ㅤ

Hosts Matt Johnson and Floyd Blaikie of Pivotree's Data vs. Commerce sit down with Anatolii Iakimets, Director of Product Marketing at Kibo. Anatolii takes the data side: an agent is a model plus a harness, the model is becoming a commodity, and the real work sits in the data and integration layer underneath.

ㅤ

The friction is clean. Floyd keeps reframing agentic commerce as a familiar platform decision, agents in a trench coat. Anatolii keeps pulling it back to the data. Commerce is deterministic. The price and the tax and the T-shirt size have to be exact, and an agent is only as good as the structured data it can reach and read.

ㅤ

👤 Guest Bio

Anatolii Iakimets is Director of Product Marketing at Kibo Commerce, where he focuses on B2C commerce. He has spent more than a decade in commerce and telecommunications, with earlier roles at Bold Commerce, Elastic Path, and Netcracker, and he has worked hands-on with AI and machine learning since around 2015. On this episode he takes the data side, arguing that agentic commerce succeeds or fails on data quality and integration, not on the model you pick.

ㅤ

📌 What We Cover

• The two parts of any agent: the LLM and the harness, the code that tells the model what to do
• Why the model is becoming a commodity, and why the switching cost for users stays low
• Why coding and text agents tolerate a slightly different answer every time, and commerce does not
• Why the price, the tax, and the T-shirt size have to be accurate to the point, not "good enough"
• Why an agent has to be integrated like an application, talking to your systems through MCP or an API
• How burning tokens turns into a real budget problem, with Uber's four-month AI burn as the warning
• The three things to weigh before you buy: composability, simplicity, and the ability to start small
• Why the "explain" function is the one customers reach for first, and the risk of a three or five year lock-in

ㅤ

🔗 Resources Mentioned

• Kibo (the guest's company)
• Anthropic (Claude) and OpenAI (ChatGPT) as frontier model providers
• Google as a frontier lab
• DeepSeek and Kimi K2 as open-weight models
• Model Context Protocol (MCP), APIs, and agent-to-agent protocols
• Microsoft Copilot's shift from subscription tiers toward usage-based pricing
• Uber's AI budget story
• Sam Altman on models becoming "intelligence on a tap"

ㅤ

Ryan took the data side. Before you buy the next AI tool, you finish step zero and step one: knowing where you're actually headed, then cleaning and governing the data those tools depend on. He and Matt get into why last year's shiny purchases went sideways, why distributors keep walking up to the table saying they're behind, and what happens now that customers show up with prototypes they built themselves. It's a distribution and manufacturing conversation, but retail and B2B operators will recognize the pattern.

ㅤ

👤 Guest Bio

Ryan Smith is an account executive at Pivotree, where he works with distributors and manufacturers on their AI and data transformation. He joined the company recently and spends his time helping companies figure out where they actually stand before they buy the next tool. He took the data side of this episode, arguing the foundation has to come before the technology.

ㅤ

📌 What We Cover

• Why most of last year's AI tool buyers hit a wall, and how much of it traced back to data that was never cleaned or governed
• Step zero, step one, step two: mapping where a distributor actually is before recommending anything
• Distributors building their own tools in-house, with AI engineers vibe coding bespoke solutions
• The security and governance risk when two departments build the same agent and spend the budget twice
• Customers arriving with ready-made prototypes built in Claude Code, then handing off the integration
• Why a working prototype is not a multi-tenant, secure, production-ready system
• The generational shift from tribal knowledge walking out the door to buyers who expect the Amazon experience
• AI fatigue, the 50/50 reliability problem, and why the human element still decides the outcome

ㅤ

🔗 Resources Mentioned

• Distribution Strategy Group (DSG) and the Applied AI for Distributors event
• Claude Code
• The earlier Data vs. Commerce episode with Bill Di Nardo and Joel Farquhar on RI plus AI
• Matt's breakout session on AI catalog management, on Pivotree's YouTube channel

ㅤ

👤 Guest Bio

Lauren McCullough is co-founder and CEO of Tromml, a vertically focused software company for the automotive aftermarket and industrial distribution. Tromml started on the analytics side, giving distributors visibility into what products were actually making money, and now builds field-sales tooling that captures rep conversations and turns them into next best actions. Lauren argues for keeping relationship-driven selling human while using data and AI to support it.

