What Not To Do:

Introduction

• These are based on our current understanding of the X algorithms and the /Reach quality score, which does its best to keep to what is good practice on X more broadly
  
• The general idea is to avoid looking like a bot and to encourage real connection and engagement
• These are rules of thumb—sometimes there is a circumstance where you’ll want to do one of these do-nots! Be aware you might take a hit either with the algorithm or /Reach level/earnings, but use your judgment. It’s your account!

No Bot Use!

• X is intended for human beings and brands to use to communicate with each other. From their rules: “You may not use X’s services in a manner intended to artificially amplify or suppress information or engage in behavior that manipulates or disrupts people’s experience on X.”
• This means no botting. We’re defining botting as any automated actions that are intended to look like a human took that action. X allows automated accounts that are designed to help people who use the platform—say, a weather bot or a bot that helps you book an appointment, but they’re clearly labeled as such and linked openly to a real human’s account.
• X HATES bots that masquerade as humans, and is constantly trying to ban them! So it’s important not to use bots, and not to act like a bot.

How To Avoid Using Bots

• Alongside literally not signing up for tools that offer account automation as a service and not connecting their apps to your X account, it’s important to keep a close eye on what applications you’re connecting your X account to and what permissions they ask for.
  
• X allows you to connect your account to apps that use X’s API to do all kind of things—for instance, /Reach is one of these tools. This functionality is really powerful and allows you to, for example, prove that you’ve reposted and followed the X account for an upcoming mint.
  
• It’s important to check the permissions that app is asking for; avoid apps that want to receive permissions to post, like, follow or commit any other action on your behalf. If you do need to use these apps, revoke permission on X as soon as you’re finished using them.
  
• Even if an app is offered by a reliable company, it’s still risky to keep an active connection if it allows these permissions, as apps do get compromised and a malicious actor could use connected accounts to spam, post malicious links, and other antisocial behaviors. It’s not only embarrassing when this happens, it can get your account banned from X!
• It’s a good idea to regularly review permissions for apps and remove the ones you don’t use—and, if you see something you don’t recognise, revoke immediately and change your passwords in case your account has been compromised.

How To Avoid Looking Like A Bot

• Alongside not using bots, you’ll need to avoid genuine behaviour that makes you look, to X’s algorithms, like a bot.
  
• Since X’s moderation teams don’t have the ability to directly observe every one of us using the platform, they implement rules and systems that they hope will catch automated accounts, and not real users. They’re constantly adjusting, but if you’re acting like a bot, you might get caught in the dragnet—and your account can be deboosted (your content shown to fewer people) or even suspended temporarily or permanently!
  
• You also need to avoid looking like a bot to fellow platform users—the report feature, which any account holder can use to flag a suspicious post or profile, is an important tool X uses to weed out bots and spam.
  
• In addition to being mindful of your own account’s activity, you should also avoid bots that artificially inflate engagements with your profile and your posts.

Botlike Activity Don'ts

• Looking first at activity—avoid making excessive amounts of posts, reposts and replies. There’s no precise number available, but more than a few hundred total per day risks triggering X’s anti-botting measures.
  
• The same goes for likes: best guess is a few hundred are okay, but doing thousands might look automated.
  
• For replies, content matters: X uses algorithms and machine learning tools to assess relevance and quality. Avoid low-effort and/or irrelevant replies; these risk triggering anti-bot measures as well as attracting reports.
  
• Short replies, like a single word or emoji, are also risky—a sentence or more, ideally with good spelling, punctuation and grammar, looks much less bot like.
• Non-post activity also matters—mass-following or unfollowing, or adding hundreds of accounts to a public List, can flag your account and attract reports. Keep these to a few dozen changes a day. The same goes for sending direct messages—even messaging someone you’ve not spoken to before can trigger your account being locked until you pass a CAPTCHA. Keep new message reach-outs to a minimum.

Botlike Engagement Don'ts

• You’d think nobody would penalise you for things other accounts do to yours, but you’d be wrong! Loads of scammers and bot accounts will pay for unethical services to inflate their account’s apparent prestige. For less than $100, they can get networks of thousands of bots to follow their account and engage with their posts. X hates this, and this will trigger their algorithms to make your account way less visible or even flag it for rules violations. It’s incredibly tempting, but never use these services!
  
• If you see a post with lots of low-effort engagement, or discover an account with what looks like a botted follower count, never engage with it as it means X’s systems might think you’re part of the horde of bots! Unfollow these accounts if you discover you’re following one.
  
• Sometimes, malicious or aggrieved individuals will target the accounts of people who they don’t like—including spending money to bot followers and posts of their target in the hopes that they’ll be flagged and suspended.
• The more your timeline is full of human, natural interactions, the more natural the account as a whole will appear to humans and algorithms—meaning you’ll have more leeway with occasional botlike posts and activity.

Avoid Looking Like A Spammer/Scammer

• Avoiding looking like a bot isn’t enough: from the perspective of the Web3 community on X, ! While many spammers and scammers are also bots, there are some additional measures you can take to make it less likely that systems, or fellow platform users, will flag you.
  
