Machine Learning

Likelihood to Engage

The Likelihood to Engage model finds correlation between the past email behavior and the near-future email engagement. The model predicts the probability of the user opening an email in the near future, based on the past email interaction behavior.

Based on the user behavior, CDP defines the likelihood to engage segments as Newbie, Opt-Out, Forgotten, Phantom, Enthusiast, Mainstreet, and Sleepy. These segments help marketers create the most relevant strategies for each group.

Use cases

Enthusiasts - individuals most likely to engage on email

  • Understand these users, their origin, and their likings, and leverage the information in lookalike audiences.

  • Leverage their engagement to create value and content for your brand in social media, surveys, and product ratings.

Mainstreet - individuals with a medium likelihood to engage

Increase engagement by more personalized email content such as newsletter segments or personalized product recommendations.

Sleepy - individuals with a low likelihood to engage

  • Lower contact frequency.

  • Re-engage through special programs to drive engagement.

Newbie - individuals who subscribed within the last 60 days

  • Implement a welcome email series to start a relationship with newbies on a right footing.

  • Maintain proper hygiene in email frequency to avoid spamming and prompting the fresh contacts to unsubscribe.

Opt-Out - individuals who opted out of email

Re-engage the high value customers through other channels such as Facebook, Instagram, direct mail, or online.

Forgotten - individuals who did not receive an email within the last 60 days

Understand the reasons for not contacting such users.

Phantom - individuals did not open an email in their lifetime

  • Reduce frequency.

  • Run special winback campaigns with aggressive email subjects and offers.

  • Identify disposable email addresses.

No-Email - individuals without an email or a valid email

  • Engage through other channels.

  • Inform individuals about the benefits of sharing their emails at the store or on the website.

Target audience

The Likelihood to Engage model predicts the engagement behavior of individuals based on past email interactions. This model scores all individuals in the customer pool.

How does the model work?

For each contact, the model calculates the email behavior features such as send, open, click frequency, volume, and recency.

CDP trains and tests the model on the historical data where the outcome is known. Post training, the model is deployed to predict the future email engagement.

The output of the model is the likelihood to engage for each customer. This probability is segmented in 10 equal-sized deciles, 1 being the most likely decile to yield engagement:

Decile

Description

Email segment

1

High

Enthusiast

2, 3, 4, 5, 6, 7, 8

Medium

Mainstreet

9, 10

Low

Sleepy

NA

Does not engage

Phantom

NA

Does not qualify

Newbies, Forgotten, Opt-outs

Using the Likelihood to Engage model

CDP displays the output from the Likelihood to Engage model in:

  • 360 > 360 Profiles

  • Actions > Campaigns and Actions > Campaigns+

  • Analytics > Metrics

To learn more about how the Likelihood to Engage model can enhance your marketing workflow, contact us.

Did not find what you were looking for?

If this content did not answer your questions, try searching or contacting our support team for further assistance.

Acquia Help

Filter by product:

Machine Learning common questions