Introduction
MessageGears direct data connection gives marketers the ability to use their complete data profile of customers to create highly segmented groups for activation on any channel. Keep reading for a full end-to-end walkthrough of the ways that users effectively segment and activate their customer base using their data.
In this article, we will cover:
- How Marketers can use Blueprints to manage campaigns across all channels, both through our waterfall segmentation as well as using labels to enhance their view of the customer
- The ways in which full access to customer data creates more timely and accurate access to customer identity attributes, ultimately providing better matching on ad, social media networks, and other activation destinations
- How MessageGears network of over 250 activation channels gives marketers the power to create limitless customer journeys
Benefits of MessageGears' Segmentation and Activation Solution
- Using the brands' full view of the customer without data syncing or mapping creates seamless and real-time journey creation for a fraction of the cost of a bundled CDP
- Cross-channel coordination empowers marketers to create highly-personalized customer journeys, utilizing your full view of the customer for a seamless end-user experience
- Behavioral events from these activation channels are available for intelligent retargeting, offer optimization and advanced campaign analytics
Background and Scenario
Imagine that we’re on the customer engagement and marketing team for City and Glory™, a furniture brand with a large nationwide presence. In this use case, we have been assigned with creating an omnichannel customer engagement campaign that reaches our customers at the appropriate channel at the right time.
In creating this nationwide omnichannel campaign, we will need the ability to:
- Segment our audience using attributes stored in their member profile table within our database - notably segmenting by engagement, internal promotional score and purchase recency to deliver the most relevant message
- Execute our campaign brief and activate audience members on appropriate channels:
- Highly engaged customers with a high purchase propensity: No coupon offer, activate via their preferred channel via our ESP
- Highly engaged customers with low purchase propensity: Segment based on Lifetime Value and send a BOGO offer to High LTV customers
- Low channel engagement customers should be A/B tested between receiving different promotions or being held out to understand how promotions effect that target audience
- Create holdout groups for different sets of content and channel pairings
- Record audience state within our database to allow for audience logging and enhanced visibility
- Use activation analytics to influence downstream marketing activity and messaging
Let’s Get Started.
Visual Segmentation
To begin, we’re going to build a blueprint from an existing audience connection. In this example, we’ll find our City and Glory Member Profile Table and start from there.
Our resulting blueprint is blank - after counting we are met with our starting population of 1.6 million audience members.
The starting population of a blueprint is the total number of customers available to segment and filter. For information on configuring your starting population, see our article on creating audiences.
To begin segmenting audiences, we’ll create new nodes by clicking the kinds of segments we’d like to make. For blueprints, there are four basic segment operations users can take:
- Target: Creating logic to help target specific audiences within a population to allow for segmentation
- Filter: Filtering down nodes based on available data attributes (e.g. selecting audience members in a few specific states from a starting population of the USA)
- Split: Randomly split the node into different segments, helpful in A/B testing
- Merge: Merging previous nodes into a singular node, used for labeling and appending data to audiences
In this example, we’ll start by targeting our users based on engagement. Let’s start by creating two segments - high engagement and low engagement. High engagement can be any customers that have either visited a store or opened an email within the last week, and low engagement customers can be everyone that hasn’t fulfilled either criteria:
After hitting ‘count’ in the top right, we can see the resulting population of the customers in our resulting segments.
We can continue to apply more filtering logic to create even more tailored audiences. For instance, we can continue to target the ‘low engagement’ segment on purchase recency, and even A/B split our low engagement population that haven’t made recent purchases for activation on different channels or to provide different offers:
In this high-level example, we’ve ended with a total of 6 activation nodes, or the ‘edges’ of the blueprint, that will be available for activation in our campaigns. Before we do that though, we’ll want to add attributes to our audience to understand the offers and channels they’re activated on. For that, we'll go to our next step: applying labels.
Creating tags and updating our audience data
So far, we’ve been able to create segmented populations of our users by applying filtering logic to our population - but there might be reasons that relying solely upon the data in our data warehouse isn’t enough to fulfill our marketing needs:
- Creating attributes for easier personalization in downstream messaging platforms
- Recording and understanding marketing treatments based on enhanced segmentation logic (i.e. “which people received the BOGO coupon?”)
- Creating marketer-friendly ‘names’ of channels and programs to more easily generate time-series analytics
For this example, we’d like to create 3 different labels:
- The Segment ID that each recipient will belong in
- The Coupon or offer that they will be sent, and
- The kind of hero image copy they should receive in downstream messaging
MessageGears labels work in a two step process: Defining the labels for our population, and applying them to segments.
Creating labels
To begin, we need to identify the additional attributes we’d like to append to our data, and place them at the root of our blueprint to be used within segments later.
