this video: sounds we expect to go viral, not sounds that are already trending. You do not want to post on a sound every other influencer is already on. If we are right 50 or 60 percent of the time and we get there early, there is less competition. The output is a daily list of TikToks each creator should make. This base then ticks whether those got posted, and the numbers on them.

HD no-watermark files, named by view count

The lowest table is one row per video: views, created time from TikTok, comments, shares. The scrape API returns a lot more. I left most of it out. Too much data and you stop seeing the account.

Every new TikTok is downloaded HD, no watermark, into Google Drive. Automated. We repost on TikTok and on Instagram Reels. One client bought the reposting service; a slave account passed his main. The main had about 7,000 followers.

The last table is a follower snapshot at 11:59 every day: posts, followers, total views, which model, which account. One record per account the day I recorded. The graphs need that history.

The hourly Make scenario

If the model is new, the scenario creates her Drive folder once. After that: scrape followers, bio, and link, write the account table.

Then the feed. About 30 clips. For each video: if we already have it, update views, shares, comments, likes. If it is new in the last hour, create the row, download the HD file, tick the no-watermark checkbox, upload to Drive.

The filename starts with the view count so the team can sort highest first. They repost onto our TikToks and Reels, then drag the file into a "posted on Reels" folder so the queue stays clean. Same scenario, every hour.

Airtable Interfaces: one page per model

Airtable Interfaces is a dashboard on the same data. Not enterprise. If you host on Postgres and you want something heavier, Retool. This is the lightweight version.

Each model is a page, generated automatically — thousands of models is fine. Posts today, yesterday, last 7, last 14. Account count. TikToks created by date: 3 on 11 December, 32 on the 12th, 21 on the 13th, 34 yesterday. Six TikToks on that model. A follower bar per account, plus the network total — most of hers sat on two accounts. Views per day.

Those were new accounts: 7,000 views one day, 37,000 yesterday, 26,800 today with no posts yet. One day of history when I recorded. The bars stack as the snapshots land.

Arc workspaces, then onboard without me

I deleted Chrome. Arc lets you put each model in a workspace with its own proxy. The Easel is a board: images, text, notes. You can drop in an embed of the dashboard that updates live, not a screenshot. Same board for crypto prices, Stripe, YouTube analytics, Slack. Not OnlyFans-only.

I was documenting the whole agency so a new Colombian model can apply, get onboarded, get accounts, get her jobs, register OnlyFans, add banking, and start TikTok and Reddit — theoretically with no input from me. We were taking multiple models a day. The content house pivoted. The process stayed. I am more confident we can hold more models because of this base, not because I watch more tables.

I wanted a Scriptable iOS widget on the home screen with each model's daily count. I had not sat down with the docs. The alert that exists: if she misses the daily target, the assigned manager gets a message and I get one. Make can do iOS, Slack, or SMS.

Hundreds of models and thousands of social accounts only work on a system like this. I will put the dashboard in version two of the course for active members. Same tables on Instagram. On Reddit you can track upvotes against clicks over time.

OFM is OFM Insider. Automation is Automation Academy. Email is Newsletter. A one-off session is Consultation. Monthly is Monthly Coaching.

Transcript

Good morning guys just thought I'd share like a couple of interesting things that I built over the past few days they were basically going to be used for managing large-scale um content creation which is what I was expecting to take place inside our villa unfortunately it didn't quite go to plan um but nonetheless like I built out these tools which I still think are very interesting and very useful but now I can see like a lot of other use cases for them um so I'm just going to jump straight into

Them this is a airtable database that is designed to be able to manage hundreds or even like thousands of um social media accounts this one is actually built for The TikTok um however it could very easily be translated across to Instagram even to like some form of like Reddit tracking uh or several other platforms but yeah just to walk you through how it works so the first field uh the first table is like a high level overview um so in here we have three models but again this I just built

