Are you an e-commerce company that sells hundreds of products in various iterations and your customers come back to order from you time and time again? Are you pulling your hair out trying to house all the data you have on your customers into one database? Stop. Relational databases and tables will become your new best friend.
We welcome Digital Marketing Manager, Ali Nazam to explain what a relational database is and why as an e-commerce marketer you should use them.
What’s in a name?
You may have heard the terms child table or database, relational table, or relational database thrown around, but wondered how they’re different. I’ll let you in on a little secret – child and relational databases are the same and child and relational tables are the same thing. Still confused? Let’s break this down further.
Simply put, a child or relational database is a separate database that contains data expanding on an existing record in a parent or master database.
The data in the relational database is dynamic, meaning that it is changing often, with many databases reaching multi-million rows of data. The dynamic nature of this data prompts you to keep it separate from your more static customer information such as email, physical address, customer number, etc. in the parent database. Keeping the expansive, dynamic data separate from your master database facilitates more flexible and easily accessed customer records.
If you were to store all your customers’ order history within the master database, your data would look something like this:
Let’s say you have 250,000 customers in your parent database. If 100,000 of those customers placed 5 orders in 2016, and you want to record order date + total purchase amount of each order. That’s potentially (100,000 customers * 5 orders * 2 pieces of data) 1,000,000 extra pieces of information you have to store in your database! Relational databases ease the amount of data your main, parent database must house.
A relational database can either expand on a single record or many records within a master database. There are two different relationships that can exist between your parent and child database depending if the relationship is expanding on one or many records.
1) Many-to-One: Many records in the relational (child) database relate to a single record in the (parent) master database
Example:John Smith is a record in your master customer database. John Smith has purchased from you 5 times in the last year. John Smith has 5 rows of data in your order detail relational database that relates back to the master database via his customer number.
2) One-to-Many: One record in the relational (child) database relate to many records in the (parent) master database
Example:You manage a sales team of 50 agents each representing 1,000+ customers. You need each customer to have an agent’s email address and phone number assigned to them. Instead of updating 1,000+ records every time an agent’s information or territory changes, you only have to update a single record in the relational database.
One record in your agents’ relational database relates to 1,000+ records in the master database. Therefore, updating that single record in the relational database effectively updates the 1,000+ customers/leads records as well.
Now that we’ve gone through what a relational database is and what are the type of relationships relational databases can have with parent databases; let’s explain how marketers can leverage data in relational databases using relational tables in emails.
Using Relational Table Data to Segment Audiences & Personalize Content
So we know relational databases are really helpful to marketers to keep their data at a manageable level and that depending on need and desired outcomes, parent and relational databases can have two different relationships.
Expanding on the examples above, let’s see how you could use relational table data in your emails to segment your audience and personalize content for a one-to-one marketing experience.
Personalized product recommendations
Fourth of July is right around the corner. Say you want to send an email to all customers in your database that purchased Fourth of July or patriotic supply from last year?
First of all your relational databases would already need to be established and you’d need to be recording the product category for each purchase in your order detail relational database. If this is in the works, you can segment your list by customers who purchased patriotic products last year.
Additionally, you could personalize the email with new stock relevant to products they have previously purchased in the past. If the customer purchased Fourth of July glassware in 2016, why not include the updated 2017 glassware or related products in their email? You can achieve just the right amount of personalization with relevant products they may be interested in, without crossing the “creepy email marketers” line.
Personalized agent assignment emails
Let’s say that late-spring/early summer are your busiest seasons. You recently had an agent of yours leave the company, but you don’t want to miss out on the busiest season for this territory. To take advantage of the many-to-one database relationship, you can update the old rep’s information with reps on hand to take over for the 1000+ customers the old rep had in their name.
Then you can send out an email to those 1,000+ customers introducing the new agent with a picture of them and their contact information.
Achieving the 1-1 Marketing Experience
The possibilities are endless when it comes to using relational databases, not even for promotional emails, lead generation emails, but also transactional emails. Think about how you can implement relational tables into your order and shipping confirmation emails to include shipping addresses, tracking numbers, and product details and pictures. Or brainstorm other ways you can leverage having multiple databases all working together.
As marketer’s we strive to reach as many people as possible while still maintaining a 1-1 experience for our customers, and relational databases empower us to do so.