eslachance/thread.md

## thread.md

      
    Raw
  

              thread.md
            
          
    Inorganic Denials Explainer

Note: This document tries to provide clear and detailed explanations only. It does not and cannot contain ways to "fix" this error because there is no fix. Read further to understand why this is happening to you.
Also, a disclaimer: I do not represent Discord, I don't have access to inside information that hasn't been said in public, I am by no means an authority on this subject but I wanted to clarify things for people that were wondering. Also it's easier linking this than copy/pasting messages all the time.
In recent weeks and months there has been an increasing number of bot owners that have seen their bot denied verification through an email sent by Discord. The wording of this email is:

Based on your bot's growth, it's clear that you joined a number of servers inorganically in order to grow your bot and qualify for verification. Therefore, we cannot verify your bot.
This denial is final and generally cannot be appealed. Your bot will remain in a pending verification state on the Developer portal. We typically do not offer secondary reviews of these denials, but if you do have other questions or concerns please let me know.

TL;DR bullet points:


Discord found there was a clear pattern with spam/scam bots, and they're denying everyone that fits that pattern.
The amount of bots being denied is very small but the people getting denied are very vocal
The real targets of these actions are spammers and scammers in it for the money (they sell the verified bot) and will absolutely lie because they're losing potential profit over this.
The potential collateral damage is a small number of bot developers that randomly, and "innocently", grew their bot in the exact same way as spammers and scammers, enough to be highly suspicious to Discord.

Now let's get into the details!
Why?

The inorganic denials stem from Discord's attempts at fighting back scams and spam on their network. Specifically, fighting against bots that grow in specific ways in order to get verified, after which they are either turned into spam/scam bots, or are directly sold to someone else that will take the same action.
Want an example? Here's one from today as I write this. This is a scam to get people to click a shady link which then infects their computer, steals the user's Discord token, and uses it in other shady actions like raids and scams:

Yes, "verified bots" are a commodity that some people value, since a lot of users automatically assume a verified bot can be trusted. They are sold and bought and then turned into bad bots. Some greedy people will take advantage of that in order to make money.
Why does that affect you and why were you rejected? Because the patterns by which your bot grew were similar or identical to how spam bots are being grown.
How?

It's important to know that Discord has kept most of the details of how they reach the conclusion that a bot is dangerous secret. The reason for this is clear: if they revealed what they were looking for directly, the bad actors would easily get around those specific checks and continue doing their thing.
There have been a number of "hints" peppered here and there however. While I cannot guarantee the accuracy of the information below, it feels valid overall, if not complete:

The bot is added to a lot of "bot farms" or "bot collection servers" where most of the members are bots. These servers can have thousands of "users" (but mostly bots), they are not valid, authentic guilds that want your bot.
The bot is added on specific bot lists and then promoted in certain ways that ensure growth at a specific speed that doesn't trigger "suspicious growth" errors.
The bot's usage patterns are abnormal, indicating the bot isn't being used "legitimately" by actual humans.

Now, again, this isn't a full list, it's not an accurate list, it's just things I gathered from staff discussions. The overall conclusion here is that these are "shady" actions, things that put you at a very high risk of being sold or turned into a spam/scam bot.
Who?

It's important to understand these denials are manual. That is to say, there isn't a cold, unfeeling algorithm somewhere that's doing the checking. These denials are done by a human being, who does a lot of checks of a lot of data points and numbers, and decides after careful consideration that the risk to Discord and their users isn't worth it.
On the one hand, it makes the person doing the bulk of these denials (Cavokarag) is now the target of a lot of anger and hatred, because they're the one that's sending out the email. On the other hand, no one can claim this is an unfeeling automated system that's broken. The system isn't broken, Cavokarag isn't targetting you specifically and isn't just blanket-denying a bunch of bots. Cavokarag is actually studying each of these bots carefully and decided after investigation that the risk was too great for Discord and its users.
"But I'm innocent!"

I'm not here to accuse you of anything, but to explain how you were looking extremely suspect. Let me make a bit of a metaphor, which you'll immediately understand if you've ever watched any cop show on TV:
You get arrested for a burglary, even though you're not a thief and have never stolen anything. You claim your innocence, but the police present you with the following proof:

There are a dozen eye witnesses that saw you around the burglarized house many times in the week before the day it happened.
These witnesses saw you take pictures of the property from various angles, and even from the back alley.
Those pictures of the house and its security features are found on your phone and backed up online.
Tools and equipment used in burglaries (lockpicks, gloves, dark clothing, duffel bags, etc) are found in your vehicle's trunk.
Your bank account received a large sum of money deposited into your account, abnormally high for your usual income. This happened a few days after the burglary.
Your fingerprints were found around the house, especially around entry points.

