Meet Pér
Consumer behavior doesn’t wait for a brand to catch up.
A loyal customer begins to disengage. Demand shifts unexpectedly. A campaign reaches the wrong audience. A promising opportunity appears in the data, then disappears before someone can act on it.
Most companies are not short on customer data. They are short on the ability to turn that data into a confident decision while the decision still matters. That gap between knowing and doing has quietly defined our industry for a decade.
Starting today, that gap closes immediately. We’re introducing Pér, an AI agent for every customer decision, and with it a fundamentally different way for companies to understand and act on their customers.
Our ambition for Pér is to deliver 10x marketing efficiency. Not by removing people from the decisions that matter, but by collapsing the distance between trusted customer context, human judgment, and approved action.
Pér understands what is happening with your customers, recommends what to do about it and, with your approval, puts that decision to work in the tools your company already uses. Trusted customer context, AI reasoning and human judgment, working together in one governed loop. Work that once moved through a back-and-forth of dashboards, data requests and meetings now happens in a single, continuous flow.
From knowing the customer, to acting on what you know
When we started Amperity in 2016, the hardest problem in customer data was knowing who the customer was. Records were scattered across dozens of systems, and no one could see the whole person. We spent a decade solving that problem, and today more than 400 of the world’s leading brands run on the multi-patented identity foundation we built. Pér solves the problem that comes next: acting on what you know while the decision still matters.
Inside most enterprises, a simple question like “Why are our best customers leaving?” still sets off a relay, that’s often both backward and forward. A marketer files a request, an analyst pulls the data, a data team builds the audience and a technology team pushes it live. Context gets lost at every handoff. By the time the answer arrives, the customer has already moved on.
The right collection and use of AI models, trained the right way, and on top of the appropriate context and governance, finally makes taking those activities out of the equation possible. That is why we built Pér.
Our customers are already seeing what’s possible
At MacKenzie-Childs, the business intelligence team asked Pér a question every retailer worries about: which of our best customers are slipping away? Pér surfaced 37,000 high-value customers who had not returned, explained what set them apart and turned that insight into a clear decision plan and an audience ready to activate. “It’s helping us rethink retention around the individual customer journey,” said Elise Edmonds Keefe, Sr. Manager, Business Intelligence at MacKenzie-Childs, “so we can identify the moments that matter and move from insight to action through one connected workflow.”
For Johnnie-O, Pér audited segments the team had relied on since early in its Amperity journey, surfaced overlaps and inefficiencies, and let the team fix them on the fly, significantly improving the accuracy and performance of those segments. All without anyone needing to ask Pér to go investigate. It just did it overnight, for fun. As Geoff Clark, CRM Manager at Johnnie-O, put it: “Most AI tools are reactive. Pér is proactive.”
These are early days. But across our customers, the pattern is clear: when an agent truly understands the customer, the distance between a question and an action essentially disappears.
Pér begins with the business question. It helps teams examine relevant customer and business context, identify meaningful changes, explain the evidence behind a recommendation and prepare a supported next step. The user can review the reasoning, revise the recommendation, reject it or approve it for use in an enabled workflow.
The person remains accountable for the decision. Pér makes the path to that decision shorter, clearer and easier to act on.
Context is what makes an agent trustworthy
Many AI experiences begin with the model. We believe enterprise AI must equally consider context.
Customer data is especially demanding. Identities change, signals arrive at different speeds, permissions vary and important history lives across systems that were never designed to work together. If an agent cannot reliably understand who a customer is, what has happened and what the business is permitted to do, faster reasoning will not produce better decisions.
When the answer is wrong, the agent is wrong with total confidence, millions of times a day. AI doesn't fix bad data. It just acts on it faster.
Pér operates within the data, permissions and workflows made available to it. Teams can inspect the evidence behind a recommendation and keep human approval substantive. Trust is not a disclaimer added after the product is designed. It is fundamental to how it works.
A new operating model for customer data
The most important changes created by AI will not come from adding a chat box to existing software. They will come from redesigning how work gets done.
Pér creates a governed path from question to evidence, recommendation and approved action. People can review, revise or reject what it recommends before work moves into a supported workflow. The goal is not to remove human judgment, but to give it better context and a clearer path to action.
Customer relationships are shaped by thousands of decisions: who receives an offer, when a customer needs help, which audience should be suppressed and how a company should respond when behavior changes. Those decisions are too important to make from incomplete data and too time-sensitive to remain trapped in lengthy analytical and operational processes.
The companies that lead in this era will not be the ones that automate the most decisions. They will be the ones that improve the decisions that matter.
For more than a decade, Amperity has been in the fortunate position to help the world’s best brands know their customers. With Pér, they can finally act on everything they know, at the speed their customers move. It is by far the most important thing we have built since we started the company, and we can’t wait to show you what it can do.
Kabir and Derek