SupplierWiki curates detailed articles to help suppliers win at retail.

Don't let Walmart Post-Audits catch you off guard in 2025. Join Allie and Bekah to learn best practices for preventing and disputing Post Audits.

Prepping for Walmart Audits

Post Audits are preventable and disputable! Join Allie and Bekah to learn how to best prepare for Walmart Post Audits.

Articles

Why Item Setup Matters: A Guide for New Retail Suppliers 

For many first-time suppliers, item setup feels like a one-time form to complete as quickly as possible. In reality, it's a key foundation of your retail business. Once your data moves through a retailer's systems, many setup decisions become difficult or impossible to change. A small mistake, like entering the wrong number or assigning the wrong product code, can disrupt the entire supply chain: failed orders, invoice deductions, shipment delays, and lost online visibility. 

Product Codes: The Most Common Failure Point 

UPCs and GTINs (product identification codes) are one of the most common causes of setup failures because they act as the unique digital identifier for your product. Retailers use these codes to recognize what a product is, who supplies it, how it should be shipped, and how it should be sold. If the code is wrong, the retailer's systems cannot correctly process the item. 

This is why major retailers like Walmart and Target require suppliers to use codes registered through 

, the global organization responsible for product identification standards. Registering through GS1 ensures that every code is unique and tied to a specific product and packaging level. Retailers regularly verify these codes against GS1 databases to confirm that supplier data is legitimate and standardized. 

A kimchi company registers three SKUs with GS1: 

If they accidentally submit the 16 oz code for the 32 oz jar, Target's system gets confused. The 32 oz shipment scans as the smaller size, causing inventory miscounts, wrong pricing, and shelf stocking errors. The retailer flags the supplier and requires manual corrections before accepting future orders. 

 One wrong digit disrupts the entire supply chain, which is why GS1 registration and code verification are non-negotiable. 

Physical and Digital Data Must Match Exactly 

Item setup involves much more than administrative work. The dimensions, weights, case pack sizes, and pallet configurations you enter become the digital instructions retailers use to move, store, and replenish products throughout their supply chains. 

Large retailers operate highly automated distribution networks. Their systems use the dimensions and weights you provide to determine warehouse placement, truck loading, replenishment forecasting, and shelf space allocation. When the physical product does not match the digital item file, the system cannot process the shipment correctly. 

This is why "close enough" measurements create serious operational problems. If your actual carton dimensions differ from what you entered during setup, retailers may trigger receiving exceptions at distribution centers. These exceptions slow down shipments, create warehouse bottlenecks, and sometimes result in rejected freight. 

Cultured Roots enters their kimchi case dimensions as 12" × 10" × 8" during setup. Their warehouse actually packs them at 12" × 10" × 9". 

When a pallet arrives at Target's distribution center, the automated system expects cases to stack a certain way based on the 8" height. The 9" cases don't fit the calculated pallet pattern. The system flags a receiving exception, the pallet gets pulled for manual inspection, and the shipment is delayed by 24-48 hours while warehouse staff resolve the discrepancy. 

 Even small measurement errors trigger costly delays and manual interventions. Physical products must match digital data exactly. 

Each packaging level requires its own unique product code. These levels include the individual unit, inner pack, case, and pallet. Retailers use these codes to understand exactly what configuration is moving through the supply chain. Using the same code for multiple packaging levels is one of the most common and expensive supplier mistakes. 

For example, Target distinguishes between an outer shipping carton (called a "case pack") and an inner carton (which they call a "store pack"). Each one needs its own unique code. If you don'tunderstand these distinctions, you may assign the same code to both, which will cause your orders to fail or shipments to be rejected. 

Cultured Roots sells individual 16 oz jars that are packed with 12 per case. They need two codes: 

If they mistakenly use the same code for both, Target's system can't distinguish between ordering a single jar and ordering a full case. Orders fail or shipments get rejected at receiving because the code doesn't match the actual quantity being shipped. 

 Each packaging level needs its own unique code, or the retailer's system cannot process orders correctly. 

