Why you should segment your database

Segmenting your email database will allow you to send content which is better targeted and more relevant to your customers. This video looks at how you should define your segments for the best results.

Click here to hide or show the transcript


When starting an email database, it’s important to remember that, just because everyone in your list has subscribed to receive emails, they’re unlikely to be interested in receiving the same emails. As you have a number of different types of customer, the most effective email marketing will include content that’s relevant and interesting to each one.

In order to be able to send more targeted emails like this , you’ll need to segment your database. In this video I’m going to talk about how to define these segments.

The number of segments, and their purposes will be dictated by your customer types and will be different to each business. Where you sell a range of products, your product categories will usually form the basis of your. segmentation For example, a beautician’s may segment by service and create separate segments for customers who have nail treatments, massages, waxing and facials. These would be set up as lists within the email system so that any email created can be sent to one or more lists. Customers may be included on more than one list - if you send to multiple lists, your email system will reduplicate at the point it sends the email out. You can of course choose to send an email to your entire database - for example if your beauty salon started offering a new service such as spray tanning, they would want to let all their customers know. But segmentation gives you the option of choosing which people the email is relevant to and targeting only those people. Also, by splitting your database into different lists within your email system, you’ll be keeping your customer types highly visible, so that when you create content, your audience is at the forefront of your mind. If you can name your segments in a concise way that will remind you of the customer personas they represent this will also be helpful.

Even if you only sold one product, the range of your customers would span different life stages and buying habits. Imagine you sell coffee. Your brand is at the expensive end of the spectrum and you currently only have one blend. You might find that you need to create three segments to cover the majority of your customers. One segment is made up of affluent, middle-aged customers, who entertain regularly, and buy every few weeks or so. You might have other customers for whom your coffee is a luxury treat and they tend to buy only once or twice a year. They may predominantly be retirees, on a comfortable pension. The third and final segment may be young and aspirational. They see your coffee as a cool brand, which they identify with, and like to be seen to buy it. They’re keen to talk about your coffee but in terms of buying frequency may again only be able to afford to occasionally. However because they’re spreading the word amongst their peers, this segment is quite large.

When we break it down like this, you can see that the types of content and promotions which would be most relevant to each group would be different, because their interests and the type of discount that would be most useful are different.

So far, we’ve spoken about reviewing and categorising customers by their characteristics. This can be done manually, by using your existing customer database profiles, although it is good practice to ensure that anyone signing up to your email database is aware that you are using personal data in this manner by including this in your privacy statement. A more transparent and easier way to segment your customers is to allow subscribers to choose for themselves which categories they belong to. By asking the customer to select types of email they want to receive on sign up, you give them control, and will see fewer unsubscribes because the content isn’t relevant. Remember, when someone unsubscribes, they don’t just unsubscribe from one list, but from the whole database and you will no longer be able to send them emails. The best way to manage this is to set up a preference centre for your email database, so customers can update which types of email they receive. I’ll cover preference centres in another video.

If you’re asking customers to select which emails to receive, you’ll still need to map out which segments to ask them to choose from. Again, in the case of product or service categories this is very straightforward, but what about our coffee example where we want to segment on life-stage and buying frequency? In this instance, you may need to think a stage further along the process and think about the kind of content you would create to appeal to each segment.

For my affluent customers who are frequent entertainers, I plan to send them recipes in which coffee is a key ingredient. The retirees who see my blend as a treat might be focused on savouring the moment - maybe they’re sharing my coffee with family on special occasions. So I want to send them content about all the ways they can make coffee part of more moments - I don’t want to downgrade their perception of my coffee by suggesting they make a flask up and take it on a family day trip or something; I want to show them that their moments can be upgraded by the addition of the coffee. My content might be things like the ideal way to spend a lazy day, starting with breakfast in bed with coffee, or about ways to reward yourself with some me time, accompanied by coffee. Finally, my young aspirations are trying to consume my coffee as publicly as possible so they might want to know about events and experiences which my company runs or they can create - so the content will be around coffee and food pairings, barista-style training sessions or coffee socials. As they’re not purchasing much at the moment this content will hopefully bring them further into the brand, give them more opportunities to talk about my coffee with friends or on social channels and help to create some buzz and engagement for the brand.

So I have some very different content strands there - I can ask customers on subscribing if they want to receive emails with recipes, lifestyle, and event information.

The other element I wanted to segment on was buying frequency. Because I want to offer discounts to email subscribers it’s really important to get this information, so I’m going to have to ask my customers for it. This will add an extra question on sign up, which may lead to fewer sign ups, but I think it’s worth this trade off because the discount codes are likely to increase my revenue and show some return on investment for my email efforts. Obviously when you ask customers to self report things like this, it won’t be as accurate as using the data, and people may over-report because they’ve bought into the brand and think of themselves as valued customers, or under-report because they simply forget how often they purchase. Because of this, I may need to experiment with the discount thresholds at first before I find the sweet spot. And I shouldn’t mix self-reported values with actual ones or this data will become a jumbled mess. But so long as I’m consistent and use this information as a guide it should still give me a fairly accurate range. As it’s never going to be an accurate value, I need to think about the options I give my customers to choose - I could ask them to say whether they buy frequently, occasionally and almost never but whilst this covers the range I need, it’s very subjective. I want to get them to report as accurately and unambiguously as possible so I’m going to ask them whether they buy a packet of my coffee more than once a month, once a month, every few months, a couple of times a year or less often. This gives me 5 tiers, which is more than I had when I was using actual data, but it presents a clear timescale for people and makes it easier for them to mentally gauge. Also, I don’t want people to overthink which band they fall within so I want quite precise definitions. What this doesn’t give me is the size of packet they buy, so I’m going to have to assume an average, or again I may be able to get a feel for this across the segments by experimenting with my discount offers.

I can then use these segments to nudge my customers into buying a bigger pack than usual, buying more frequently, buying more items in one transaction or buying direct, if I’m also selling via retailers. So people who already buy frequently, might receive a discount code for free delivery when they reach a certain spend threshold. Or I might offer them a multi-buy discount. For my occasional purchasers, I might try and incentivise them to buy a bigger pack by offering a percentage discount on that pack size only. I want to persuade people who almost never buy to make the leap, so I might offer them a money off coupon which they can use online or with my stockists.

So I now have my three content segments and five buying segments. I can go ahead and create eight separate segments for each of these, but as these don’t operate independently of each other, it’s probably better to create some as sub-segments. This will help me to identify more patterns in the way people respond to my emails. I will send the content emails more frequently than the discount codes, so I’m going to use these as my main segments and then each will have five sub-segments for the buying behaviour. I can then even send to just the sub-segment, for example people who are interested in receiving lifestyle emails and only purchase a couple of times a year.

Once you have your segments set up, you can monitor responses across them, and as a whole, and experiment with content and offers. Looking for patterns and insight will help you to improve future email campaigns. I may find for example that in future the free delivery discount is very popular, but has much greater impact amongst people who are also interested in events, as they are in the habit of purchasing directly from me without the delivery charge at the events they attend. You may also find that some segments are quicker to respond and respond in higher volume than others. As your database grows, you may choose to do a phased email for something like a limited offer and send to these more highly-engaged segments first, to make the most effective use of your budget. Depending on their response you can then decide whether to expand the campaign to other segments.

I hope this has made the value of email database segmenting clear. Thanks for watching. If you have any questions, drop me a line in the comments.


0 Comment

Leave A Comment