Joe Pizarro works on product data and AI-powered commerce. Their posts describe building FeedZip.ai to help brands validate product data for agentic commerce, and developing Kato to assess supplement products using formulation, evidence, and safety context.
Joe Pizarro is useful for product data and agentic commerce insights, especially validation and recommendation limits.
Best for
A B2B ecommerce data platform shaping agentic commerce go to market and buyer education.
Not for
A consumer wellness brand seeking broad reach or a guaranteed supplement endorsement.
Key takeaways
Trusted product data is the control layer for agentic commerce.
Understanding a product and being allowed to recommend it are separate decisions.
They report internal results from 548 Shopify product cards, separating understanding from recommendation accuracy.
Check first: Confirm audience evidence, campaign reach, deliverables, disclosure, and willingness to endorse before contacting them.
Main ideas
Trusted product data is the control layer for agentic commerce · Opaque platform enforcement silently erodes trust · Understanding a product and being allowed to recommend it are separate decisions
Joe Pizarro works on product data and AI-powered commerce.
In their posts, they connect hands-on e-commerce operations with the newer problem of helping AI systems interpret products reliably.
They describe building FeedZip.ai to validate product listings and Kato to assess supplement products using formulation, evidence, shopper restrictions, and safety rules.
Their work offers a grounded view of agentic commerce, covering both the data infrastructure these systems need and the limits they should respect before recommending a product.
Pizarro’s distinctive contribution is treating product understanding and permission to recommend as separate problems.
In their Kato post, they describe a system that can identify missing or conflicting product information and abstain when supporting information or safety context is insufficient.
They report internal results from 548 Shopify product cards, distinguishing product-understanding scores from recommendation accuracy.
That qualification makes the example more informative than a broad claim that AI can shop for consumers.
The operational perspective runs through their other posts: they recall manually checking supplement labels at iHerb and managing catalog work at scale, then argue that accurate, trusted product data remains essential even as commerce tools change.
Readers interested in agentic commerce can follow how they turn that practical concern into product-validation and governance ideas.
Their posts establish work associated with FeedZip.ai, ShopLister, and Kato.
Joe Pizarro's posts argue that agentic commerce is arriving faster than the product data underneath it can support.
Read more
Their recurring priority is trust: product data that AI agents can read, validate, and rely on, with platforms, brands, and protocols aligned around that layer.
Pizarro pairs diagnosis with named fixes, including FeedZip.ai for listing validation, Kato for supplement product intelligence, and a proposed Universal Compliance Protocol.
A separate thread from their iHerb years presents catalog accuracy at scale as an old discipline built by careful people, not just better tools.
Trusted product data is the control layer for agentic commerce
Pizarro argues that finding a product and completing a purchase are no longer the hard parts.
Read more
Google, Shopify, Amazon, Meta, Walmart, and Stripe are aligning around standards such as the Universal Commerce Protocol, the Agent Payment Protocol, and a Gemini-powered cross-platform cart.
The missing layer, in their account, is trusted, machine-readable product data.
When data cannot be validated with confidence, agents skip the product rather than troubleshoot it, and demand moves to whoever has cleaner data.
Their reasoning: agents do not infer intent or give brands the benefit of the doubt, so data quality determines whether a product is considered at all.
In a post about Google I/O 2026, Pizarro lists the new commerce rails, then calls the product-data layer completely fragmented.
They argue that the industry needs more than transaction protocols and propose a Universal Compliance Protocol covering automated validation, policy alignment, machine-readable trust, and cross-platform governance.
They name FeedZip.ai as the compliance and trust engine they are building for that layer.
Pizarro argues that each major platform runs its own enforcement system without clear feedback to sellers: feeds get throttled, listings are suppressed or removed, and products are misclassified based on incomplete signals.
Read more
Most brands are not alerted; revenue drops, and teams blame creative fatigue, attribution, or seasonality.
Large brands absorb this with internal integrity engines and policy operations teams, while everyone else operates blind.
Pizarro wants to make that hidden layer explicit and governed, because the system will not scale when agents rather than people do the shopping.
They present these platform behaviors as a market argument, not as claims backed by platform documentation.
