Let’s suppose that you have two different incentive programs, but you don’t want to allow anyone to take part in both. In other words, you want to make sure that if someone has entered either of the two programs, they cannot enter the other, ever. Proceed as follows:

1. Create both programs and add an Exclude Segment node after both entry nodes.
2. Add an Update contact list node after each of the two Exclude Segment nodes, set the same contact list name for both, and choose the Add contacts to the contact list option.
3. Create a segment and choose Based on contact list. For the Source list select the contact list named above.
4. Select this segment in the Exclude Segment nodes.

Now, each contact who enters either program will automatically be added to the contact list, which will update the segment and exclude them from entering the other program.

If a large-volume transactional program contains a Wait node which is set to a specific time (such as “in four days at 8:00”), it can lead to performance issues since all contacts will be processed individually. To improve the performance of these programs, you can add a Update contact list node before the Wait node and move the rest of the program into a new program with a Recurring filter entry point which runs to the same schedule (e.g. every four days at 08:00).

Proceed as follows:

1. Create the transactional program which ends in a Update contact list node to collect the contacts who would otherwise enter the Wait node.
2. Set a contact list name for the Update contact list node and choose the Add contacts to the contact list option.
3. Create a segment and choose Based on contact list. For the Source list select the contact list named above.
4. Create a program with a Recurring filter entry point and select the segment created above. Schedule it to run every day at 23:59,
5. Add a Wait node after the segment with the same settings as the original Wait node and continue with the rest of the program.
6. Set the participation in both programs to allow contacts to enter them only once ever.

Now, at the end of each day the second program will pick up all the contacts who would have reached the original Wait node and queue them, but they will be processed as a batch and proceed must faster through the program.

Suppose you have a program which tries to reactivate defecting customers. You could have a branch for highly valued customers who usually spend a lot. If they do not react to emails, SMS or push messages, you may want to call them via phone and see if your sales representatives can reactivate them. Proceed as follows:

1. Create a daily recurring program designed to reactivate contacts.
2. Place a Update contact list node at the point where you would like to make a call to the contact and choose the Overwrite existing list with contacts who arrived at this node (even with single contacts) option.
3. Schedule a daily contact list export to run about an hour after the contacts are expected to reach the Update contact list node. This will export the list of contacts who entered the node on that particular day. You can set the export to send an email with the list of customers to be called to a sales representative.

Use the predict last session date and predict last purchase date fields to identify leads who have visited your site but not yet bought anything.

Combine them with predict last session time spent and predict last session categories to gauge their level of engagement and the products they are interested in, and target them with incentives accordingly.

Use the predict last session categories, predict last abandoned categories and predict last purchase categories fields to segment contacts you know to be interested in these products and target them with stock-clearance and back-in-stock campaigns, as well as cross- and up-sell campaigns.

How many of your shopping carts are abandoned? Use the predict last abandoned date and predict last abandoned categories fields to trigger timely reminders to customers of what they left behind.

Combine it with predict last abandoned total price to offer different incentives based on total cart value. In this way you can recover up to 20% of your abandoned purchases!

If you don’t want to keep sending recovery campaigns for items that have been left in the cart for a long time, check if the predict last abandoned products field has changed or not.

The fields predict last session date and predict last purchase date give you a good indication of when customers are beginning to defect. Combine them with predict last purchase categories to offer targeted incentives.

See our example on how to use the web behavior fields to create a great lookalike audience.

You can also use these fields to personalize sections of ad hoc email campaigns with dynamic content that will increase click-through-rates and conversions. For example:

• Use the predict last session/abandoned/purchase categories fields to select the right content based on their last preference.
• Use the predict last abandoned session/cart/purchase fields to segment your audience according to their level of engagement and treat them accordingly.

Targeted lists can be generated based on the data collected through our JavaScript API and subsequently exported for use in other marketing activities. For example:

• A high value for predict last session time spent with no corresponding predict last purchase date may indicate problems with your shopping experience.
• A high predict last purchase total price may merit one-to-one interaction from your Customer Services, such as invitation to a loyalty scheme.
• You can also match purchase activity with email responses and consider targeting individuals through other channels for better conversion.