• A spammer violates the basic understanding of social media—we’re there to get interesting information and engage in conversation with legit people and organisations. While a lot of this is commercial, there’s a difference between connecting with your customers and, for example, posting a link to your NFT drop in an unrelated conversation.
• A scammer will use a number of social and technical techniques to trick users into clicking a malicious link; that can lead to your device, X account and other social media being compromised. Scammers also try to get users to transact with wallet drainers.

Spammer Don’ts

• If you’re promoting something, like a drop, airdrop-farming referral links or other farming content, context matters in your replies! There are plenty of X threads that ask outright for replies like this—posting there works better for your engagement. Self promotion in other threads really annoys the community; they might flag your post or account as spam.
  
• Another reason context matters is that the airdrop farm you signed up to wants to promote their own messaging in your content, not be associated with spammy posts. Good, in context replies will get engagement from your fellow farmers and potentially boost your airdrop, as well as growing your account. 
  
• Avoid copy/pasting identical replies. Systems and users see this as spam, and it looks low-effort. Customise the text of every post, and if you can’t relate what you’re promoting to the thread, don’t post there!
• It’s also advisable to make sure your own timeline isn't a constant repetition of posts about the thing you’re promoting—have some variety. 

Scammer Don’ts

• Obviously nobody taking this course would deliberately scam someone—but even for law-abiding X users, it’s important to avoid associations with scammers where possible.
  
• If you share a malicious link, or even share someone else’s scammy post, you can attract flags from community members and put your account at risk. Check posts carefully! Make sure you’re not sharing content from an impersonator of a big account, and if you find out that you’ve shared a scam post, remove your share.
  
• Checking posts and links is especially important as scamming is one of the most sensitive violations of X’s rules, and if your post with a bad link gets mass reported, you could wind up with your X account suspended or banned. 
• Avoid following scam (and spam) accounts, and if one follows you, block it.
  
  

Looking Spammy/Scammy to Humans

• X systems and human X users are always on the lookout for obvious signs of scamming and spamming—there are also some more subtle signals that sophisticated fellow users look for.
  
• Your account should have a healthy mix of original content and reposts! All-repost accounts look spammy to the human eye. (While an account can have a significant amount of reposts before looking spammy to humans, algorithm analysts like Alex Finn are also of the view that reposts in general are penalised by the algorithm.)
  
• If you’re earning $reach by completing missions, you’ll have some reposts—that’s why it’s important to also use your /reach connected account for regular activity! For a balanced account, have a good mix of original posts and reposts, and space out your /Reach mission shares throughout the day.
  
• Conspiracy, strongly controversial, or offensive content is a big turnoff to human readers. Use moderate language, back up assertions, and cite facts and sources.
• Don’t spread disinformation, either. If there’s a Community Note on a post, the original post will likely be seen by most human readers as false. Because disinformation and conspiracy content is largely spread by bots and scammer accounts, sharing this sort of content can undermine your reputation.

 Avoid A Timeline Of All Farming Posts

• Airdrop farming on X (tokens like $PARAM, etc) are a great way to earn while rapidly building your account by connecting with other farmers. But an account filled entirely with airdrop posts (particularly original posts) that are all airdrop farming content looks like spam!
  
• Keep the airdrop content mostly to replies, and even do those in moderation. Again, original content, with variety and congruence that make your account feel real and human make it more likely that the well-connected and interesting accounts will follow you, not mute, block or report!
• If you were to complete /Reach missions in the brainless way many people approach social airdrop farming, this would also apply to /Reach—but if you do your mission replies consciously they don’t look like airdrop farming at all!

Avoid Getting Hacked

• If your X account gets hacked and used for spamming and/or scamming, both X systems and human users will take a long time to trust and boost your account again!
  
• Set up robust security for your account—multi factor authentication (ideally not using SMS as one of the authentication methods, as SIM swapping is rife among scammers), strong passwords not used anywhere else, etc.
  
• Malicious websites that come up when navigating the world of crypto do more than drain wallets—hackers are just as glad to get control of a crypto-focused X account so they can post their bad links from trusted accounts! Even if you encountered the website via Discord, a web search or another app or user, the hacker may try to use keyloggers, etc to grab X account credentials. Never re-use passwords anywhere online, as hackers are smart enough to try passwords across platforms!
• A password manager is a good way to manage strong, secure and unique passwords without needing to remember them—using one for your X account is a quick way to strengthen security.

Discussion Questions

• Find three spam posts and three scam posts (look in replies under the posts of, for example, Yuga Labs/Bored Apes, crypto newspapers like Coindesk, and you’ll find some good examples).
  
• How did you identify that they were spammers/scammers? What caught your eye first?
  
• Check the timeline of these accounts, including both the posts and the posts and replies tabs. From looking at the timeline, can you tell whether these accounts are entirely spam/scam accounts? Why or why not?
  
• Review your own account. What, if anything, are you doing that is likely to make your account look spammy or scammy to X systems and to humans? How will you change your activity to avoid these?
• Review your account security, and your own security practices when viewing X. What are you doing that is risky? What will you do to improve your security?