The same Google I/O post describes quiet feed throttling, suppression without warning, instant listing removals, and misclassification based on incomplete signals. Pizarro says trust is silently degrading inside platform systems, and brands need standardized validation and cross-platform governance instead of discovering the damage through falling revenue.
Understanding a product and being allowed to recommend it are separate decisions
Pizarro describes building Kato to bring a decision layer to agentic commerce.
Read more
Kato applies a governed supplement ontology to catalog data to understand ingredients, forms, and formulations; compare products using available product information and safety context; preserve shopper restrictions; identify missing or conflicting information; and decide when to clarify, abstain, or recommend nothing.
Their key argument: comprehension should not automatically grant permission to sell.
Eligibility, information, and safety are separate gates governed by deterministic rules.
Abstaining is intentional design, not a failure.
Pizarro connected Kato to Shopify Catalog through the Universal Commerce Protocol and reports an internal benchmark of 548 real Shopify product cards: 73.7% product-level understanding, 74.1% recognition of active components in multi-ingredient products, and 1.1% completely unknown product identity. They state that these figures measure product understanding, not recommendation accuracy, and that Kato understands far more products than it is willing to recommend.
Standards matter at the decision layer, not only the specification layer
Pizarro treats agentic commerce standards as practical infrastructure, not jargon.
Read more
They explain the market shift in plain language, then show what a specific protocol leaves unsolved. Their recurring priority is to connect protocol news to product-level decisions: what an agent can access, what it understands, and whether it should move a shopper toward a purchase.
One post explains agentic commerce in simple terms: AI agents compare, validate, and recommend products automatically.
It then says most brands are not ready because their data is not structured or compliant enough.
Another post shows the Universal Commerce Protocol in practice, describing how Kato was connected to Shopify Catalog through UCP for discovery, carts, and checkout.
Pizarro argues that UCP provides the commerce connection, while Kato adds supplement intelligence and decision boundaries.
Catalog accuracy at scale is an old discipline, sustained by people
Pizarro's origin story argues that careful catalog work is not new.
Read more
Before automation, AI, or systems thinking, they scanned supplement labels and typed ingredients, dosages, and claims by hand, then checked every word.
The post presents that repetitive verification as the foundation of trust at scale.
Pizarro prioritizes investing in people over polished setups: the founder invested in them, and they later taught HTML to warehouse colleagues.
They say different tools now sit atop the same problems, which can make broken systems easier to hide.
At iHerb in the early 2000s, Pizarro worked from a CRT monitor, a flatbed scanner, and a folding table overlooking the warehouse.
The team managed roughly 20,000 SKUs by pulling, verifying, updating, and repeating.
Pizarro taught HTML to warehouse colleagues; some became editors, photographers, or leaders.
Marshawn Knight is named as still a senior designer there.
This is the origin story behind Pizarro's later product-data work.
Read more
It shows the operational discipline they say still matters when tools change and explains why the agentic-commerce pitch centers on accuracy and trust rather than novelty.
Pizarro describes starting an e-commerce career at iHerb in the early 2000s, pulling supplements from a warehouse, scanning labels, and typing ingredients, dosages, and claims by hand, then checking every word.
The post says the team managed about 20,000 SKUs this way and that they taught HTML to warehouse colleagues, naming Marshawn Knight as a designer still on the team.
They frame the work as slow, repetitive, accurate, and necessary, and link it directly to today's problem of messy inputs and accuracy at scale.
For brands planning a campaign
For a B2B ecommerce data platform, this post offers ready-made credibility for a go-to-market narrative: it grounds Pizarro in catalog operations and people development, not just AI commentary, and supports messaging about data quality as an old problem with new urgency.
This is Pizarro's clearest statement of the market thesis: what problem the company is solving and how they connect protocol news to a commercial argument about data trust.
Read more
Pizarro argues that Google, Shopify, Amazon, Meta, Walmart, and Stripe are aligning around agentic commerce foundations such as the Universal Commerce Protocol, Agent Payment Protocol, and a Gemini-powered Universal Cart, but that the critical layer is still missing: trusted, machine-readable product data.
The post says platforms enforce opaque rules, quietly throttle feeds, suppress listings, or remove products, while brands often blame creative fatigue, attribution, or seasonality.
They call for a Universal Compliance Protocol covering automated validation, policy alignment, machine-readable trust, and cross-platform governance, and say this is the layer being built at FeedZip.ai.
