The Agile Brand Podcast: How AI Agents are transforming commerce execution

The biggest bottleneck in commerce isn't strategy. It's the time it takes to execute it.

That's the central theme from a recent conversation between Greg Kihlstrom and CommerceIQ leaders Himanshu Jain (Co-founder and Head of Product) and Bill Schneider (VP Product Marketing) at eTail Palm Springs. The discussion explored why brands are increasingly moving beyond traditional agency models toward agentic AI and what this fundamental shift means for commerce teams.

The data paradox: Drowning in insights, starving for action

According to recent CommerceIQ research surveying 250 CPG ecommerce leaders, 80% feel overwhelmed by data. But here's the surprising insight: the core challenge isn't company culture, strategic alignment, or process. It's the gap between having data and being able to act on it.

"When you think about an ecommerce leader that is managing hundreds of SKUs, thousands of SKUs across multiple retailers, there are just so many different data points that they have to keep in their awareness," explained Schneider. "Making actionable decisions from all of that is a real challenge."

Jain illustrated the problem with a timely example: "Valentine's Day just passed. How many brands did you notice curate experiences to capture those moments? Very few. The challenge isn't that they don't know Valentine's Day is coming, it's the operationalizing of changing the PDPs or changing their experiences. It's very, very hard."

The culprit? A fragmented tech stack. To make a simple content change, teams often need to:

• Coordinate with agencies (who may be booked months in advance)
• Pull content from PIM systems
• Access digital assets from another platform
• Use syndication tools to push changes to retailer sites

"There are like tens of different systems that they need to log into, extract information, and update," Jain noted. "It's manually impossible to do it."

Beyond automation: What makes agentic AI different

Automation has existed for years, but agentic AI represents a fundamental evolution. The distinction? Planning and intelligence.

"Automation is basically following a series of steps. if X happens, do Y," Jain explained. "What agents bring to the table is the ability to optimize towards a goal."

Consider a competitive pricing scenario: A competitor drops their detergent price from $12.50 to $12.20. A traditional automation would simply match the price. An AI agent takes a smarter approach:

1. Evaluates relevance: Is this competitor actually relevant? Does this 30-cent change matter?
2. Draws on memory: What happened last time this competitor dropped prices? Did it affect sales?
3. Checks multiple systems: Is there enough inventory? Enough margin? Trade funding for promotions?
4. Considers business strategy: Is the goal to increase cash flow or gain market share?
5. Simulates scenarios: What happens at $12.40 vs. $12.30?
6. Recommends with rationale: Presents options with supporting analysis for human review

"Imagine this whole process could be very dumb: just match the competitor, which can lead to margin erosion," said Jain. "Or you would ignore it because you can't run all that analysis. Now an agent can do 90% of the work and a human applies 10% judgment on top."

The long tail opportunity

One of the most compelling benefits of agentic AI is unlocking value from previously neglected areas of the business.

"It helps you get to the long tail," said Schneider. "There's no scale limit. They're running all the time in the background, and you can apply that to your whole SKU catalog."

This extends beyond products to retailers. While teams typically focus on Amazon and Walmart, Jain pointed out that retailers three through ten often combine for 50% of revenue and present better economics.

"Winning a bid on a snack keyword on Amazon or Walmart might cost you $10 per click. On a Hy-Vee or Ahold, it might only cost you two bucks," he explained. "There's a massive amount of opportunities available on these long tail retailers and long tail SKUs."

Measuring the agentic shift

The efficiency gains are dramatic. Tasks that once took 30-60 minutes—like PDP adjustments—now take less than a minute with agentic support.

But the true KPIs go deeper:

• Input metrics: Conversion rates, search rankings, traffic, out-of-stock rates
• Output metrics: Sales, share, and margins

And there's a powerful second-order effect: "Most employees can now spend 80% of their time on strategic activities that agents aren't good at," Jain noted. "Negotiating with a retailer, creative work: an agent won't negotiate with Walmart for better deals."

The human-agent partnership: New skills for a new era

With AI agents handling execution, what skills should commerce professionals develop?

Jain offered a practical framework, comparing AI agents to junior analysts: "When you hire an intern, you provide business context, tell them who to talk to, what systems exist, how decisions are made. The same way, you onboard an agent."

The key skills for humans in this new model:

• Judgment: Knowing when to accept recommendations and when to override
• Deep expertise: "Five years back, a generalist was considered great. Now, depth in a particular area matters a lot more because you're teaching an agent to do that."
• Continuous feedback: Training and refining agent performance over time

"In not so distant future, maybe a year from now, I expect every white-collar employee to onboard a bunch of these agents and become 10X more productive," Jain predicted.

The path forward: Change management is critical

The research is sobering: 80-90% of AI pilots are failing. The culprits? Change management and lack of business context.

Successful implementations, according to Jain, require what CommerceIQ calls "forward deployed engineers" putting technical resources within customer organizations to onboard agents, understand unique processes, and integrate across siloed systems.

"The most successful brands and retailers will be the ones who embrace this change and are agile enough," he emphasized. "It's important to have the right governance, but it's also important to move really, really fast because the pace of change is massive."

Key takeaways for commerce leaders

1. The bottleneck has shifted: From insight generation to execution speed
2. Agentic AI isn't just automation: It brings planning, intelligence, and goal optimization
3. Think beyond the top 20%: Agents unlock the long tail of SKUs and retailers
4. Invest in human judgment: As agents handle execution, expertise becomes more valuable
5. Prioritize change management: Successful adoption requires organizational readiness, not just technology

As Schneider observed, "We're in a fundamental shift, like the browser, like mobile, AI is another huge shift. Things that used to take me days or maybe a week to complete are now done in less than a day."

The brands that thrive will be those that recognize this shift isn't about replacing human expertise. It's about amplifying it at scale.

‍

Want to learn how CommerceIQ's AI-powered platform can help your team move from insight to action faster? Request a demo to see agentic commerce in action.