This to be like super scalable to give like a high level overview so you could easily plug in hundreds or potentially even like thousands of models up here we have a high level overview how many posts were done today uh it's 11:44 so the girls don't start working for another 16 minutes but post yesterday we have 34 posts yesterday from this model 45 from this model but again I just want to be able to see very very quickly uh like why did this model not hit her like Target criteria next

Field post from the last seven days and here we have 19, 34 and you could easily build this out to go back much further the kind of the data and the logic is already there so you could do like a 28 day um backlog like a 60-day um lookup as well and here we have the total number of accounts that they're managing so here we have six and and here we have five and again just to add in a new model into this system and for them to be tracked you

Would literally just create a new row, type in the name of the model and add in The TikTok accounts that she's working we come down into the next field this is kind of more in detail and we're looking at an account by account basis the first field was looking at an overview of every account the model works on and this is giving the specifics of each individual account um so first off we have like the usernames and then we have the profile pictures the reason that I put the profiles in

Is actually very um specific I want to be able to see every single profile picture of every single account that our model is working on our team in Georgia will make TikTok accounts in a lot of cases and there may be a case where they just don't use a good profile picture and I do think that profile pictures are kind of like a underestimated piece of the puzzle on onlyfans by having a great profile picture on there that that maybe Sparks a guy's interest and gets him to click through onto

That profile if you can really optimize the click-through rate of when someone finds your profile compared to where they click through onto your profile and they either turn into a follower or a subscriber profile pictures are like a great way to do that if you can have a girl who looks like a little bit hot from a distance and it kind of gets them to want to click through uh it just Sparks it Sparks the guy's interest to want to click through onto the profile to see like a full view

Of her um that means you're doing really really well so like you're just going to see like tremendously better results compared to a normal profile picture or even a profile picture where everything's fully visible I have found the ones that work the best are profile pictures that kind of spark a little bit of Interest like this one here is like a great great example actually then we have the bios so this database sorry I should say it's updated to every single hour you could run more frequently than that however I

Think it would just be a little bit too much Overkill I think one running every single hour gives a really clean evolution of every single account you can really see the progression of each account um and you're not going to be wasting like a ton of API credits so this checks the bio every single hour and we also make sure that the Instagram is entered in there correctly of course we don't want to be sending people over to the wrong Instagram with even something as small as like a spelling mistake

Stupid as it sounds we have had issues like that before so again I just built an automated system to verify that the correct Instagram is added on there and to make sure that there is actually an Instagram added on there um and here we have an overview of The TikTok account and how many posts they did each day so today again all accounts are on zero but yesterday we have 11 10 11 10 and then this account only had three so I can very easily build like an automation to trigger

Something to send over to the account manager uh why did this account uh not reach the the target criteria and again we have for another model here so you can see this model is actually a lot more active and posting a lot more content uh than this one here we check if we've put a link in there um generally I don't really use links like uh we find that it just increases Shadow bands and just a little tip for you as well we're testing around currently not even logging into the

Instagram account simply just having it connected in the bio to check if that has an increase on on reach if we come over here you can see the total number of followers of each account and as well you can see an overview of yesterday followers so here we can see this account yesterday had 123 followers and it's now on 651 uh this account was on 1062 is now it's on the 1078 so we're we're seeing progression sorry that wasn't a good example this one is on 145, this one is 779

Today um so almost 600 followers grown in a day just nice to have like a visual a really easy and clear overview if you're managing thousands of accounts there's kind of no other way to do it there's no way that you're going to retain the information in your head then we have a total number of views so this is quite simply a sum where we add together the total amount of views of all of the TikToks that this account has ever created and again this is updated every single hour and

We can also do a comparison of yesterday but you see yesterday this account only had a total of 459 views and now it has 24,715. looking back further through the data there's a seven day views 14-day views 28-day views and as well a backlog of seven day followers 14 day followers and 28 day followers you could do 60 days 90 days 180 days I only Built out this database yesterday which is why all those fields are empty so I'm not too worried about um about looking back six months because quite

