We have mentioned before that segregating email marking lists is the quickest route to better returns. Research shows it will improve, amongst other returns, open rates, completion rates and unsubscribes. It also ensures that you don’t waste a window by sending a general, untargeted offer.
There are a number of straightforward ways you can split your email marketing lists, such as by gender, age and location. Segregation at this level allows you to target your emails to a certain extent.
Most email marketers, and this should include you, separate new subscribers. By keeping them on their own list in the early stages you can send offers that encourage them to complete, developing loyalty. Further, you might want them to buy a basic product in the anticipation that they will expand their purchasing and buy other items based on the original one.
This is basic email marketing. For better returns you need to move onto different criteria. The next step is to use returns to predict subscriber behaviour in detail. Using purchasing history to highlight differences ups the stakes as it works on feedback, so becoming a sharper tool over time.
You should keep a record of the activity of each individual subscriber; for instance, opens, clicks, leads and purchases. Once you have established a database you will find that you will be able to segregate them into groups where they have similar behaviour.
There are other pointers. For instance you can classify customers by their monetary value, their frequency of purchases or completion and how long it has been since their last purchase. Once you have a sufficiently detailed database you will be able to target you next campaign with greater precision as predicting their likely response will be more precise.
Prediction is based on what others with similar characteristics have done in the past. If they have factors in common then email design, mode of address and nature of offer can be similar.
A person who has a lot in common with others will tend to react in the same way. If you can predict the product that a subscriber is most likely to buy, and possibly at a particular time, then most of your work is done, apart from choosing the best price for the offer.
Age does not describe who might buy a medium range greenhouse. Subscribers might well have been pleased with the cheaper, budget greenhouse bought from your gardening supplies company a year ago, just before the start of the season. It provided all the functions promised but is somewhat limited in its range.
You find that those who had bought the item fall into two distinct groups: those who respond only to bargain basement offers, and those who steadily upgrade their equipment.
There will be a trigger time when those in the second group will be vulnerable to an offer of a replacement as the budget greenhouse is not suitable for the serious hobby horticulturist. You will be able to predict the best time to make an offer going by how those with similar characteristics performed in the past.
Collect data and refine, and when you’ve done that, refine again.
