Recently, we had a large food & beverage e-commerce company come to Cotera looking for a strong segmentation model. The Head of Marketing needed to bring down a high one-and-done and churn rate. She knew they were facing an abundance of customer churn but that her team wasn’t proactively identifying these at-risk customers until it was too late. With a company-wide focus on retention, she knew they needed a dedicated tool that specialized in customer segmentation.
Recently, we had a large food & beverage e-commerce company come to Cotera looking for a strong segmentation model.
The Head of Marketing needed to bring down a high one-and-done and churn rate.
She knew they were facing an abundance of customer churn but that her team wasn’t proactively identifying these at-risk customers until it was too late. And on the other side of the spectrum, AOV for their top customers wasn’t growing.
They tried to use the pre-built RFM model in their ESP, but it was too generic and missed too many churning customers. And this wasn’t changing the behavior of their top customers either.
With a company-wide focus on retention, she knew they needed a dedicated tool that specialized in customer segmentation.
They wanted a model that could:
On churning customers
On top customers
Prior to Cotera, the brand faced several problems:
1) Existing customer segments were too broad
Here’s what the food & beverage brand would have sent out in a marketing campaign before working with Cotera:
The issue with this is, although their at-risk customers might need a promotion to nudge them toward their next purchase, their top customers were likely going to buy from them regardless.
And as a result, emails were underperforming with an increasing opt-out rate, and AOV stayed static.
2) Segmentation and custom emails require too much time and work from the retention team
Manually having to analyze customer data and pull individuals into different segments was more than just time consuming. Work like this often led to simple mistakes and oversight — from mis-categorizing customers to sending out the wrong emails.
3) Customer segments would not be updated in real time
Another downside — customers weren’t automatically moving across segments as their behavior changed. This was due to a combination of segments being too broad along with the use of a generic logic that didn’t apply to the nuances of a food & beverage brand.
For this brand (and many others!), Cotera set up real-time customer segmentation that measured each customer’s brand affinity using historical order data, website click data, and marketing interaction data.
1) Every day, each customer gets labelled with custom segments
For the food & beverage company, we split customers into 3 different segments:
Let’s say this is some of the data our model automatically took in for a random customer:
Our model then processes this information and would assign this customer to a segment. In this case, this customer would likely be placed into the “Highly Loyal Buyer” segment.
2) These segments are automatically integrated into our companies’ email and marketing platforms (Klaviyo, Iterable, Braze, etc.)
Instead of one general message, they could now automatically send multiple customized campaigns to the segments we built for them:
To Group A (At-risk Customers):
To Group B (Top Customers):
And the best part about it all is they didn’t have to change a single thing about their current SMS/email systems to deliver these messages.
Using our Segmentation Model, Cotera was able to:
1) Increase upsell rate by 5% on average
2) Increase reactivation rate by anywhere between 14% up to 20% per month
3) Lower churn rates more than 2x
Interested in bolstering your segmentation and personalization efforts? Book a demo here!