Many new suppliers estimate dimensions or weights before final production samples are complete. However, this creates a serious problem: Once your data is entered and synchronized through certain retailer systems, it can be cumbersome or problematic to change it later. 

If you estimate a weight or dimension and need to correct it later, you cannot simply enter a new number. Instead, you must resubmit the entire item information through the system, which requires creating new setup spreadsheets and going through approval again. This causes launch delays and increases administrative costs significantly. 

Retailers also actively validate this information before approving products. For example, Target uses specialists to review the dimensions, weights, and packaging attributes you provide. If the information appears incomplete or inaccurate, the item may never move into "approved for ordering" status. 
 
Real World CPG Example 

Cultured Roots estimates their kimchi jar weight at 18 oz during initial setup. After production, they discover it's actually 19.2 oz. They can't simply update the number in Target's system

— they must resubmit the entire item file, create new setup spreadsheets, and go through approval again. This delays the launch by weeks and adds significant administrative costs. 

Target's specialists also validate all submitted data before approval. If dimensions or weights appear incomplete or inaccurate, the item stays stuck in "pending" status and never reaches "approved for ordering." 

 Get your data right the first time. Corrections after submission are costly and time-consuming. 

Many suppliers treat product hierarchy and category information as simple back-office administration. However, this data plays a critical role in controlling how retailers organize, replenish, search, and merchandise products across their systems. 

Retailers use category structures to determine where products belong within departments, assortments, and reporting categories. If an item is categorized incorrectly, retailer systems may struggle to process replenishment forecasts or authorize you to sell within that category. It can impact businesses in the following ways:  

Retailers also use this information to build shelf placement plans and track supplier performance. If setup data is inaccurate, products may appear in the wrong section online or in-store, while your performance metrics may become distorted. 

Cultured Roots lists their kimchi under "Condiments" instead of "Fermented Foods & Probiotics." Target's system doesn't recognize them as authorized to sell in the Condiments category, so purchase orders fail to generate. Online, the product disappears from probiotic searches entirely because the category hierarchy doesn't connect it to that attribute. 

: No orders, visibility, or sales, even though the product is in Target's system. 

 Incorrect category data blocks purchase orders and hides products from customers. Category structures control authorization, replenishment, search visibility, and performance tracking. 

Compliance Requirements: Hazmat, Country of Origin, Ingredients, and Materials 

Hazmat, country of origin, ingredient, and material disclosures are among the most important compliance fields in item setup. Unlike product descriptions, these disclosures are tied directly to legal requirements, safety regulations, and retailer approval workflows. Missing or inaccurate information can delay or completely block an item launch. 

Hazardous material disclosures are required for products containing chemicals, aerosols, batteries, pesticides, or other regulated materials. Retailers use a third-party compliance assessment/platform called WERCS to evaluate hazardous materials before products can move through the supply chain. 

In many retailer systems, you must complete a hazmat review before you can finish item setup. Importantly, if you incorrectly flag a non-hazmat item as requiring a hazmat review, the system becomes blocked. Fixing this mistake requires contacting retailer support to reverse the flag. This correction process is time-consuming and delays your launch. 

Different Retailers, Different Requirements 

Every retailer uses different systems, terminology, and item setup requirements. Walmart uses its own supplier portal, Target uses different systems, and other retailers maintain their own supplier platforms and data structures. 

As a result, a product setup that works for one retailer may not work for another. Differences in category structures, operational attributes, packaging requirements, and code standards often force suppliers to manage multiple versions of the same product data. 

Without centralized item data management, suppliers usually end up manually updating information across several retailer systems. This increases the risk of inconsistent data, shipment rejections, compliance fines, and payment deductions. Many suppliers use centralized data management platforms to maintain consistent product information across retailers and reduce manual entry errors. 

Cultured Roots completes their item setup correctly before submitting to Target.  

They obtain official GS1 codes for each packaging level, measure their actual production samples (not estimates), and map their products to the correct category hierarchy. They also complete all compliance requirements upfront: country of origin, ingredient disclosure, allergen information, and hazmat review (if applicable).  

They provide Target with a single, comprehensive item file with no missing or inaccurate data. 

