How to Imporve Campaign Effectiveness Using Best Send Time

The content of the message plays a huge role for successful communication with subscribers. Yet the proper sending time is equally important.

What can it influence?

Increase in Open Rate — if the email lands in the inbox when the user is most active, the likelihood of the message being opened increases.
Growth in engagement — subscribers interact with the content more often (click on links, perform targeted actions) if they receive the message at a convenient time, and the messages themselves are perceived as more relevant and less intrusive.
Reducing the likelihood of emails landing in spam — if the messages are rarely opened, email services may mark them as unwanted. Sending emails at the best possible time helps reduce such risks.

The Best Send Time mechanism based on machine learning takes subscriber behaviour into account and determines the optimal sending time to help you make each newsletter as effective as possible.

Machine learning (ML) is a technology that allows a computer to independently find patterns in data and use them for decision-making or forecasting, without explicit instructions from a human.

How the Best Send Time Works in the Altcraft Platform

The best sending time (one or more days of the week and hours when the message is most likely to be opened) is determined individually for each user based on activity analysis. The platform selects the closest of such time slots and sends the message. This increases the likelihood of emails being opened, may improve Open Rate and Click Rate, and may reduce the number of unsubscribes.

The model uses profile activity data from the last six months to determine the optimal time for sending messages. If a profile has no sending or open history for the last six months, the algorithms will not be able to calculate the best sending time for it.

Read the interview with Altcraft CEO to learn more about how the Best Send Time model was created and how it works.

Activating Best Send Time

To gain access to this type of segmentation, first you need to enable the ML service. To do this, contact your platform administrator.

Next, enable machine learning in the database you plan to work with. For this, open the database editor, scroll down the page, and in the «Machine learning settings» block, click the «Activate» button:

After that, the Best Send Time model will become ready for further use.

By default, the data about the best sending time is updated regularly at midnight according to the account's time zone. If the update time needs to be changed, contact your platform administrator.

The time of the next update can be found in the database settings. If machine learning was enabled recently, wait for the first data recalculation.

How to Use Best Send Time for Mailings

Best Send Time can be used in two ways: through a segment for a regular mailing or through a node in a scenario.

If the newsletter is simple (for example, one-time or periodic, without complex conditions and automatic data updates), a segment may be enough: the system will select the optimal sending time at the moment of launch. If the mailing includes dynamic data updates, a chain of emails, or trigger events, it is better to use a scenario where Best Send Time is configured through a special node.

Sending Mailings: Integrating Best Send Time Into Segments

To launch a regular newsletter at the best time, open the «Data» tab, then in the «Segments» section choose «Dynamic segments» and create a new segment.

Select the profile database in which machine learning has already been activated. Click the «Condition» button, select «From profile» → «Best send time». Then choose the selection condition:

If the «for current hour» condition is set, the segment will include profiles for which the ML service considers the current hour the best time to send the message. For instance, if the segment is calculated at 11:30, it will include those subscribers who should receive messages at 11 a.m.

When choosing the «select hours» condition, you can select several hours. If a profile's best sending hour coincides with one of the selected hours, the profile will be included in the segment.

As an example, the settings below allow you to cover profiles with the best sending time at 9, 10 or 11 a.m.:

After setting up the segment, switch to the «Segment saving» page, enter a name and click «Save».

Next, select the «Mailings» → «Regulars» section and create a new mailing. In the recipient settings, choose «Segment» as the subscription source:

Select the segment created earlier:

Complete the setup and save the mailing. From now on, the messages will be sent in accordance with the Best Send Time model.

Integrating Best Send Time into Scenarios

To launch a mailing with more flexible settings, open the «Scenarios» section and create a new scenario. For sending the message, you will need an element from the «Channels» category:

Add the «Best send time» element before your message. The element can be found in the «Operators» section:

«Best send time» will set a pause before the next element in the scenario, calculated based on the ML service data. This way, the message will be sent to the profile at the most convenient time.

If the best hour has already passed when the profile enters the element, the system will wait for the nearest suitable time to send. If there are no such time slots, the profile will immediately move to the next element.

If a profile for which there is no best sending time data enters the element, it will also be immediately directed to the next element.

Connect all the elements, activate the scenario and save it.

Conclusion

Best Send Time helps to choose the optimal time for sending messages. The Altcraft Platform's ML algorithms allow you to more accurately predict when subscribers will open your email. This solution improves customer interaction and reduces the risk of sending messages at an inappropriate time.