Tips to avoid automated DMs from chatbot getting flagged as spam

[andevo]

Hi,

I am extremely new to the Twitter API having only gained access yesterday.

I have made a chatbot that sends a greeting and a random fact in a single message in response to every DM received, as a test. I expect a lot of followers to try to use the chatbot at once in bursts (>1000 / 24 hr). Before we start using it I would like to confirm a few things…

I have also read the automation rules and I’m wondering if part of the reply message (the greeting) is the exact same and there are only a limited number of phrases, can the app and/or account get suspended for sending similar things to many people and is this considered “spamming”? It seems like chatbots would be prone to saying the same thing to many people by definition. Noting that there won’t be misleading links/sensitive media/asking for private info etc. I may include a link in the DM reply (for example an invite to a Discord server later, though) if this is allowed.

In addition I had a question about the policy here (from Twitter's automation development rules | Twitter Help):

• you provide a clear and easy way for such users to opt-out of receiving automated Direct Messages, and promptly honor all such opt-out requests.

In such a case where there’s no opt-in and people just message the account when they want to interact with the chatbot, how can people opt out?

Thanks!

[suvpen]

As far as I know, the Twitter spam filter is adaptive based on account variables like followers, account creation date, number of tweets, etc. Also, it ruled all API hits from all twitter clients including apps from the developer. I suggest starting slowly, don’t burst all at once. Once your account scores passed the filter threshold, you can go all out.