 Target's systems immediately recognize the products, purchase orders generate without delay, inventory counts accurately, shelf space allocates correctly, products appear in the right online searches, and compliance approval completes without blocking the launch. No exceptions, no rejections, no corrections needed. 

Item setup is the operational foundation of your retail business. Every mistake in your data creates friction later with blocked orders, warehouse rejections, payment deductions, and lost visibility. Get it right the first time by verifying codes through GS1, measuring actual products, assigning correct categories, and completing compliance requirements fully. The time you invest now will save you from costly delays and rejections that could derail your retail launch. 

Luckily, you don’t have to do it all by yourself. 

 manages item data across your entire retailer network, ensuring codes stay synchronized, dimensions stay accurate, and compliance requirements stay current. Learn how SPS helps you keeps item data reliable across retailers.  

Common Failure Points in DIY Supplier Integration Workflows 

____________________________________________________________________________________ 

Most DIY supplier integration workflows don't fail at launch. They fail days, weeks, or even months later, when issues compound to a level that can’t be ignored any further.  

The failure modes are predictable. They follow a consistent pattern: complexity that wasn't anticipated at build time (edge cases in data formats, exception volumes, trading partner spec variations) compounding through organizational gaps that show up when something goes wrong. 

This article maps those failure points, so that IT directors and operations managers can recognize their own situation and react accordingly. It’s a pattern recognition guide built from what actually breaks, in what order, and under what conditions.Teams that understand these failure modes before they hit them are better positioned to either prevent them or make a clear-eyed decision that managed infrastructure is the right fit for their situation. 

Why DIY Integration Looks Fine... At First 

At first glance, DIY integrations typically seem to be performing well. The “happy path” (a clean purchase order comes in, maps correctly, transmits without error) is what the team tested, and it performs exactly as designed. The trading partner count is small; volume is manageable; the spec documents are current because you just pulled them. Everyone who built the system is still on the team and knows how it works. 

This is the condition that creates false confidence. The edge cases in your trading partners' data haven't appeared yet because you haven't seen enough volume. The exception queue is empty because you don't have enough transactions to generate meaningful exceptions. The spec documents are current today; the question is whether your process will catch it when they change.  

The problems are waiting for the conditions that surface them: a new trading partner with a non-standard implementation; a volume spike that exposes a scalability issue; a retailer update that lands sideways on a map nobody has touched in a year. 

None of this is an argument against building in-house. DIY integration is a legitimate approach, and for the right team with the right resources, it can work well. The goal here is to build it with clear eyes and to understand where these workflows characteristically break down so you can either design against those failure modes or make an honest assessment of whether your team has the capacity to manage them long-term. 

The following failure points are the patterns that surface repeatedly as DIY supplier integration workflows age. 

Failure Point 1: Underestimated Data Format Variability 

The assumption most DIY builds start with is that each retailer has one EDI spec, and once you've mapped to it, that trading partner is handled. The actual picture is more complicated. X12 transaction sets and GS1 standards define a framework, but implementation varies significantly:  

This gap is usually invisible at launch. Your first few trading partners are chosen carefully, onboarded slowly, and tested thoroughly. The problem surfaces when you scale (e.g., more trading partners, more SKUs, more transaction volume) and the variations you never encountered in testing start appearing in production. 

Your workflow handles Target without issues. Then you onboard a regional grocer running a non-standard EDI 850 implementation. The fix requires modifying logic that was built eighteen months ago by someone who has since moved to a different role, and the documentation doesn't reflect the decisions that were made at the time. 

Typically, ERP systems don’t include EDI. Instead, EDI runs alongside your ERP in a separate layer that translates data between your internal systems and your trading partners. In a managed integration environment, that layer stays synchronized with your ERP by design. In a DIY build, keeping them in sync is a responsibility that belongs to your team, and it's one that's easy to underestimate at build time. 

The structural problem is dual maintenance. When a product code, ship-from location, or anything else changes in your ERP, someone has to make the corresponding update in the integration layer. There's no automatic propagation. The integration layer doesn't know your ERP changed but rather keeps running on the old data until something breaks. 