Simply we we have haven't got the data here this is total posts this is scraped off of the account itself this is how many posts TikTok says that accounts has and this is how many posts that we have linked in our database so here we have the correct number of accounts and here we do as well here TikTok is saying that there are 12 posts but we only have nine in our database and maybe the model has recently done three posts or maybe the scraper failed on the the last time

It ran uh one of the two, I will I'll have to to check in into that and then if we go down into the next level of the database sorry I should mention as well this is a relational database which means that every single table um each of these tables models is able to communicate and pull data in and send data to tiktok accounts and vice versa and the same TikToks like each table that you see here funnels data up uh to create like a higher level database so TikToks contains

Like all of the raw information all the raw data um and then that funnels up into TikTok accounts which gives us like a cleaner overview of the total activity that was taking place today which then funnels up into models so we go deeper and deeper into the database and each time more and more raw data is kind of just dumped on us and each table um sorts through that as well if you see the color scheme here right now these accounts are red because they have no posts just really simple

But kind of nice like nice to work with features uh when a TikTok has no posts it will be red when it has two posts it'll be orange when it has five posts it will be yellow when it has a plus sorry when it has that many posts it it will automatically change into those colors airtable also offers built-in automation or you could build this through that API or through integromat where you could go and build an automation uh which is what we have done so I actually get a message

On my phone uh iOS notification through the integromat app if a model hasn't posted any content I get like a warning that goes through at 1:30 so if for the first hour and a half of the day nothing has gone up it tells me that the model isn't working and that will also get triggered through to the account manager as well to kind of let them know to Ping the model just to see what's going on this is like completely automated and I built this out as a way to be

Able to manage thousands of accounts in an automated way so the other thing that I've been working on I won't share that today I'll save that for another video but we've been working on it all for identifying I I wouldn't call them trending sounds but I would call them like up and coming sounds I guess would be the right word like the sounds that we expect to go viral based on parameters that we're tracking and so it's kind of it's like it's a work in progress let us put it that

Way um and the reason that we do that is because you want to you don't want to be posting on a trending sound when every other influencer and Creator is already on there you want to be ahead of the curve so of course it means that we don't always get it right but if we're able to guess in 50 or 60% of cases that this is a sound that is starting to pick up traction and we're able to get in there early um it really unlocks the the opportunities what is

Possible with um with TikTok and it means that you're seeing much less competition so we're building out that tool which then goes in each every day sends a list of the TikToks that each Creator needs to create and then again we can track here automatically has that actually been done and we can track all the analytics and results onto a from a specific account level up to like an overview of the um how many TikToks the creators made and the results that we're seeing then if we come across into the

Lowest level of the database this is like a raw dump of The TikTok data so these are actual TikToks this is the number of views that they had this is the created time according to TikTok this is how many comments they have how many shares they have the metadata TikTok actually gives you like a ton of data like with this like scraping API however I actually chosen not to take most of it I think you can kind of overwhelm yourself with information um whereas these are kind of the core metrics

And core data points that I'm interested in um and so yeah and then the only other thing is that we have like a check box here we download every single TikTok that is created um without a watermark and then we store it in a Google Drive and again that's completely automated and I'll show you how that runs as well um the reason that we do that is because we do TikTok reposting which is going incredibly well we actually had I I thought it was quite funny anyway we had a funny

Case the other day where a client ordered our reposting service and one of his slave accounts actually ended up having more followers than than his main account his main account had about 7 000 followers and we were reposting uh and doing like metadata adjustments and uh onto the content that we were reposting and one of the slave accounts ended up having more followers than his mother account uh which anyway I thought was pretty funny um so we we do reposting on TikTok and then we also do reposting onto Instagram reels

Which I'll go through in a second as well and then the final table is just a followers snapshot so this is taken every single day at 11:59 and it gives me an overview of each account it shows me how many posts it has how many followers it has and the total number of views and who and this links into the model and tells me the name of the TikTok account and again I only built this yesterday which is why there's only one record for each account but this will continually update

