A guide to segmentation: 4 techniques for effective email campaigns

But what does it mean exactly? How does this translate in practice? The answer lies in segmentation, that set of activities that is useful for splitting your database into relevant groups. After we explained how to increase sales through automatic emails, we move on to discover the fundamental activities in segmentation.

Before separating your contacts and creating custom emails, an in-depth analysis of the data should be performed, in order to understand what was collected, the amount, quality, type, and whether they are updated or obsolete data. Only by knowing the database’s value can you determine the best type of segmentation.

Secondly, the marketing strategy and customer profile with which you want to communicate must be defined. Data analysis helps to understand what information is needed, if the data in our possession is sufficient or whether it is necessary to acquire new data. A complex activity, which MailUp allows you to face thanks to the platform’s advanced segmentation functions, along with the help of questionnaires, digital forms, and cross-channel acquisition campaigns.

After defining the objectives and analyzing your database, you’re ready to segment your contacts. Let’s see some examples.

Segmentation based on personal data is the easiest way to divide contacts: gender, age, address, are just some of the useful data for creating clusters. This data is easily available, often provided by the user during the registration process.

If you want to create a data collection form and insert many fields, remember to make the only mandatory request field the user’s email address, thus leaving the user the choice of what and how much personal information to share with you. Because at this level of interaction, your interlocutors are less inclined to share too much information about themselves.

Therefore, make sure that the information you are requesting is necessary for your market strategy. Excessive demands can lead to minor conversion rates.

Segmentation based on behavioral data focuses on actions that users perform, for example, in response to an email you sent or from browsing on your site. This kind of segmentation helps to understand which stage of the conversion process users are located in.

If we consider the different interactions that a recipient may have with an email, we can identify four clusters based on four behaviors:

• the subscriber opens the message;
• the subscriber does not open the message;
• the subscriber opens and clicks in the message;
• the subscriber opens and does not click in the message.

If you’re wondering what your contacts’ level of involvement is, you can calculate the fidelity index with a simple calculation: compare the number of messages opened with the total number of messages sent. Then compare the data with these three categories:

• loyal subscriber, with an index greater than or equal to 75%. In this case, reward their loyalty with special content, a free service, or a special discount;
• uncertain subscriber, if their fidelity value is between 25% and 74%: they are potentially loyal customers, to convert by optimizing the content, subject, timing of sending, and informing them that loyalty to your communications will be rewarded;
• unloyal subscriber, with a score below 24%: these contacts are difficult to reactivate, and there is no single motive behaind their scarce involvement; mainly, their perception of the message’s value is what stops them from opening the messages. Leveraging promotions and special offers could be a double edged sword, creating fidelity to the offer rather than the company.

It is all the information related to buying behavior, in both physical and online channels: type of product chosen, purchase frequency, number of orders, total value of purchases, preferences for brands, colors, and more.

Navigating through and using this information is not easy. Our advice is to analyze the data that makes your database particular and combine them to develop an effective segmentation. Here are some examples:

• special offers, for customers who clicked or purchased repeatedly on products of a specific brand;
• email to recover abandoned carts, for users who have selected products without ever completing the purchase;
• cross-selling campaigns: if a user bought a certain color dress and bag, send an email (with or without a discount) to suggest related products of the same color;
• up-selling campaigns: propose a selection of products according to the average the customer spends;
• pre-sales campaigns: anticipate seasonal discounts with an email to the customers who have made a certain number of orders in recent months.

RFM analysis is very popular in marketing and email marketinga: a sophisticated example of a segmentation that employs predictive statistical methodology on behavior, based on three variables, thanks to which you can associate customers with a score for each metric request:

• recency: the date of the last purchase made;
• frequency: the purchasing frequency;
• monetary: the average expenditure in a given period of time.

The three principles of RFM analysis are:

• customers who bought recently are more receptive to new promotions than customers who bought further in the past;
• regular customers are more receptive than occasional ones;
• customers who have high average spending are more receptive than those who spend less.

The best customers to direct dedicated and tailored messages to are of course those who obtain high RFM scores. We also recommend setting a threshold score, under which it is no longer convenient to continue sending campaigns and would be better to experiment with new re-engagement strategies.

What criteria do you use to divide your contacts? Share your segmentation techniques with us in the space below.