This failure mode is particularly nasty due to its silence. The workflow continues to process transactions. It just processes them with stale data. Nothing flags an error until a trading partner rejects an order or a chargeback surfaces downstream, at which point the discrepancy may have been present for weeks. 

A product code gets updated in the ERP as part of a routine SKU rationalization. The integration layer doesn't get updated. The next 850 that comes in for that product maps to a code that no longer exists in your system. The partner rejects the order. By the time the rejection routes back to someone with the context to diagnose it, the fulfillment window has passed. 

Failure Point 3: Exception Handling That Was Never Built 

DIY integration builds are designed around the expected path. A purchase order comes in, maps correctly, transmits without error, and a functional acknowledgment comes back: That's the workflow that gets built, tested, and signed off on. 

What doesn't get the same attention is what happens when the path isn't clean, such as when a transaction fails validation or when an acknowledgment comes back with a rejection code. 

At deployment, this feels like a manageable gap. Exception volumes are low, the team knows the system well enough to handle issues manually, and the few problems that do surface get resolved quickly. But six months in, with more trading partners and more SKUs, the picture is different. The exception queue is growing. Resolution is inconsistent. What started as an occasional interruption becomes a recurring burden. 

The hidden cost is that manual exception resolution at volume introduces its own errors. A transaction that gets resolved incorrectly under time pressure creates a downstream problem that's harder to diagnose than the original failure. 

A functional acknowledgment (EDI 997) comes back with a rejection code. There's no automated alert, no routing rule, no assigned owner. It lands in a shared inbox. Three days later the purchase order is in limbo, the buyer is asking questions, and the fulfillment window is closing. The resolution, when it finally happens, is manual and undocumented, which means the next time the same rejection code appears, the process starts over from scratch. 

Failure Point 4: Monitoring That Flags Nothing Until It’s Too Late 

Most DIY integrations have some form of monitoring. There's a log file, an alert if the process crashes, and enough visibility to know whether the workflow ran. What's usually missing is visibility into whether it ran correctly. 

The gap between those two types of monitoring is where a significant share of DIY integration damage accumulates. Adequate monitoring for a supplier integration workflow requires more than uptime alerts. It requires field-level validation against trading-partner-specific rules, and alerting that's tied to business outcomes (e.g., a rejected transaction, a compliance score drop, a chargeback) not just whether the process completed. 

Incorrect ASN data gets transmitted for weeks. The workflow runs cleanly every time. Nothing alerts. The problem surfaces when a chargeback cycle closes and the deductions hit, at which point the damage is done and reconstructing what happened requires digging through log files that weren't built to answer that question. In some cases, a chargeback or deduction will be dealt with by a finance team that isn’t considering that the workflow is the root cause of the problem or doesn’t have time to dig in to root cause, and simply writes off the chargeback as a cost of doing business.  

Failure Point 5: Retailer Spec Updates That Hit Without Warning 

Retailers update their EDI requirements on their own schedules, and the communication isn't always timely. For a managed integration provider, tracking these changes is a core function. For a DIY team, it competes with everything else on the plate. Most DIY teams learn about a spec change when a transaction fails, not before. 

The window between when a spec changes and when a DIY team catches it is where compliance score damage accumulates. Transactions keep transmitting in the old format, nothing flags an error, and the problem surfaces in a scorecard review or a chargeback cycle. 

Walmart updates its 856 ASN requirements. The change is published in the supplier portal but never reaches the person who owns the integration map. The workflow keeps transmitting in the old format for two weeks before a scorecard review surfaces it. The fix is straightforward. The compliance damage is already on the record. 

Failure Point 6: Ownership Gaps When Something Breaks 

DIY integration projects are built by someone. That person understands the system, knows where the edge cases are, and can diagnose problems quickly. The risk is what happens when they move on, change roles, or accumulate enough other responsibilities that integration issues stop being their priority. 

Runbooks and escalation paths are usually the last thing documented and the first thing that matters when something breaks. Most DIY builds have neither or have documentation that was accurate at launch and hasn't been touched since. The institutional knowledge lives in one person's head, and that person isn't always available. 

