From Customer Question to Next Move: How MacKenzie-Childs Is Exploring Pér
MacKenzie-Childs is exploring a new way to work with customer data: ask a question, find the context behind it, and decide what to do next.
As an early design partner for Pér, Amperity’s AI agent for customer data, the team is testing how AI can help uncover answers across existing data, take on analysis that once required hours of manual work, and turn those findings into potential customer experiences.
For Elise Edmonds Keefe, Sr. Manager, Business Intelligence at MacKenzie-Childs, that means spending less time figuring out how to get to an answer and more time exploring what the answer could mean.
“Pér is able to add in more strategic thought and layering, and it’s a little bit more conversational. The responses are a lot richer.”
One question. More ways to find the answer.
Finding an answer Elise didn’t know was there
One of Elise’s first surprises came when she asked Pér about a recent bounce-back coupon campaign.
She wasn’t expecting an answer. MacKenzie-Childs didn’t have promotion data in Amperity, or so she thought.
Pér recognized that the campaign audience had originally been created in Amperity and found another route through the campaign recipients data.
“I had forgotten that we had created the audience in Amperity. I was thinking, we don’t have promotion data here, so we wouldn’t actually be able to learn anything.”
- The question
- How did our bounce-back campaign perform?
- What Pér found
- The relevant data was already there, just not where Elise expected to find it.
Pér could reason across the customer data MacKenzie-Childs already had without requiring Elise to know exactly where the answer lived before she started.
Turn customer feedback into customer context
MacKenzie-Childs also collects free-form feedback through a post-purchase survey. Historically, analyzing those responses meant exporting the data into spreadsheets and manually categorizing answers before Elise could identify patterns.
Once that survey data was brought into Amperity, Pér gave her another way to explore it.
Elise could ask questions directly against customers’ written responses, then examine those answers alongside information such as purchase history, buyer type, demographics, and product data.
One question focused on customers who said they had purchased a gift for a wedding or related occasion.
Pér analyzed the free-form feedback alongside customer and product information, helping Elise understand who those shoppers were and which products were more closely associated with the occasion.
- 3K
- Distinct customers identified around wedding, bridal, registry, or engagement-related gift purchases.
- ~90%
- Already opted into email.
Pér also surfaced characteristics of the audience, including a high proportion of first-time buyers and product affinities associated with the occasion.
Work that once started with manually coding survey responses could now become a broader conversation about the customers behind them.
Keep the conversation going
From “who are they?” to “what should we do?”
Elise didn’t stop once the audience had been identified.
She asked Pér whether MacKenzie-Childs should target those customers.
Pér proposed a potential next step: a dedicated welcome and registry-engagement journey for the reachable audience. The recommendation brought together the audience definition, supporting analysis, and a potential email treatment for the team to consider.
- Customer feedback
- Audience identified
- Customer context analyzed
- Potential next move
The analysis became the beginning of a decision, rather than the end of the workflow.
Pér learns the language of MacKenzie-Childs
Customer data alone doesn’t capture everything a team knows about its business.
MacKenzie-Childs can give Pér rules and definitions that reflect the way the company actually works. Elise can reference something like the “2026 Barn Sale,” for example, and Pér can understand the corresponding dates without requiring her to explain them each time.
The same approach can apply to collections, collaborations, and other company-specific language.
“Being able to train it with our own vernacular has been critical, and it allows other teams to use the tool efficiently using terminology we use every day.”
That context can make customer intelligence more accessible to people who may not be comfortable writing SQL or navigating the underlying data themselves.
From broad segments to individual journeys
Elise’s north star is personalization.
Today, MacKenzie-Childs might think about experiences in broad groups, such as new versus existing customers. The opportunity ahead is to understand more of the context around each person: what they bought, what they said, where they are, how they behave, and what similar customers tend to do next.
The post-purchase survey offers an early glimpse. A customer’s own words can become another signal alongside behavioral, demographic, and product data, giving the team more context for deciding what experience could come next.
“It’s literally about eliminating the grunt work and actually being able to have accurate, valid findings that are actionable.”
For MacKenzie-Childs, the promise of Pér starts there: fewer manual steps between a customer question, the relevant context, and the next decision.