Test with Einstein Frequency and Send Time Optimization

Learning Objectives

After completing this unit, you’ll be able to:

  • Analyze email send performance using Einstein features.
  • Configure Einstein Engagement Frequency.
  • Use Send Time Optimization in Journey Builder.

Analyze Performance with Einstein

You’re learning more and more about your customers with Einstein Engagement Scoring and are sending personalized content with behavior-based recommendations. Your content strategy is looking good. But what about your delivery strategy? That’s right—when you send is just as important as what you send. Say hello to Einstein Engagement Frequency and Send Time Optimization. Let’s learn how it works.

Einstein Engagement Frequency

Engagement Frequency uses AI to help predict how often you should send your email communications. Why? Because sending too many emails might mean more unsubscribes (which can impact overall deliverability), while sending too few means leaving potential money on the table. So, adding this Einstein feature takes some of the guesswork out of email sending by identifying the optimum number of email messages to send. Why wait? Let’s activate. 

Activate Einstein Engagement Frequency

There are a few things you need to do to activate and configure Einstein Engagement Frequency at the business unit (MID) level for your account. Follow these steps.

  1. Navigate to Einstein.
  2. Select Einstein Engagement Frequency.
  3. Click Activate Einstein Engagement Frequency.

All good things take time. Wait up to 72 hours to allow the activation process to complete.

Get Configured

Once activation completes, head back to Einstein and Einstein Engagement Frequency.

  1. Click Get Started.
  2. Select how many days of data you want the system to analyze. Select from: 30, 60, or 90 Days to Review. Data filter and email type. 90 days selected
  3. Next, select email classification from Commercial or All in Email Type (includes commercial and transactional messages).
  4. Optimize the suggestions to consider Open Rate, Click Rate or Advanced.
  5. When completed, click Save.

Choose Your Advanced Settings

You can adjust these settings using the advanced option. Let’s review. 

  • Open Rate and Click Rate Sliders: Your organization might put higher priority on clicks vs. opens. If so, you can adjust the Open Rate and Click Rate sliders to assign more importance or the appropriate level of emphasis. The total of both percentages always equals 100 percent. For example, you can emphasize the open rate at 70 percent and the click rate at 30 percent to give opens more priority within the algorithm.
  • Unsubscribe Penalty: You can also choose how to consider unsubscribe rates in your suggestions. This metric can help differentiate between two email frequencies that have similar open or click rates. You can choose to give more emphasis to a lower unsubscribe rate. Adjust the Unsubscribe Penalty slider to assign the percentage of emphasis you give to the unsubscribe rate. Open rate, click rate, and unsubscriber penalty sliders.

Create Data Extensions

Unlike in Einstein Engagement Scoring, where a data extension is automatically created for you, you need to create the data extensions for Einstein Engagement Frequency at first—but it’s as easy as the click of a button. After you finalize your threshold values, click Create Data Extensions. Review the names and statistics for your data extensions and click Create.

Two data extensions, one for undersaturated subscribers and one for oversaturated subscribers, are created in the Einstein folder. Undersaturated subscribers received fewer than the suggested number of emails in the time frame you chose, and you guessed it, oversaturated subscribers received more emails than suggested. 

Screenshot of frequency data extensions.

Using Send Frequency to Target Personas

The beauty of these frequency data extensions is that you can fine-tune your marketing strategy by targeting the engagement personas (that we discussed in the first unit) along with the over- and undersaturated subscribers that have been identified. Here are some possible strategies to combine this frequency data with personas. 

Window Shoppers
Selective Subscribers

Create a suppression list to use with daily or weekly sends to reduce the number of emails.
Attempt a reengagement campaign, then add them to opt-down segment, only sending important communications.
Attempt a reengagement campaign, then add them to opt-down segment, only sending important communications.
Add them to an opt-down segment that only receives quarterly messages to guard against losing them entirely.
Send an exclusive  campaign that provides an offer and makes them feel extra special.
Send a targeted campaign with product recommendations to improve click rates.
Test a campaign with various subject lines to improve open rates.
Send them a win-back campaign with an offer.

Update Configurations

Once you have configured, you can view your suggestions on the Einstein Frequency homepage or dashboard. You can also change the settings by clicking Update Data Extensions. Einstein Engagement Frequency dashboard with Update Data Extensions circled.



The data extensions are refreshed daily with the threshold values that were chosen at the time of the last update.

Einstein Send Time Optimization

Einstein doesn’t stop at send frequency. We take this analysis even further with Einstein Send Time Optimization (STO). Timing is everything, but how do you know when it’s the right time to send that email? Should you send at 9:00 AM local time, when subscribers may be getting to work? Or is 8:00 PM local time better—when they’re likely at home and looking at their phones? Take the mystery out of these questions and let Einstein do the work.

Einstein Send Time Optimization determines the best time to send emails to each of your contacts in your marketing journeys within Journey Builder. Similar to frequency, STO uses machine learning and 90 days of email engagement data to determine the best time (within a 24 hour period) to send a message to each contact.

How It Works

Einstein gathers data for approximately 20 factors, such as the number of emails sent to a contact, the number of emails a contact opened, and the day of the week the email is opened. This feature uses machine learning to assign a weight to each factor according to that factor’s influence on predicting future engagement behavior. Then, Einstein analyzes this data and assigns a likelihood score to each of the 168 hours in a week for each contact. Finally, Einstein uses this information to recommend send times that lead to opens. So smart, that Einstein. 

Activate Einstein Send Time Optimization

Complete these steps for each business unit where you plan to use the feature.

  1. In the Marketing Cloud Setup app, expand Einstein.
  2. Click Send Time Optimization.
  3. Click Activate.

Allow up to 72 hours after activation to complete the data transfer and analysis before heading to Journey Builder.

Use STO in Journey Builder

Once activated, the Einstein Send Time Optimization activity is placed before an email activity in a customer journey. This helps to maximize the chance your subscriber opens the email. JB map with Einstein STO.

How do you do this? Glad you asked.

  1. Drag the Einstein STO activity to a position immediately before the email activity you want to optimize.
  2. Click the added activity.
  3. Select the time period that Einstein optimizes send times for either: Best Hour in Next 24 or Best Hour in Next Week.
  4. Click Done.

Send Time Optimization Dashboard

Feeling curious? Head to the Einstein Send Time Optimization dashboard where you can view suggested send times by day of the week along with hour of the day. You can use these overall predictions in Journey Builder via the Einstein STO activity like we illustrated or simply to adjust the timing of your scheduled campaigns according to trends. 

Einstein Send Time Optimization dashboard.

Now you know the ins and outs of Einstein in Marketing Cloud—your very own marketing genius. What’s more? Einstein features only continue to get smarter. So, even while you’re sleeping, Einstein is working away—accelerating your marketing strategy by predicting your subscribers’ behavior, serving up the products and content they want to see, and sending that content when and how often they want to see it. And for marketers, that’s even better than pulling a rabbit from a hat.

Rights of ALBERT EINSTEIN are used with permission of The Hebrew University of Jerusalem. Represented exclusively by Greenlight.