A transaction fails at 11pm before a Monday morning delivery window. Nobody knows who to call. The person with system access is out of office. By the time the right person is reached, the window has passed. 

Failure Point 7: Scalability Assumptions That Don’t Hold 

A workflow built for 10 trading partners and 200 weekly transactions behaves differently at 30 partners and 2,000 transactions. Performance issues, queue backlogs, and database limits that never appeared in testing start surfacing under real volume. Scalability testing is usually skipped because day-one volume feels manageable, and it is. The problem is that the workflow doesn't stay at day-one volume. 

This ceiling gets lower when large batch files are involved. A workflow that handles a single PO cleanly may lag or fail unpredictably when asked to process a 5,000-line ASN or a batch of 500 invoices simultaneously. General-purpose integration tools aren't always built for that kind of throughput, and if the infrastructure isn't configured for external binary storage, memory exhaustion becomes a real concern.  

Peak season rolls around, doubling your transaction volume. Processing that ran in near-real-time is now running six hours behind. ASNs are transmitting after shipments have already arrived, which means the advance ship notice isn't in advanceanymore, and the compliance hit follows. 

Failure Point 8: Undocumented Workflows and Governance Gaps 

Low-code integration tools have made it possible for operations teams, logistics coordinators, and procurement staff to build their own workflows without involving IT. That's the appeal. The governance problem is what happens when those workflows become mission-critical and something breaks. 

IT is now responsible for fixing a workflow they didn't build, weren't told about, and have no documentation for — no version control, no peer review, no record of what decisions were made or why. And because the workflow was never formally scoped or resourced, there's no clear owner to call, no runbook to reference, and no way to know whether a fix in one place breaks something else downstream. 

A logistics coordinator builds an order routing workflow in a low-code tool. It works, gets adopted, and quietly becomes load-bearing for a key trading partner relationship. Eighteen months later it breaks. IT gets the call, but they're starting from zero, since there was no clear governance of the workflow.  

Modern integration tools are built around HTTP-based communication, such as webhooks, 

, and standard web protocols. That works fine until you hit a trading partner who mandates AS2, which is a different protocol entirely. AS2 requires digital certificates, encryption key management, and non-repudiation receipts, which is a mechanism that cryptographically confirms a message was received and authenticated by the intended partner. General-purpose integration tools often don'tsupport this natively, and bolting it on isn't straightforward. 

For suppliers working with Walmart, this isn't an edge case. Walmart mandates AS2 for EDI transmission. If your DIY build isn't built on infrastructure with native AS2 support, you're not meeting the requirement, and there's no workaround that gets you there without rebuilding the communications layer. 

A team builds a workflow using webhook-based transmission logic. It works cleanly with several trading partners. Then a major retailer requires AS2. The tool doesn't support it. The choice is rebuild the communications infrastructure or find a different solution entirely. 

What Separates Workflows That Hold From Workflows That Don’t 

What separates successful workflows from unsuccessful ones is whether the team built for failure, not just for success. The workflows that hold up over time aren't necessarily more technically impressive. Instead, they're built with an honest accounting of what breaks and what needs to be in place when it does. 

There are four general markers of a resilient DIY build: 

These aren't simple to build, and they're not simple to maintain. As Eric Shelton, supply chain strategist at SPS Commerce, stated, “I think the vast majority [of DIY builds] will need help in some capacity. And I think that's where the opportunity is.” 

Teams that evaluate the full operational picture honestly (not just the build cost, but the ongoing staffing and oversight required to keep a DIY integration healthy) sometimes conclude that managed infrastructure makes more economic sense. 

When To Reconsider Your Integration Infrastructure 

If the failure modes in this article look familiar, that's useful information. It means the exposure is visible, and visible problems can be addressed, whether that means shoring up the current build or reconsidering the infrastructure approach entirely. 

 provides managed EDI and integration infrastructure where spec monitoring, transaction-level validation, exception routing, and protocol compliance are core functions. For teams that have mapped their failure points and concluded that managing all of it internally isn't the right fit, that's what a managed solution is designed to handle. 

