Also, Alexandra urged the audience to use data science by using the lifecycle touchpoint from the Klaviyo library, noting that one customer sent a flow with 9 emails and still got a 32% Open Rate on the ninth email.

Predicted Customer Lifetime Value and Predicted Gender also enable much more targeted flows.

She underscored the importance of continuing to test. Even if you just send to a small sample set an email that reflects an untried idea, you will statistically grow revenue by testing. For example, optimize your series length by testing and adding an additional email or two while monitoring engagement.

Finally, Alexandra recommended tagging profiles with the first purchase date. One flow tags each profile with the specific date of the initial purchase, then a second flow follows up with specific content to celebrate the anniversary of the first purchase. Depending on how many years are involved and whether additional purchases have taken place since, the content and discount change.

Another example of tagging profiles includes adding categories to target initial purchases, continued purchases and categories abandoned altogether. Flows can also tag customers into particular loyalty tiers: customers who have purchase X times, X+Y times, etc.

To design a flow, consider your goals, triggers, customer personas and benchmarks for open rates, clicks and conversions.

Besides smart sending, tagging recipients as “in flow” (and then running a separate flow to remove the tag) will prevent a prospect from receiving too many emails.