For brands planning a campaign
A platform hiring for go-to-market can use this as a thought-leadership spine: position the product around compliance and trust as the control layer of agentic commerce, and use platform-enforcement examples to make the cost of bad data concrete for brands.
This post moves from diagnosis to a working product with numbers and guardrails.
Read more
It shows how Pizarro evaluates an AI commerce system and why they treat the ability to abstain as a feature, not a failure.
Pizarro describes connecting Kato to Shopify Catalog through the Universal Commerce Protocol so that an AI agent can understand supplement formulations, compare products using product information and safety context, preserve shopper restrictions, identify missing or conflicting information, and decide when to clarify, abstain, or recommend nothing.
The post reports an internal benchmark of 548 real Shopify product cards: 73.7% product-level understanding, 74.1% recognition of active components in multi-ingredient products, and 1.1% completely unknown product identity.
They specify that these figures measure understanding, not recommendation accuracy, and that understanding and permission to recommend are separate gates governed by deterministic rules.
For brands planning a campaign
For a product data platform, this shows evaluation discipline and safety-gated recommendation logic. It supports go-to-market angles about agent-readiness, evidence review, and the difference between understanding a product and being allowed to sell it.
This is the shortest, most accessible entry point to Pizarro's commercial offer.
Read more
It explains what the tools do in plain terms before readers encounter protocols, benchmarks, or ontology language elsewhere.
Pizarro explains agentic commerce as the next wave of ecommerce, in which AI agents compare, validate, and recommend products automatically.
The post says most brands are not ready because their product data is not structured or compliant enough for AI to understand.
It positions ShopLister and FeedZip.ai as ways to prepare brands and check whether listings are AI-ready, with links to ShopLister.io and the FeedZip.ai validation tool.
For brands planning a campaign
This shows how Pizarro translates a technical shift into simple demand-generation copy and connects it to a validation-tool funnel. A go-to-market team could use this style for top-of-funnel content and to test whether buyers understand their own AI-readiness gap.
Publishing style: plain framing, structured lists, named products4
Product data quality as the blocker for agentic commerce3
Protocol literacy across agentic commerce standards3
Platform enforcement as a trust and compliance problem1
Supplement intelligence with explicit decision boundaries1
Accuracy at scale, built by people1
Product data quality as the blocker for agentic commerce
Pizarro's central argument is that AI agents can already discover and buy products, but the data underneath them is fragmented and unverified.
Read more
One post describes brands whose product data is not structured or compliant enough for AI to understand.
Another argues that agents do not troubleshoot or infer intent: if a product cannot be validated with confidence, the agent skips it and demand shifts to whoever holds cleaner data.
A third notes an agent can reach a product without understanding its formulation, the support for its claims, or whether it fits the shopper's restrictions.
Each diagnosis is paired with a named fix: a FeedZip.ai listing-validation tool, a Kato decision layer, and a proposed Universal Compliance Protocol for automated validation, policy alignment, and cross-platform governance.
For brands planning a campaign
For a B2B ecommerce data platform, this is the core narrative an expert can lead go-to-market with: agents are the demand channel, data trust is the constraint, and validation is the product. Useful for positioning, analyst-style content, and sales enablement around AI-readiness audits.
Platform enforcement as a trust and compliance problem
One post describes enforcement as opaque and uneven: feeds quietly throttled, listings suppressed or removed without warning, and products misclassified on incomplete signals.
Read more
Most brands are not alerted; revenue drops, and teams blame creative fatigue, attribution, or seasonality while trust degrades inside platform systems. Pizarro argues that large brands absorb this with internal integrity and policy teams while everyone else operates blind—and that this does not scale when agents, not people, are the shoppers.
The post attributes specific enforcement behaviors to named platforms without citing documentation. Treat this as a stated argument about the market, not as verified platform policy.
For brands planning a campaign
A strong content angle on feed governance, listing integrity, and policy operations for brands selling through marketplaces. For reuse, keep it as the speaker's claim and pair it with verifiable platform documentation the client holds.
