Email Segmentation, Personalization, Automation: Getting Important Data Into Your ESP

What kinds of data should you be uploading into your ESP? Columnist Andrew King outlines that and more to help boost the relevance of your emails.

As an email marketing consultant, I frequently talk to marketers about the benefits of segmentation, personalized content and automation. Marketers are generally in favor of all these things and usually agree with most of my recommendations.

However, when it comes to implementation, we frequently run into the issue of how to get the required data into their email platform. This issue doesn’t just come up once in a while; it comes up all the time.

So, I want to discuss the types of data that you should be uploading into your email service provider (ESP), how this data can be used to enhance the relevance of your emails, and finally, how to get that data into your ESP.

Useful Types Of Data

This is your basic A/S/L (age, sex, location) data that you might collect from subscribers when they sign up or make a purchase. This type of data generally doesn’t change very often; people don’t change their name, gender or location too often (although all of those things could change), which makes this type of data ideal for basic segmentation, personalization and automated emails.

  • Name
  • Age
  • Location
  • Gender

Preference data is usually collected through a preference center that your subscribers fill out during the signup process and might include data regarding their preferred products, services, brands, size or frequency of mailing.

You need to be careful how you use this data as it can go out of date very quickly. For example, someone might tell a travel company that they’re interested in visiting Thailand, but that doesn’t mean you should only send them emails about Thailand for the next two years!

I also want to note that many people won’t update their preferences if they do change, unless prompted, so it might be worth setting up an automated email to remind subscribers to do this periodically.

  • Product
  • Service
  • Destination
  • Category
  • Size
  • Frequency

Transactional/RFM (Recency, Frequency, Monetization)
Liberating your transactional data from its e-commerce platform prison is one of the best things you can do for your email program. Importing this data into your ESP will dramatically improve your segmentation and personalization, and opens up a new world of automated emails.

This data allows you to identify who your best customers are, who’s about to lapse, or who hasn’t made a purchase yet.

  • First purchase date
  • Last purchase date
  • Total amount spent
  • Number of purchases
  • Average order value
  • Past products purchased

Recent behavioral data is the most reliable indicator of what your subscribers are interested in right now. This type of data might be collected from an email (opens/clicks) or from your website (pages browsed/items carted).

The most common type of behavioral email is a cart abandonment trigger, which usually goes out within 24 hours of subscribers abandoning their cart. However, many technologies are now available which allow you to follow up pretty much any action that a subscriber takes on your website. That sounds a bit stalker-ish, but I assure you that it can be used to send helpful, relevant emails based on the products, services or topics they have recently browsed.

  • Product/Service/Web page browsed
  • Cart abandonment
  • Form abandonment
  • Email opens/clicks

What Are You Planning To Do With This Data?

There’s no point in doing a data integration if you don’t know what you plan on doing with that data — unfortunately, this is a far-too-common scenario.

I suggest that you create a very specific plan and mock examples of what you’d like to do with the data available to you and then work out how you are going to get the required data into your ESP. Typically, your ideas will fall into segmentation, personalization or automation.

Having a rich data set within your ESP opens up some great opportunities to create highly targeted segments, which combine demographic, preference and transactional data.

For example, you could target people of a particular gender, who are interested in a particular product category, have made a purchase in the last year, and spent over a certain amount. Now that’s a targeted segment!

2015-04-27_11-58-49 used segmentation to send this email offer to some of its best customers:



Personalized emails deliver six times higher transaction rates than non-personalized emails, but 70 percent of brands fail to take advantage of that, according to Experian Marketing Services. I assume this is due to a lack of data, easy-to-use personalization tools in many ESPs, and the time-consuming nature of testing personalized email content.

However, with the right tools and data, it can be very easy to implement. A few examples of how I’ve seen brands use personalization include:

  • Offering discounts in bulk emails which only certain subscribers can see
  • Personalizing the content of a birthday email, based on the year that the subscriber was born
  • Showing products based on the subscribers brand and size preference
  • Aggregating data from mobile apps for end of week/month/year summary emails

Fitbit has integrated the data from its mobile app with its ESP to send this highly personalized weekly summary email. (Disclosure: Fitbit is a client of my employer.)


Automated emails

Sending automated trigger emails would not be possible without data. Whether it’s an email address entering your list, a birth date or a cart abandon, data is required to trigger the correct email.

The timeliness of these emails is also extremely important — there’s no point in sending a shipping confirmation email after the package has arrived. This is why you ideally need an automated way of uploading your data into your ESP as frequently as possible, ensuring that these trigger emails are delivered on time.

Illy Coffee takes timing seriously when it comes to its automated emails. (Disclosure: Illy is a client of my former employer, Lyris.) The example below is sent within hours of a subscriber browsing its website and not adding anything to his or her cart or making a purchase. As you can see, the company has taken a customer-service approach to seeing if the subscriber had any issues finding an item or making a purchase.


How Do I Get This Data Into My ESP?!

Now that you’ve determined what data you have access to and what you’d like to do with that data within your emails, it’s time to figure out how to actually upload it into your ESP. I wish there were an easy answer to this question. Unfortunately, there are a lot of variables, and it will most likely depend on how easy it is to get data out of one system and into another.

Generally, these will be your main options.

Manual upload

Manually uploading data into your ESP is by no means an ideal solution. However, it still seems to be a fairly common practice, as it is easy and anyone can do it. I don’t recommend this approach, though. It can be very time-consuming if you have a large list; it increases the risk of making a mistake; and it doesn’t allow you to send automated emails in a timely manner.

If you need to justify to your boss why this shouldn’t be an option, add up the amount of time that you’ll spend in a year manually uploading your data and estimate how much that will cost your company. You can then compare that amount to the cost of a one-off integration project. The savings should be substantial!

FTP batch data upload

This method involves your IT team setting up an automated process to drop your data onto an FTP (File Transfer Protocol) site at an agreed upon time each day. Many ESPs offer a built-in upload from the FTP feature, which simply allows you to input the FTP location details, or you should be able to use an API call to upload the data into the ESP from the FTP.


Most ESPs provide developers with an API (Application Programming Interface), which they can use to upload data from a variety of locations. This solution will require a developer to hardcode specific API calls to pick up and upload the data into your system.

It’s a good solution for email marketers who have access to a developer. But because the process has to be hardcoded, it can become painful if you want to change your systems or add additional data fields.


Connectors are pre-built integrations between an ESP and the main e-commerce, analytics, ERP and CRM (Customer Relationship Management) systems. Many ESPs are now offering these pre-built connectors to make it easier for marketers to execute an integration without the help of IT. These types of integrations work great, unless you have a homegrown system or very specific requirements.

Hopefully, I’ve helped outline the huge benefits of having a rich set of data within your email platform. I know that an integration can seem like a very daunting task for many marketers at first, but it doesn’t have to be overly complex — even simple pieces of data such as location, gender, and product preference can open up a number of possibilities.

Remember to determine what data you have access to in your other systems, outline exactly what you plan on doing with it, and then develop a plan with your IT team to help you import that data on a regular basis.

Some opinions expressed in this article may be those of a guest author and not necessarily Marketing Land. Staff authors are listed here.

About The Author

Andrew is an email marketer and Sales Engineer at Campaign Monitor. He has designed, coded and deployed hundreds of emails (to millions of people) and most recently consulted on email strategy for some of the world’s largest retailers, non-profits, financial institutions and publishers. He regularly speaks at industry events and can often be found in San Francisco, pondering the future of email over an artisanal coffee.

(Some images used under license from


Marketing Land – Internet Marketing News, Strategies & Tips


Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.