Are you looking for clearer signals about which channel to send a message on? Are you getting good engagement from your existing campaigns, but wonder if users might convert more readily on a different channel? Let Channel Decisioning help take the guesswork out.
When a user arrives in a Channel Decisioning send tile in a journey, Iterable analyzes their historical data and sends the message to the channel it determines they’re most likely to engage with. Data is analyzed on a weekly basis to account for changes in channel preference.
Possible uses for Channel Decisioning include:
Alerting users to time sensitive or in-the-moment messaging on the channel where they’re most likely to see it. For example, a retailer might inform customers that an item they’ve been watching is back in stock in a limited quantity, so they see the message before it sells out.
Redirecting users to an application. For example, a news agency might steer users who respond best to SMS messages to a Breaking News app for more information about a story the SMS message highlights.
# In this article
# Channel Decisioning considerations
To evaluate channel preferences for users in an Iterable project, the project should have at least three months of campaign data and a minimum of two active channels. Currently, Channel Decisioning (formerly, Channel Optimization) supports email, SMS, and push notifications.
As with other journey send tiles, when you add a Channel Decisioning tile to an existing journey, you’ll need to create new campaigns from the tile–you can’t copy or move existing campaigns into a send tile. Be sure to update any references that exist in your project from the previous campaign IDs to the new campaign IDs.
Channel Decisioning is supported for use with:
- Marketing messages
- Email, push, and SMS
- Journey campaigns
- Rate limiting and Quiet Hours
Channel Decisioning is not supported for use with:
- Transactional messages
- Blast campaigns
- Triggered campaigns outside of a journey
- Send Time Decisioning
# Channel selection and fallback options
When you’re setting up a Channel Decisioning tile in a journey, you can tell Iterable which channels you want to be available for optimization and how you want to prioritize those channels when a user’s preference can’t be determined (referred to as fallback options).
Consider how the fallback options that prioritize sending first to email, then to push would be handled under various circumstances.
# Fallback options and new Iterable projects
While Channel Decisioning can be used with new projects, it’s best to use it when you have at least three months of campaign activity. Until then, there’s not enough data for Iterable to determine channel preferences, so it sends messages using a randomized algorithm, not the fallback settings. Once there’s sufficient data to analyze, Iterable starts using fallback options, as described in the following sections.
# Fallback options in established projects
If you create a journey in a project that has several months of historical data and users who frequently open email and push campaigns, Channel Decisioning sends campaigns to users on the channel Iterable determines they’re most likely to engage with.
Fallback options are used only when a user’s data is inconclusive (for example, when a user engages with different channels with similar frequency) or incomplete (for example, when a user is missing a necessary data point). For our example, which specifies sending messages first to email, then to push, if a user is missing an email address, Iterable sends the message to the next fallback option, push.
# Fallback options with new users in established projects
When a new user is added to an established project that uses Channel Decisioning, Iterable sends messages to that user according to the fallback options. Once there’s enough historical data to make the user’s channel preferences clear, Iterable stops using the fallback options and sends messages on the channel the user is most likely to engage with.
If you plan to experiment with campaigns that new users receive, it’s best to set up those campaigns outside of the Channel Decisioning tile to ensure that the data you track as part of the experiment reflects the same channel, consistently.
# Opt outs and Channel Decisioning
Like other Send tiles in Journeys, Channel Decisioning doesn't need to account for user opt outs, because the Iterable platform automatically prevents messages from being sent to users on channels from which they’ve opted out.
# Experimenting with Channel Decisioning
The campaigns you create inside of a Channel Decisioning tile are like any other campaign you create—you can include them in A/B experiments just as you would any other campaign.
If you plan to experiment with campaigns in a journey that uses Channel Decisioning, set up those campaigns outside of the Channel Decisioning tile so you can control the channels that are used.
# Want to learn more?
For more information about some of the topics in this article, check out these resources. Iterable Academy is open to everyone—you don't need to be an Iterable customer!
Iterable Academy
Support docs