And every single day we're adding more and more data which allows us to build out more kind of complicated dashboards and graphs which I will walk you through in um in a moment as well so now to go into like the actual automation uh I guess people will be interested to see this so I'll just kind of walk you walk you through step by step uh first off we grab from the um first airtable database and if this is a new model that we've added in we will go and generate

A Google drive folder for her for us to and this is again it's done automatically this creates a Google Drive for us to be able to go and upload content automatically when I mention we download the non-watermark content this will just go and get the Google Drive and um for us to dump the content into so that only happens the first time that we create a model it only needs to be done once every time after that we scrape the followers of the account and we checked the bio of the

Account we get the link all the information that you saw in the second table and then we update airtable next up we scrape the feed we grab all of the videos this returns us I think they're up to 30 clips and then we iterate through each of those um iterate just means to process each of those results one at a time first off we check for duplicates as I mentioned the database updates every single hour so if it's a TikTok that we scraped already previously we will check for a duplicate

And if we register a duplicate we will go down this path and we will update the record with the current statistics of views, shares comments likes uh and everything else if we haven't found that TikTok before if it's a newly posted TikTok and it's been uploaded in the past hour we just go and create a record we then download the HD video we update and airtable the field as I mentioned that says The TikTok uh HD um no watermark has been done we put a little checkbox in there download the

Actual file we download the actual data of that video file and then we go ahead and upload it to Google Drive the format that I used to upload into Google Drive is we put the view count at the start of the uh of the file name so that it can easily be um so that our team can prioritize the highest performing content on TikTok they can prioritize reposting that on TikTok and also reposting that onto Instagram reels again very very simple so we just sort by um highest to lowest sort

Of uh the TikToks with the highest views are going to appear at the top as soon as our team have reposted it onto their onto our TikTok accounts and onto our Instagram Reels account it's just dragged into posted on reels so that they they know not to keep keep reusing it of course we can go back in there in future and always use it whenever we want but it just keeps this table quite clean and this is essentially a queue of um TikToks that need to be reposted so that is

The automation again it runs every single hour um yeah nothing else really to add there so now I can walk you through some of the more interesting parts of what you can do with this data I think some people weren't actually aware but airtable actually has an interfaces um feature where you can build out dashboards based on the data that you have within air table um it's not like Enterprise grade solution like if you were looking for something more complex or if you host your data or natural databases such as

Or Postgres um maybe you want to look at using a tool like retool which I am also a fan of but it's a lot more it's going to be more work to kind of set everything up this kind of gives us a lightweight overview of everything that is taking place within our airtable database so on the side here I can navigate through into each of our models again these are generated automatically so we can literally process a thousands of models uh without any issues um post today post yesterday post last

Seven days post last 14 days total accounts added number of TikToks created daily so this is based on the date so it's from um the 11th of December she posted three videos on the 12th 32 videos on the 13th 21 and again this will update every day so yesterday she did 34 records she has a total of six TikTok accounts linked to her and this gives me an overview of the followers of each of the accounts and the total number of followers within her own network of accounts so I can

See that most of her followers are made up by these two to main accounts and again this will update every day so every day it will add in a new bar I only made this uh yesterday which is why there's only one record in here but this will continue to add on and you can kind of see the evolution of all of the accounts and how how it changes over time so I think this is really really interesting data to have and then finally views generated per day self-explanatory I think

Nothing nothing really for me to add so yeah these are all new accounts seven thousand views generated one day thirty seven thousand uh yesterday and today so far we've had 26,800 with no posts being made yet so yeah I think this is all like just again just to give you like a high level overview of everything that's going on with your account I think this is just super interesting data and as I continue to collect more data there's not much information in our database at the moment as it continues to

Build out I think this data will become more and more useful and there will be more and more um kind of use cases that I can and things that I can build on top of uh in order in order to make more interesting dashboards essentially one other tool I would actually make a full video on this another time but like because we're here I'm just going to show you like another way that I I've actually used this so I'm going to show you this is this is a browser that I've