ㅤ

📌 What We Cover

• Why reps in distribution aren't just selling parts, they're selling relationships and trust, and why that changes how AI fits
• The difference between a system of record (put information in, pull a report out) and a system of action (surfaces the next best action and gets smarter over time)
• A rep's actual Monday workflow: prioritized accounts, optimized routes, a human-digestible briefing before the visit, and voice-note capture after
• The field signal that never reaches the boardroom, like a shipping delay or a launch that isn't landing, because nobody reports bad news up the chain
• Hiring implications: when the data lives in the system, you hire for emotional intelligence instead of category memory
• Where to point AI in a high-trust industry, and Lauren's blunt take on automating the customer relationship away

ㅤ

🔗 Resources Mentioned

• Tromml (Lauren's company; mobile conversation-capture app for field reps)
• Pivotree
• AI for Distributors event, Chicago (Matt's reference; the event's formal name is Applied AI for Distributors, run by Distribution Strategy Group)
• Claude (AI note-taking tool referenced by Sam)
• Salesforce (referenced as a CRM)
• Genuine Parts Company (referenced by Sam; see flag below)

ㅤ

His argument: taking an order is the easy part, and almost none of your hard problems live there. The friction starts on the execution side, when the promise date has to hold across phone, web, and field sales running on the same inventory. Keith makes the case for order management as the orchestration layer that sits on top of the ERP instead of working around it.

ㅤ

👤 Guest Bio

Keith Gorney has spent 25 years in direct-to-consumer, B2B, and order fulfillment. He started at Best Buy overseeing their enterprise order management platform, carried that into consulting roles touching URBN brands including Urban Outfitters, Free People, and Anthropologie, and now leads order management work at Pivotree. On this episode, he took the commerce side, arguing that the order is where commerce becomes real.

ㅤ

📌 What We Cover

• Why taking an order is easy, and the execution side is where the friction really starts to manifest
• The "real-time version of the truth" problem when phone, device, salespeople, and manual orders all race the same inventory
• What happens to the customer service rep stuck inside a legacy ERP, jumping between systems, emails, and phone calls
• Why canceling and recreating a six-figure order to change one line is effectively unacceptable
• How an OMS reads an order as pieces and parts, with line statuses and quantities, instead of one all-or-nothing object
• Selling a job three months out without freezing 1,000 units of inventory for a quarter
• Why this is a component-by-component migration, not a big-bang rip-out, with implementations Keith has seen go live in three months

ㅤ

🔗 Resources Mentioned

• Pivotree
• Best Buy, Urban Outfitters, Free People, Anthropologie (Keith's career background)
• ERP, OMS, global inventory visibility, BOPUS / curbside, DoorDash (referenced in conversation)

ㅤ

Dan took the data side, and the friction surfaced fast. The marketers want brand, lifestyle photography, and emotional copy to carry the search. The machine starts with UPC codes, GTINs, inventory, and shipping policy before it cares about any of that. The conversation covers the robots.txt settings that quietly block AI crawlers, the attribution problem when a shopper leaves ChatGPT and goes straight to a store, and why content marketing still earns its place once the data foundation is solid.

ㅤ

👤 Guest Bio

Dan Ornstein is Pivotree's Retail Industry Leader, focused on helping retailers grow revenue through unified commerce, customer experience, product data, and practical AI. Before Pivotree he was a Partner at KPMG Canada and a Director at Publicis Sapient, working across e-commerce, omnichannel, and loyalty. On this episode he took the data side, arguing that product data completeness, not brand copy, is what gets a retailer surfaced by AI shopping agents in the first place.

ㅤ

📌 What We Cover

• Why retailers suddenly see traffic from ChatGPT, Perplexity, and Gemini without doing anything to earn it
• The order an LLM works in: product data first, then price and availability, then third-party trust signals from sites like Vogue, GQ, or Reddit
• The robots.txt problem, where fraud and denial-of-service settings block the AI crawlers before they ever reach your catalog
• How subjective attributes like "soft" or "puffy and warm" have to become data the model can read, like down fill rate and temperature rating
• The attribution gap when a shopper exits ChatGPT and goes straight to the store, and why LLM referrals still convert at a higher rate
• Which categories suit agentic shopping now (grocery, hardware) versus where brand still drives the decision (fashion, home furnishings)
• What an e-commerce or merchandising leader should check tomorrow to confirm they show up at all

ㅤ

🔗 Resources Mentioned

• ChatGPT, Perplexity, Gemini (AI assistants surfacing retailer recommendations)
• Shopify (embedded LLM referral analytics)
• Vogue, GQ, Reddit (third-party reference sites the models check)
• Amazon (marketplace-seller comparison)
• TikTok, YouTube, Instagram (social channels referenced)

ㅤ

👤 Guest Bios

Bill Di Nardo is the Chief Executive Officer of Pivotree. He founded Grocery Gateway, one of Canada's first major pure-play ecommerce brands, and was named EY Young Entrepreneur of the Year in 2000 before leading the merger that became Pivotree.

ㅤ

Joel Farquhar is Pivotree's Chief Architect, with more than 25 years in enterprise commerce architecture and a seat on the MACH Alliance Technology Council. He came up through software development and now owns Pivotree's commerce architecture and platform strategy.

