Understanding donor channel recommendations

Know your donors' channel preferences can help you meet each supporter where they are.

Some donors have a preference for a particular channel (Mail, Email, SMS, or Phone) and when asked on that preferred channel are more likely to respond. Some channel preferences are obvious (if you do not have a donor’s phone number you can’t reach out via phone, or if a donor has opted out of certain channels like email), but some are more difficult (if a donor has given to both email and mail outreach, which is best?) 

Favorite Channel

The Favorite Channel field in Dataro is a simple calculation of the channel that a donor has most often donated through. 

Channel Recommendation

Dataro’s Channel Recommendation is an AI model that predicts which channel a contact will contribute the most value through. The Channel Recommendation may often align with a donor’s Favorite Channel, but not always. In general, we recommend using the Channel Recommendation field for segmenting targeted audiences, as it provides a more robust analysis of donor behavior.

How the Channel Recommendation model works

There are three unique aspects of Channel Recommendation, compared to Dataro's other propensity models.

1. Multiclass classification

The first novel aspect of this prediction is that it is a "categorical" prediction instead of a binary prediction (i.e. we are predicting Mail / Email / Phone / SMS vs. True/False). In the ML world this is called a ‘multiclass’ problem. The output of the model is a probability distribution across the four potential classes. For each contact we are publishing the class with the maximum probability.

2. Defining the model target

As mentioned above, if a donor has given through multiple channels, how can fundraisers decide which is best? We have defined the model’s 'target' (i.e. what we are training the model to predict), as the channel through which a donor can be expected to contribute the highest total value in the next twelve months.

What does this look like in practice, as we model a real donor database?

  • For example, if a donor has donated $10 6 times via SMS, but given a $75 mail gift, the model will recommend contacting this donor via Mail, while noting that the donor’s Favorite Channel is SMS.

  • If a contact in the training set doesn't make any donations in the next 12 months from the reference date, then Dataro will look at which communication channel they have the highest number of responses to in order to determine their recommended channel.

3. Model Factors

The factors for the model are relatively similar to the other propensity models, however, we also include additional binary factors about contactability: has_address, has_mail, has_phone.