When creating labels, we can choose to have them be appended (or concatenated) to one another, or overwritten:
Appended labels will create a comma-separated list of each value that applies to that audience member. Appended labels are useful when understand which segments an audience member was in each day, and how they make their way through the customer journey - for instance if a person was ‘recently engaged’ and had a low ‘purchase propensity’ on a given day, it could be useful to record both label values.
Overwritten labels will only apply a single label into the data, meaning the ‘final’ target node that a recipient falls into will be the only label written back to that audience member. A good example here might be which offer an audience member received on a given day - although they may qualify for many, they ultimately only received a single offer, so it only makes sense to record one.
After identifying the labels we want, we can click on our starting population and define them there:
Applying labels to segments
After we’ve created our labels, our next step is go through our segments and define where we add our attributes. To add a label, simply click ‘add label’ above the segmentation logic, and define which labels to add to which segments:
Notice how we’re overwriting the labels for ‘HeroCopy’ and ‘OfferId’, but are appending the ‘SegmentId’ label, as we’d like to record each segment that each recipient flowed through.
Labels unlock the ability to define multiple segments within nodes as well - Notice how within the same node of ‘High LTV’, we can have multiple segments based on lifetime value to better understand and work with our audience:
After appropriately segmenting and labeling our audience, we’re finally ready to begin activating our recipients.
Activating on Third-Party Marketing Channels
After segmenting our audience and adding the appropriate marketing attributes, we are ready to activate our recipients through some of MessageGears 250+ activation channels.
To begin, we want to create an external campaign for each activation channel we'd like to feature for our audience. In this use case, we'd like to activate:
- High LTV and Promotional score audience members through emails, via Salesforce Marketing Cloud
- High purchase likelihood individuals through Meta Ads/Facebook Custom Audiences
- All other recipients through Google Ads/Google Customer Match
We can simply select the blueprint as our audience and the destination from the dropdown in the external campaign section to set up the initial campaign. Navigating to the 'Blueprint' tab to filter which nodes in the blueprint you'd like to segment or activate for the specified channel.
Prior to sending our audience, we can optionally customize which data attributes we send to our third-party for activation. In many cases, we'll be segmenting on personal data that we may not need in deliver - items such as:
- Monetary spend
- Recent purchases with our brand
- PII that should not leave our data warehouse
Personal data like this can be useful for segmentation, but likely won't be needed in personalization aside from the easy-to-use data attributes we created with our labels. To select only the columns that we'd like to send to our third-party, we can select the 'column customization' tab and use the WSIWYG column picker to select the fewest number of data attributes necessary:
After creating, we'll navigate back to the 'settings' tab to schedule our campaigns to this destination. Each campaign having its own schedule gives us the flexibility of activating different members of this audience at different times, allowing everyone to be on their own schedule.
Once we have set up all of our external campaigns, we can view all of our activations for each segment from our initial blueprint by clicking the 'planning' tab in the top right and seeing the associated campaigns.
Creating a Customer History and Viewing Activation Analytics
At this stage, we've defined our audience, appended semantic data for use downstream, and activated our audience via our third-party channel destinations - the last step in our use case is to use the resulting analytical information to influence our campaigns in the future. At MessageGears, there are two ways to view analytical data: short term destination-level analytics and long term send logging for internal analytics.
Activation Analytics
For a high-level look at how our scheduled transmissions are taking place, we can navigate to the 'analytics' tab of any external campaign. There we'll get a snapshot of how we are activating our data and if our attempts to transmit data and activate are successful.
Field-level values are as follows:
Destination: The name of the activation channel for this campaign, with ancillary information available for certain destinations
Start Date/ Extraction Time: The time at which MessageGears began the data query to send to the third party. In certain situation, the extraction time can be hours - given the unpredictable nature of the database we display both the start time and the total time to query and transmit data
Size: The number of recipients in this campaign
Action: For certain destinations, which recipient-level actions were taken for this list (whether to add/remove/replace)
Name: The name of this campaign execution
Status/Request ID: Whether or not the activations were successful, and the MessageGears request ID to be used for any support requests
Job ID: An internal ID used for campaign tracking
Audience Recording and Time-Series Analytics
Additionally, MessageGears allows users to record audience state on any given cadence - similar to send logging in many platforms. The benefit to direct audience recording is the ability to record every recipient's state and the segments they were in whether they were activated on a channel or not.
To enable audience recording, we simply select a pre-specific audience recording function in the 'advanced settings' of the external campaign settings tab, seen here:
Recording the state of each recipient in our audiences unlocks virtually any analytical view of our activation data we'd like, as we now have a full time-series database of our audience and the different ways we view them. These segment-level analytics are incredibly valuable in understanding both who our audience is today, and the different ways our customers flow through their customer journey over time and how our messages impact their journey:
Comments
Article is closed for comments.