Supplement intelligence with explicit decision boundaries
Pizarro describes connecting Kato to Shopify Catalog through the Universal Commerce Protocol, applying a governed supplement ontology to real catalog data so it can read ingredients, forms, and formulations; compare products using information supporting their claims and safety context; preserve shopper restrictions; flag missing or conflicting information; and decide when to clarify, abstain, or recommend nothing.
Read more
An internal benchmark of 548 real Shopify product cards is reported at 73.7% product-level understanding, 74.1% recognition of active components in multi-ingredient products, and 1.1% completely unknown product identity. The post states these measure understanding, not recommendation accuracy.
Understanding and permission to recommend are deliberately separated gates, governed by deterministic rules for eligibility, evidence, and safety. The post presents fewer recommendations than products understood as intentional.
For brands planning a campaign
Shows how a data platform can sell its way of working: define the ontology, quantify with bounded metrics, and show where the system refuses rather than guesses. Good raw material for a campaign about trustworthy recommendation, abstention, and category-specific intelligence.
Protocol literacy across agentic commerce standards
Two posts trace the 2026 alignment around agentic commerce standards, naming the Universal Commerce Protocol, Agent Payment Protocol, and a Gemini-powered cross-platform universal cart, with Google, Shopify, Amazon, Meta, Walmart, and Stripe involved.
Read more
One post uses UCP practically to connect a catalog to agents; another names Google, Shopify, OpenAI, and TikTok as forces moving the market and explains agentic commerce in plain terms as AI agents comparing, validating, and recommending products.
Standards are explained at a decision level rather than a specification level: what the protocol enables and which layer it leaves unsolved.
For brands planning a campaign
Supports explainer and education content that places a platform inside the emerging standards stack. Date any reuse and check protocol details against current specifications.
An origin-story post describes early-2000s work at iHerb: a CRT monitor, a flatbed scanner, and a folding table beside the warehouse, pulling products and hand-typing ingredients, dosages, and claims, then double-checking every word.
Read more
At one point roughly 20,000 SKUs were managed this way—pulled, verified, updated, and repeated.
The founder invested in Pizarro, who later taught HTML to warehouse colleagues; some became editors, photographers, or leaders, and Marshawn Knight is named as still a senior designer there.
The post closes on trust, value, and simplicity.
Repeated manual verification treated as a system before it had a name, alongside investment in people rather than tooling. Pizarro notes that different tools now sit on the same underlying problems, making broken systems easier to hide.
This is a reflective personal narrative without dates or externally checkable metrics; it should be read as stated experience rather than verified record.
For brands planning a campaign
Useful for founder-brand and values content that humanizes an infrastructure pitch: catalog accuracy as a long-standing discipline, and team-building as part of what a platform brings to a go-to-market role.
Publishing style: plain framing, structured lists, named products
The posts share a readable format: a short hook, a plain-language definition of the trend, then arrowed or checked lists of the problem, the missing layer, and the fix, closing on a product name and a question to the reader.
Read more
Technical posts use protocols and benchmark numbers; the reflective post uses scene-setting narrative. FeedZip.ai and Kato are named directly each time, and supplements appear as a deliberate specialization rather than a one-off.
For brands planning a campaign
Copy-friendly structure for campaign assets: short explainers with a clear problem, a named solution, and a call to compare data readiness. The category-specific depth in supplements is a differentiator worth carrying into briefs aimed at health and wellness brands.
Joe Pizarro's posts describe four commercial threads: they describe working at iHerb in the early 2000s as the start of an e-commerce career, ongoing work on product-data tools under the ShopLister and FeedZip.ai names, and building Kato, a supplement product-intelligence tool, then connecting it to Shopify's Catalog through the Universal Commerce Protocol.
Read more
They say “we” built the FeedZip.ai validation tool but do not specify their role or relationship to that business. Ask about any sponsorship, affiliate, or client relationships before engaging.
iHerb
Employment
Pizarro describes iHerb.com in the early 2000s as “the foundation of my e-commerce career,” recalling scanning supplement labels and typing ingredients, dosages, and claims by hand across roughly 20,000 SKUs, then teaching HTML to warehouse colleagues.
Pizarro writes that “at ShopLister, we help brands prepare for the agentic era,” positioning the business around making product data structured and compliant enough for AI agents to read.
Read more
Ask what their role and relationship to ShopLister are.
