[Reply] Setting up Rules to automate common tasks

Rules are only available on our Business Plan. More information on our plans can be found here.



Rules are a powerful way to automate common tasks like tagging, moving, assigning, or setting the status of, incoming conversations. In their simplest form, you'll setup a filter to match specific incoming conversations that you can then apply an action to.

Setting up Rules

To get started, click on Settings on the left hand side of your dashboard and then select the Rules tab.


Click New Rule


The first thing to do is select the filter you'd like to use. Filters allow you to target specific conversations, based on criteria such as keywords or language, that you can then apply automated actions to. Here's a run down of the filters you'll see:

  • Contains: Filter conversations by keyword, hashtag or phrase. This will search the body text of a thread item, e.g. a tweet or Facebook comment. This is one of our most popular filters.
  • Contains (Include Links): Similar to the Contains filter, but also searching within links. For example, you could use this filter to search for keywords within links like UTM parameters, or to filter conversations that contain shortened links, e.g. buff.ly or bit.ly.
  • Contains Only Mentions: Filter conversations that only include mentions, but no actual text content. For example, when a Facebook user only tags a friend in a comment on a post.
  • Begins With: Similar to the Contains filter, but only searching the body text that starts with a given keyword or phrase. For example, pre-populated Tweets or other messages.
  • Recipient Is: Filter conversations sent to one of the social accounts you have connected to Buffer Reply.
  • Language Is: Filter conversations based on detected language. Reply currently supports the following languages: Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Hebrew, Hindi, Hungarian, Indonesian, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swedish, Thai, Turkish and Urdu.
  • Type Is: Filter conversations based on type. This includes Twitter @mentions, direct messages and searches, Facebook comments and private messages, and Instagram comments.
  • Is In Reply To Tweet: Filter conversations that were in reply to a specific tweet (based on the tweet's ID).

Please note, the Contains filters do not currently support comma separated lists. Rules containing multiple words will only find conversations with an exact match. For example, a Contains filter with "shipping, ships, delivery" would only find conversations with that exact phrase, it would not find conversations containing any of those keywords. We hope to include support for more advanced searching soon!


The next step is to choose an action, i.e. what should happen to conversations matching the filters you've set? Here's a run down of the actions you'll see:

  • Assign To: Assign matching conversations to a specific team member.
  • Move To: Move matching conversations, either to the Spam folder, or one of your custom folders. You can learn more about custom folders here.
  • Tag With: Add a tag to matching conversations. Please note, only existing tags can be added.
  • Set Status To: Set the status of matching conversations to Closed.


Once your filter and action have been defined, click Save Rule.


Automatically assign conversations containing a specific keyword

You might like to assign specific enquiries to a certain person. For example, you might assign all “shipping” related enquiries to Katie. In this case, you'd have a rule set up with the filter “Contains = shipping” and the action “Assign to = *your team member*”.

Automatically assign Spanish conversations to your Spanish speaker

You might like to assign conversations in a particular language to a person on your team who speaks that language. In this case, you'd have a rule set up with the filter “Language is = Spanish (or other language)” and the action “Assign to = *your team member*”.

Automatically move conversations to a custom folder

You might like to segment your conversations by type. At Buffer, we have a dedicated custom folder to hold comments from both Facebook and Instagram. In this case, we have a rule set up with the filter "Type Is = Facebook Comment" and the action "Move To = *your custom folder*".

Automatically change the status to Closed

You might like to automatically close conversations that only contain mentions (i.e. no actual body text). In this case, you'd have a rule set up with the filter "Contains Only Mentions" and the action "Set Status To = Closed".

Automatically move conversations to Spam

Maybe the name of your brand or organization is real word. For example, at Buffer, we’ve seen tweets about Netflix taking a while to “buffer”. Since these aren't relevant to us, we have a rule set up with the filter "Contains = netflix” and the action “Move To = Spam”.

Conversations from social contacts that you have  muted will also be automatically routed to the Spam folder.

Order of Rules

At the moment, it is not possible to combine multiple filters and actions. For example, if you wanted to automatically assign and tag conversations based on keywords, you would need to create two separate rules.

One thing to bear in mind is that rules are generally run top to bottom, which is sorted by creation date, oldest on top. If you have conflicting rules, the one created most recently (i.e. sitting closest to the bottom of your list) will take precedence.

Example A

You have the following two rules:

  1. Filter “Contains = shipping” and action “Assign to = Hannah Voice”
  2. Filter "Contains = shipping" and action "Tag With = shipping"

These rules do not conflict, so matching conversations would be assigned to Hannah Voice and tagged with shipping.

Example B

You have the following two rules, where the first one was created before the second:

  1. Filter “Contains = shipping” and action “Assign to = Hannah Voice”
  2. Filter “Contains = shipping” and action “Assign to = Darcy Peters”

These rules are conflicting. Matching conversations will be assigned to Darcy Peters, since that rule was created most recently and sits closest to the bottom of the list.