ㅤ

📌 What We Cover

• Why the "fewer humans, same output" pitch fails the math, and why winning teams handed experts a sharper tool instead of cutting them. ㅤ
• The reason customers don't pocket the savings from AI-assisted systems integration, and what they ask for instead. ㅤ
• How a lean company scales without hiring 100 more people, and why that doesn't mean shrinking the team. ㅤ
• Joel's matrix of domain knowledge against technical capability, and the multipliers, visionaries, and gatekeepers it produces. ㅤ
• Why Q1 was an experimentation quarter, and how some experiments quietly cost more time than they saved. ㅤ
• The move from billable hours to outcomes, and why CFOs spend when they have confidence the work will land. ㅤ
• The problem Bill says they haven't solved: where the next generation builds judgment when the entry-level work disappears.

ㅤ

🔗 Resources Mentioned

• Anthropic's Claude (the AI code assistant and tool referenced throughout)
• Adapt Relentlessly (Pivotree's core value, cited by Bill)

ㅤ

Matt Johnson and Floyd Blaikie open Episode 2 of Data vs. Commerce by tracing that delay past the product data they covered last time, into the place most teams forget to look: the integrations between systems. Matt walks through how an e-commerce system can confidently promise overnight delivery while the ERP and OMS behind it know the item is sitting in a warehouse across the country. The storefront wasn't lying. It just never got the truth from the back end.

ㅤ

From there the two get into why national retailers and distributors end up with a twisted ball of yarn: 30 acquisitions, 17 Salesforce instances, crusty middleware nobody wants to touch, and load-bearing systems you can't switch off without the business crawling to a halt. This is a solo episode, hosted by Matt Johnson and Floyd Blaikie of Pivotree. The friction this week isn't data against commerce. It's the invisible plumbing between them, and who actually owns it.

ㅤ

📌 What We Coverㅤ

• Why an e-commerce system can promise overnight shipping when the item is actually in stock across the country, in a different warehouse
• How distribution center location and on-shelf inventory live in back-end systems that often never connect to the front-end customer experience
• What growth by acquisition does to your stack: inherited ERPs, inventory systems, and disparate data that all have to talk to one storefront
• The 17 Salesforce instances problem, and why companies insist on keeping every system they acquire
• Brittle, hard-coded integrations and crusty middleware held together by one person in IT
• Who needs the macro view of the system architecture, and why nobody clearly owns it
• Where AI helps: reading APIs and connecting data faster than any human developer, plus agents that flag when an integration is about to break
• Where AI backfires: order management systems that hallucinate a part because the underlying data was wrong

ㅤ

🔗 Resources Mentioned

• Pivotree, named on the episode for its work on AI integration frameworks and processes
• Salesforce, referenced via the company running 17 separate instances
• The system categories at the center of the conversation: ERP, OMS, PIM, and the front-end e-commerce platform, plus the integration layer and middleware that connect them

ㅤ

The hosts kick off the show by naming the friction the whole podcast is built around. Data and commerce want the same outcome. They've stopped talking to each other in most organizations. Matt and Floyd, both of Pivotree, frame it less as a fight and more as couples therapy: a relationship that needs the communication rebuilt.

ㅤ

This episode is the premise. Future episodes pressure-test it with guests who take a side.

ㅤ

📌 What We Cover

• The dishwasher pump story: a manufacturer changed three pins to four mid-year, kept the same model number, and the distributor's product page never caught up.
• Why customer service catches the blame for an error that started in a supplier feed.
• The reason distributors and manufacturers run physical fulfillment at 99.9% accuracy and product data nowhere near that.
• What "operating on putting out fires" looks like inside a product data team.
• The contractor who orders the wrong part, takes the heat from their customer, and never trusts the website again.
• Why B2B e-commerce penetration at distributors still sits in the 10 to 15% range and how product data quality drives that ceiling.
• Data governance as a continuous lifecycle rather than a one-time integration project.
• The personal fitness analogy: product data discipline is a habit, not a launch.

ㅤ

🔗 Resources Mentioned

• Pivotree
• Matt Johnson on LinkedIn
• Floyd Blaikie on LinkedIn

ㅤ

The gap between the two is where modern commerce actually fails. Stockouts that aren't really stockouts. The 0.8% integration error nobody caught. The AI agent that ordered the wrong size. The B2B replatform that blew up on data nobody had cleaned in six years.

ㅤ

This show lives in that gap. Hosts Matt Johnson and Floyd Blaikie bring on the people who own the data, run the platforms, and pay the price when those two stop playing nice. New episodes drop weekly. If you're responsible for any part of how products get from a database to a doorstep, this is your show. Subscribe now.