Pizarro describes FeedZip.ai as a validation tool that checks whether product listings are AI-ready, and later as “the compliance and trust engine for agentic commerce” that “we're building”—an automated validation, policy alignment, and cross-platform governance layer for agentic commerce.
Read more
Ask whether they own, work for, or consult to the business.
Pizarro says “I built Kato,” a decision layer for supplement products in agentic commerce.
Read more
It applies a governed supplement ontology to catalog data so it can compare formulations, preserve shopper restrictions, and abstain from recommending rather than guess. The post reports an internal benchmark of 548 real Shopify product cards.
Joe Pizarro writes short, structured LinkedIn posts about the gap between agentic commerce standards and the product data underneath them.
Read more
Start with the Google I/O post, which names the Universal Commerce Protocol, the Agent Payment Protocol, and a Gemini-powered cross-platform cart.
They argue that trusted, machine-readable product data is the layer still missing and point to FeedZip.ai as the validation tool being built for it.
The Kato post is the most technical: it connects a supplement catalog to Shopify Catalog through UCP, reports an internal benchmark of 548 real Shopify product cards, and explains why understanding a product and being permitted to recommend it are deliberately separate gates.
The iHerb post is personal, recalling hand-scanning and typing supplement labels in the early 2000s, managing roughly 20,000 SKUs, and teaching HTML to warehouse colleagues.
Across the posts, recurring priorities include data trust, validation, abstaining when evidence is thin, and treating careful catalog work as something people sustain rather than something software replaces.
Read the Google I/O post for the diagnosis and the Kato post for the working method.
Both write about the catalog layer beneath agentic commerce.
Read more
Ojha describes Fegmo deploying autonomous AI agents to ingest messy merchant inputs, generate structured attributes and product imagery, and syndicate them across channels.
They also describe becoming GDSN-ready through a Commport partnership, allowing standardized product data to flow into ecommerce, marketplaces, and AI shopping experiences.
Pizarro covers the same territory from the trust and validation side.
Ojha’s posts frame company momentum and capability, announcing two new enterprise contracts and a partner certification to argue that structured catalog operations are urgent and funded.
Pizarro focuses on diagnosis and decision quality: opaque platform enforcement, why agents skip products they cannot validate, and a supplement ontology that deliberately refuses to recommend.
Ojha shows what the platform does; Pizarro shows where it should stop.
Shivers writes a similar protocol explainer for a commerce audience, walking through OpenAI and Stripe’s Agentic Commerce Protocol and then Google’s Universal Commerce Protocol: what each enables, how merchants keep ownership of the transaction, and what a structured feed now determines.
Read more
Both treat agentic commerce as practical infrastructure rather than jargon and connect standards to product feed readiness.
Shivers teaches through merchant checklists—for example, providing a structured feed, implementing checkout endpoints, and supporting delegated payments—and closes with an offer of a feed-readiness checklist.
Pizarro goes deeper, asking whether a product should be recommended at all and setting out governed rules for eligibility and safety, plus category-specific benchmarks.
Shivers is broader and more how-to; Pizarro is narrower, focused on decision boundaries.
Van Camp also addresses the readiness gap in agentic commerce, citing a GEO Readiness Analyzer that found fewer than 1% of scanned product pages scored high enough for AI agents to reliably surface them.
Read more
They describe product enrichment and direct connections to LLM platforms so catalogs can be discovered, recommended, and transacted at scale. Their core claim—that most merchants have no way to capture agent-driven demand—matches Pizarro’s.
Van Camp frames readiness through a platform launch and joint research, tying the gap to marketplace growth rates and a Mirakl-Merkle playbook that positions marketplace infrastructure as the structural advantage.
Pizarro reports an internal benchmark from their own tool covering 548 product cards, states plainly that it measures understanding rather than recommendation accuracy, and treats abstention as intentional.
Van Camp argues for an infrastructure bet; Pizarro argues for governed data trust within one category.
The audience information in Joe Pizarro's posts is a professional-following figure rather than a demographic picture: the LinkedIn account behind these posts reports 1,964 followers as of 8 October 2026.
Read more
For a B2B ecommerce data platform deciding whether to bring them in to shape go-to-market, that points to a modest, LinkedIn-centred professional reach — useful for distributing practitioner arguments, not a basis for promising large-scale campaign impressions.