"But I'm innocent!" you claim. And let's say you are, and you can justify everything with explanations. You were taking pictures of the plants, you were admiring the architecture, your tools are because your cousin is a locksmith and you're into dark clothing, you had an inheritance and maybe you wanted to buy that house by offering immense sums of money to the owner, and that's also why you were actually looking closely and leaving fingerprints...
But even if you can justify every single little thing separately with seemingly valid arguments, it still looks, to a whole lot of people including the police, like you were casing the house and preparing for a burglary, and that's why you're sitting in a room with handcuffs behind your back explaining all of this.
What is the difference here between an innocent person and a guilty one? One thing only: the action that they have, or would have, taken. And in terms of Discord and people selling verified bots, there's no way to actually tell until the time where it happens. So they took the safe route in this case.
With that metaphor being established, the people currently arguing about their innocence on the Discord Developers server can be divided into 2 groups:

The ones that are actually guilty of growing their bot in an inorganic fashion but claiming they're innocent because they're missing out on the cold, hard, cash they would have gotten from selling their bot or executing a scam.
The ones that are actually innocent but find themselves completely unable to prove it because they haven't been paying attention to what happens with their bot and ended up randomly in the same places as all the bad actors.

According to Discord, the vast majority of people being denied for Inorganic Growth are in the first group. It's sad for the tiny minority that is the second group but they're just that - an extreme minority.
How can I prevent this?

So, you don't want to look like you're "casing the house", or in Discord terms, building a spam/scam bot? Here's a few very important tricks.

Don't go on bot lists until you're verified. Being on popular bot lists isn't bad in and of itself. However, it can lead to your bot growing too fast ("suspicious growth" error when you try to verify) and also to a smaller degree, that random people add your bot to random bot farms, just to add more activity to those servers.
Make a bot people want to share. That is, word of mouth is actually the best way to grow your bot in a normal fashion, so that people actually use it.
Make sure it's easy for people to invite, setup, and use your bot. Have an invite command, or add the bot's invite in the help or about command, for example. Make your onboarding and usage simple and clean and clear, without complexity.
Make a specific-purpose bot instead of an all-encompassing one. This is a more general suggestion but it will help you distinguish yourself from the massive flood of existing bots. Make your bot do one thing, and do it right. Groovy doesn't do moderation, Dank Memer doesn't do music, Mudae doesn't do memes. Large, popular bots have specific purposes in mind and they do it well. Nobody wants a "does everything" bot that does everything badly, because there's already thousands of those around.
Pay attention to your guild membership and ownership. Be proactive into making sure your bot isn't added to bot farms, unwanted guilds, things that break the community guidelines, etc. A simple log when joining a guild can give you a lot of information (its owner, number of users, number of online users, sometimes the bot/user ratio, etc). Act upon this and know what's going on!
Make sure your bot is being actually used. A real-life example: I have a bot that does "custom commands" (I call 'em "tags"). In this bot, I know whether people use it because if they haven't created a custom tag, they're... y'know, not using the bot. So I leave any guilds that have not created a tag within 1 month of adding the bot. If after 30 days you haven't used it, then you're clearly not going to, and I don't want you, so I leave you. It's that simple.

There's a lot of owners with the same problem!

No, actually, there aren't. Following the conversations on discord-developers right now it feels like there's a ton of them but it's been clearly stated by the staff that this is actually a tiny percentage of total verifications. Yes, there has been a tightening of the verification procedure, so there are more people than before getting denied. However, for every single bot that's denied with the inorganic growth reason, there will be (probably, I don't know), a hundred bots or more are being verified.
The thing is, bot owners that do get verified won't go on discord-developers and argue for hours about their bot being approved. Of course they don't, because it's what they wanted and expected, so they just move on with their lives. But a dozen people getting denied will absolutely be very vocal in their complaints and cause long-winded conversations as they try to either understand why they've been denied, or lie and pretend they don't know what the problem is and attempt to extract as much information as possible to "get around" the error for their next bot they want to sell to the highest bidder.