Switched over to it's called arc Arc I've actually deleted Google Chrome entirely um I won't go into too much detail about it but like I've moved over to this it allows you to have like separate workspaces so you can separate out all of your models and have different proxies on each different workspace so when you're managing a lot of different girls it's becomes very very useful one feature that I'm just going to show you now this is a feature within Arc called like an easel and it's kind of like a

Little workspace so I can drag images I can drag text I can make some notes I can do whatever the most interesting thing at least in my opinion is if I come in into the dashboard if I just come in and take a screenshot it's not really a screenshot though this is a embed I can embed whatever I want on here and I can have it as a screenshot or if I press the play button this will update in real time and so you can drag in like multiple let's say

That I want to like this isn't just limited to um this isn't just limited to uh only fans by the way like you could use this for anything I use this for tracking my crypto Investments I can just copy the prices of hundreds of coins and build out these uh really simple like workspaces I can have high level overviews of all of the models that I want I could inject like stripe revenue for example YouTube analytics um my Slack messages I can embed here and again it updates in real time

Which I think this is such an interesting piece of software again I'll make a video on it in future but um I've been using it quite a lot and not quite a lot I've literally moved over entirely from Google Chrome it's like my browser over onto Arc so um yeah I'm a so yeah that is everything for me um these are to be honest some of my favorite things to to build out is really what I enjoy like doing the most in my job is like having ideas building out like

Automations and systems and processes to um to kind of make make my work easier and and to automate things uh so yeah um I've also been doing a lot of like process documentation talking of automating things like you can of course automate through software and also through um through humans through it through Manpower or woman power and I've I'll share a couple of other tools I'll make another video on this but I've been like kind of documenting everything to make a completely automated onboarding process so that we can have a

New Colombian model who wants to join our agency and start and who would have started in the content house again slight pivot on the strategy there um but I built out an entire process where a model applies to join they are onboarded completely they are assigned accounts they're given their responsibilities of what to do uh the only fans is registered banking information is added correctly through to they are creating TikToks and Reddit content and everything else that we need creating the only fans um bringing in subscribers to us in a

Fully automated process idea well theoretically with no input from my side we'll see if it actually pans out um but yeah that has been like a focus um we've been we've been taking on like multiple models um per day and now I've built this on like I'm feeling I've built built this out I feel much more confident in our ability to manage uh manage a lot a lot more models um sorry one other thing I forgot to mention as well is on this it's really easy to kind of build out

So I actually get an alert um I was trying to build out like an iOS widget so that I just always have it visible on my home screen how many pieces of content each model has built out each day however like I I haven't quite figured that out yet but there's an app called scriptable that you can use to build custom iOS widgets so I just haven't had time to to sit down and figure it out and go through their documentation but the other thing that I did build is if

A model hasn't uploaded content per day or hasn't reached her daily content Target a message will be sent to the assigned account manager I will also receive a notification as well within integromat you can do iOS notifications or Slack notifications or any other as well actually text messages whatever you prefer so I just get a message saying this model hasn't done this today and again it just makes the whole process of if you're going to be managing hundreds of of models and thousands of social media accounts uh I really think

It has to be built on top of systems um such as this so yeah I hope that it was uh that this was interesting and uh and insightful I'm going to add like a couple of more um I've got a few other ideas of things and improvements I want to add on to this system but I'm just going to publish this in um version two of the course so all of our active course members will get access to this dashboard as well and this dashboard can also like this for example

Was built on top of TikTok but it could very easily be translated across to any other social media platform it can be translated onto Instagram very very easily Reddit as well was the other one I thought it'd be quite useful for you could track the correlation of upvotes and how that progresses over time compared to a number of clicks that you get over time particularly if you're doing organic growth I think that's very very you could track some very very interesting data points um through Reddit as well so I have

Also been thinking about translating this across onto um other platforms as well to try to automate the entire in entire marketing funnel essentially but um yeah as always thank you so much for watching any questions please leave [Music]