The four posts are written for brand-side operators: catalogue and product-data teams, feed and compliance owners, and supplement or DTC merchants preparing for AI-agent shopping.
They address that reader directly and repeatedly — "Most brands aren't ready.
Their product data isn't structured or compliant enough for AI to understand" — and the recurring subjects (product-data trust, opaque platform feed enforcement, supplement formulation and safety, UCP/cart/checkout standards, catalog integration through Universal Commerce Protocol) sit squarely inside the world an ecommerce data platform sells into.
The posts don't show who reads them, where those readers are based, which companies employ them, or their job seniority.
Placement decision: they fit a campaign aimed at practitioner-facing authority with product-data and agentic-commerce decision-makers, with reach expected to come from LinkedIn posts and direct professional relationships rather than broad or consumer-scale distribution.
If the campaign requires assurance about audience geography, company size, or seniority mix, ask for a dated LinkedIn analytics report or audience survey results covering follower geography and job function.
A platform demographics report of the kind YouTube provides can inform geography, age, and gender, but no such report or video performance data were provided here; that type of data would not verify employer, occupation, or seniority.
Before committing, ask them for recent post-level results (impressions, reach, engagements) for topics, depth, and formats comparable to the planned campaign, together with follower geography and job-function breakdowns from LinkedIn analytics or a documented audience survey.
Joe Pizarro’s four LinkedIn posts describe building and promoting FeedZip.ai, ShopLister, and Kato, not paid collaborations with outside brands.
Read more
They publish no rate card, media kit, sponsorship policy, disclosure statement, deliverable specifications, or inquiry route.
A buyer would need to ask directly about availability, terms, deliverables, and disclosure requirements.
Joe’s commercial work includes building FeedZip.ai as a compliance and trust engine for agentic commerce and Kato as a supplement product-intelligence layer connected to Shopify Catalog through the Universal Commerce Protocol.
A brand considering Joe as a partner is more likely to be seeking a subject-matter expert, advisor, or product collaborator than a channel with defined sponsored-content inventory.
Good fit: a B2B ecommerce data platform needs a practitioner to shape its agentic-commerce go-to-market. Pizarro’s posts connect product-data operations to the newer problem of making catalog information usable and trustworthy for AI agents. They describe manually checking product ingredients, dosages, and claims at iHerb, then managing about 20,000 SKUs; newer posts discuss feed validation, commerce protocols, and how agents should handle missing or conflicting product information. That gives them relevant material for positioning, buyer education, and sales-enablement work—not just trend commentary. One useful distinction in their Kato post is that understanding a product is not the same as being justified in recommending it: the post describes separate eligibility, evidence, and safety gates, and explicitly says its reported benchmark measures product understanding, not recommendation accuracy. A campaign could use that distinction to make a platform’s trust claims more concrete. Pizarro’s LinkedIn following is modest. Ask for audience evidence and campaign-reach expectations before booking, and agree on scope, deliverables, and disclosure directly.
When to choose a specialist
Read more
Poor fit: a consumer wellness brand wants a broad-reach sponsored endorsement of a supplement, with a guaranteed number of impressions or a direct product recommendation. The posts show category knowledge and discussion of supplement data, but do not establish consumer audience reach, endorsement formats, or sponsored-content terms. In the Kato example, Pizarro stresses that a system should sometimes clarify, abstain, or recommend nothing, and that product understanding is not recommendation accuracy. That approach suits a campaign about responsible product data, but it is not evidence that they will endorse a particular supplement or that a consumer-facing promotion will deliver reach. Before considering a different brief, ask about willingness to endorse, audience evidence, proposed deliverables, and any product-safety or claims review.
A worked campaign example
Worked angle for a campaign aimed at a B2B platform: “An agent can find your product and still have no sound basis to recommend it.” Build a short LinkedIn explainer or live working session around one anonymized catalog record with incomplete or conflicting attributes. First show what an agent can read; then identify what it cannot safely infer; finally demonstrate how validation, posts, and an explicit decision to clarify or abstain change the outcome. Pizarro’s posts provide the narrative ingredients: their account of checking catalog details by hand, the argument that fragmented product data can make brands invisible to agents, and the description of Kato’s separate understanding and recommendation gates. Keep any benchmark figures clearly attributed to the internal test described in the posts and do not present them as proof of the client platform’s performance. A practical next step is a discovery session with the platform’s product and go-to-market teams to select a representative catalog example and verify the claims before publication. Treat this as a proposed campaign angle, not a published offer or confirmed service.
