There is immense data that can be accessed with a single prompt. When artificial intelligence is trained on a specific industry or company, that knowledge becomes more niche, and the AI agent can even become an expert. This made the burgeoning technology a great resource of company or even store-specific information.
Today’s column explores the startups that released AIs that simply know more—may it be about individual store performance for a brand that has over a thousand branches, or about customers and how they are perceived by AI.
When AI knows the stores better: Zipline
Numbers can often say what is happening, but they rarely explain why. So, when a district manager looks at a dashboard and reviews the sales of each store under their watch, they just see which ones lead and which ones lag, but they do not necessarily know why this is so.
Retail tech startup Zipline said it is launching a new AI capability that can help local leaders make sense of store-specific performance. It’s called ZipCoach, and as the name suggests, it coaches field leaders by comparing sales with behavioral data in each store and then turns these observations into recommendations.
This marks its latest effort to innovate how store operations get translated into actionable data. It’s an enhancement to the operating system the California-based startup developed for the retail industry, an OS that has so far won over more than 150 brands, including fashion brands American Eagle, Alo and Pacsun.
It’s a feature that is especially useful for fashion retail, a sector which often has to look for structure in the chaos that changes floor sets, promos, markdowns and product launches nearly every week.
“There are a few blind spots among fashion brands we see often,” Melissa Wong, CEO and co-founder of Zipline told Sourcing Journal. “When HQ sends directions for all of that through email, chat apps, printed PDFs, and phone calls, it’s not clear what actually happened in the store.”
These blind spots are in how the store is actually operating: whether a company initiative such as a floor set is done right, if at all; whether an employee has read a message or did product training; and why a store might miss its targets.
“Zipline answers all three,” she said. “Every message and task is tracked down to the store and the person, so HQ can see what got read and what got done. And ZipCoach connects that behavior to sales, so a leader can see that a store is behind on a new product line because its team hasn’t finished the training on it.”
When AI knows the customers better: Contentsquare and Amperity
How well do you know your customers?
Two AIs from two different startups want to learn more about their customer base: Contentsquare’s Sense and Amperity’s Pér.
Sense is the name of the agentic AI that runs Contentsquare’s reimagined Experience Platform to do something traditional KYC metrics can’t: learn about the customer, but through the eyes of AI agents.
The Experience Platform, which is currently only available in beta, has a menu of new features that elevates the kind of analysis it can do. It can, for example, track a single customer’s “full path across sessions, channels and devices,” as well as interactions made by their AI agents. It can also trace conversion across every touchpoint, which includes large language models and AI-referred traffic, as opposed to just crediting the last click.
Sense is getting an upgrade too. Contentsquare said the AI agent now has the ability to benchmark data against industry peers, showing who is ahead while quantifying the opportunity to close the gap.
“AI is only as powerful as the context behind it,” said Jonathan Cherki, CEO and founder of Contentsquare. “A deep behavioral and contextual data foundation makes a real difference, and what excites me most is what it unlocks: the ability to understand more of the journey, make better decisions faster, and continuously optimize experiences for both people and agents.”
“Every business dreams of truly understanding the lifetime value of its customers. Now, it’s finally possible,” he said.
Pér, on the other hand, is Amperity’s AI agent for marketers and analysts trying to understand their consumer base. It is built around what it calls the customer decision loop in marketing that starts with understanding the customer, making a recommendation that would eventually be acted on and learned from.
Amperity summarizes the loop this way: understand, recommend, approve, act and learn.
All of this starts with a conversation in plain language, then the AI takes it from there. For example, the user can ask Pér to find and explain customer data, complete with a report that can be shared with the rest of the corporate team. If a user wants to run a marketing program, they can ask the AI agent about a group of consumers and then create a campaign or predictive model that acts on the answer.
It is also important to note that Pér is trained on company context. As the startup explained on its website, an AI agent without company context can see decline and say revenues are falling. On the other hand, an agent trained on company context can see the decline, and tell whether this is because the company is deliberately winding down specific segments of the business.
“The value of AI is not simply the answer it generates. It is the quality of the decision and action that follows,” said Kabir Shahani, co-founder and co-CEO at Amperity. “Pér is a teammate that works from a shared, trusted understanding of the customer and helps teams move from insight to approved action. It outperforms a generic CDP, or an AI model alone, because the data underneath it is complete, governed, and can immediately be put into action that drives measurable economic outcomes.”
When AI knows supplier risks better: Accuris and Trigo
These are two separate company updates that both just want supply chains work like well-oiled machines.
Data firm Accuris is making it easier to use its trusted supply chain intelligence. It launched Accuris Connect, bringing Accuris intelligence on supply chains to systems that a company already uses.
Accuris said teams can work from the authoritative source rather than a stale copy, keep a traceable thread from their work to the standards and parts behind it and see change coming before it is expensive to act on.
“Accuris Connect represents a significant expansion in both the scope of what Accuris delivers and the role we can play for customers,” said Claude Pumilia, CEO of Accuris. “It gives customers a more connected way to design, build and sustain complex products.”
Meanwhile, Trigo, a global provider of quality management solutions for the manufacturing industry, is expanding its portfolio of AI-powered solutions, including its On-Time Delivery Predictor.
The company said these AI tools analyze supplier performance to identify patterns that may signal delivery risks or future disruptions, which should make it easier to identify the suppliers that may need extra attention and the necessary interventions that must be made.
By automating part of the analysis, this approach has reduced assessment times by more than 50 percent, the company said in a statement, with accuracy levels of up to 95 percent for AI-generated outputs. Trigo said human experts still have to validate the findings and define the appropriate course of action.
“AI helps organizations identify risks sooner, prioritize actions more effectively and deploy resources where they can have the greatest impact. Combined with human expertise, it enables manufacturers to strengthen supply chain resilience, improve supplier performance and optimize quality costs,” said Benoit Leblanc, co-CEO of Trigo.