Consider Abhishek Ojha when the platform needs an enterprise catalog-operations story: a founder who talks about automating messy supplier inputs, enriching product attributes and assets, syndicating across channels, and connecting to standards-based exchange.
Read more
The posts describe Fegmo signing enterprise grocery and multi-brand retail contracts, deploying autonomous AI agents for product experience operations, building a closed loop from performance signals back into enrichment, and becoming GDSN-ready through Commport so national-brand onboarding can move from weeks to minutes. That fits a campaign that needs retailer and supplier integration language, PIM/PXM/DAM functionality, and enterprise credibility rather than a general agentic-commerce explainer.
The tradeoff is that Ojha's posts focus on company and partnership announcements; they do not show the step-by-step catalog record analysis, validation gate, or abstention logic described in Joe P.'s Kato post. If the platform's differentiator is governed product understanding and safe recommendation boundaries, Ojha is a stronger fit for enterprise sales and GDSN workflow framing than for explaining how an agent should decide not to recommend a product.
Consider Tom Shivers when the need is buyer education about agentic commerce protocols and retailer readiness.
Read more
The posts explain the difference between OpenAI and Stripe's Agentic Commerce Protocol and Google's Universal Commerce Protocol, stress that ACP keeps the retailer as merchant of record, list practical steps for retailers to work with ACP, and translate UCP into the structured feed is the new homepage and the product graph is the new SERP.
Shivers also says they are preparing a feed-readiness checklist for auditing ecommerce brands.
That fits a campaign aimed at helping merchants understand protocol choices, feed requirements, and where their data determines agent eligibility.
The tradeoff is operational depth.
The posts are protocol explainers and recommendations; they do not show original catalog remediation work, validation results, or a product-data system being built.
Shivers's posts include a firsthand account of manual SKU processing and a described internal product-understanding benchmark.
Choose Shivers for clear protocol and feed-readiness education or for a practitioner narrative about catalog data quality and recommendation governance.
Consider Amelia Van Camp when the platform needs enterprise market framing and demand-generation claims rather than a builder's implementation story.
Read more
The posts cite Mirakl's GEO Readiness Analyzer finding that less than 1% of scanned product pages scored above the threshold where AI agents reliably surface products, reference Salesforce's $67 billion Cyber Week 2025 agent-influenced sales figure, present Mirakl Agentic Activation capabilities for enrichment and direct LLM channels, and describe a Mirakl-Merkle playbook on the readiness gap, AI traffic growth, and marketplace GMV growth. That suits a campaign that must size the problem, reach enterprise buyers, and position a platform against agency-led research and marketplace infrastructure.
The tradeoff is vendor and category framing.
Van Camp's posts are launch and partnership announcements around Mirakl and Merkle; they do not provide a hands-on walkthrough of validating an individual catalog record or deciding when an agent should abstain.
The posts are closer to the product-data operating problem, including a described supplement ontology and separate understanding and recommendation gates.
Choose Van Camp for market sizing, enterprise readiness arguments, and partner-led demand generation, or for practitioner detail on product data quality and recommendation boundaries.
Joe Pizarro's four LinkedIn posts cover an explainer on agentic commerce and FeedZip.ai, a technical post on Kato and its Shopify Catalog connection through the Universal Commerce Protocol, a post naming FeedZip.ai as a compliance and trust layer, and an early-career account of catalog work at iHerb.
The posts have no publication dates, so the examples are undated.
The benchmark figures are Pizarro's own internal measures rather than independently verified results, and the descriptions of platform enforcement are presented as his market argument.
The posts connect Pizarro's work to FeedZip.ai, ShopLister, and Kato, but do not cover books, shows, or other earlier public work; that remains an open gap for anyone searching the name beyond these posts.
Comparisons with similar creators draw on two posts each from Abhishek Ojha, Tom Shivers, and Amelia Van Camp; follower, engagement, or audience data is not included, and no named employer or account beyond the posts is confirmed.