Retail News: Industry Leaders Raise the Bar on AI, Retail Tech, & Fulfillment

Top Retail News You Need to Know This Week

This week’s retail news roundup looks at how retailers and retail tech providers are investing more heavily in AI automation, faster fulfillment, and operational efficiency.

Dive into this week’s retail news roundup covering retail tech automation, AI-powered advertising, and the race for faster delivery.

Retailers Turn to AI to Fix Costly Operational Bottlenecks

Amazon and Walmart are using AI and automation to tackle costly operational problems like slow AI execution and complex ads campaign management. 

Latest Walmart AI Initiative Focuses on Speed, Not Just Technology

Walmart’s AI work adds another layer to the same story. Instead of simply building AI tools, the retailer is trying to shorten the time between development and real-world use.

, SVP of Enterprise Business Services at Walmart, aims to achieve this by 

move toward faster testing, smaller task-specific AI tools, and more repeatable development infrastructure

 (building systems and processes that make it easier to create and launch AI tools over and over again instead of starting from scratch).

As we all know, retail moves quickly. A fulfillment issue, customer service problem, inventory gap, or pricing opportunity may not wait for a months-long roadmap. If teams can test and improve AI tools in days or hours instead of quarters, they can respond faster to problems that affect customers and operations.

One key idea, as Glick explained, is the use of smaller specialized AI agents, or 

instead of one large system trying to do everything. 

Rather than building one giant tool for every task, Walmart can build smaller AI tools that each handle a specific job well.

 One agent might help with finance workflows, another with operations, and another with employee support.

Between the super agent and the nano agents, 

“chunky agents” act as mid-level organizers. They handle larger chunks of work within broad business areas like finance or operations before passing each to a more specialized tool.

Overseeing the process is a routing layer, or 

 that directs requests across the system.

Walmart Reengineering AI Speed Delivery.png

For example, a user submits a “check customer refund” request, which the “super agent” recognizes as a finance-related issue. It then routes the request to a “finance chunky agent,” which passes the task to a specialized “nano refund agent.” This allows teams to resolve requests faster without manually figuring out which tool should handle each task.

When applied to retail operations, this could help Walmart’s internal teams move faster with fewer bottlenecks while improving issue resolution, stock visibility, and delivery experiences for shoppers. 

However, it could also raise expectations for suppliers and partners working with Walmart. If the retail giant can move faster internally, they may expect partners to respond faster too.

Amazon Ads Automates More of the Campaign Process

The latest Amazon Ads updates show how AI automation is becoming more embedded across its retail media network as advertisers manage more campaigns with leaner teams.

 based on campaign goals and targeting criteria.

 The feature uses machine learning to monitor campaign performance, add stronger-performing deals, and remove weaker ones throughout the campaign

Advertisers can now promote multiple related products in a single ad unit 

either by manually selecting ASINs or letting Amazon’s AI dynamically group products based on campaign targets and shopping signals (e.g., search intent, browsing behavior, and product details). This could help brands improve product discovery without building multiple campaigns manually.

Amazon introduced additional AI-powered tools for:

Overall, these Amazon Ads updates aim to reduce manual campaign setup while improving audience reach and performance optimization across the customer journey.

. Our expert-managed advertising solution leverages Amazon's first-party data to connect your brand with high-intent audiences across websites, apps, and streaming platforms. 

Amazon and Walmart Push 30-Minute Delivery Into the Mainstream

The two retail giants are competing more aggressively to get groceries and household goods to shoppers faster. What used to be a premium convenience, same-day delivery, is being compressed into 30-minute and sub-hour delivery windows across more major cities.

, its 30-minute delivery service, by using smaller micro-fulfillment centers placed closer to where customers live and work.

These Amazon micro-fulfillment centers are much smaller than traditional facilities and are built for speed, not massive selection. They stock high-demand items such as groceries, household essentials, personal care products, pet food, over-the-counter medicine, and other products shoppers often want immediately.

Walmart is answering with its own local fulfillment play.

 Instead of only relying on its supercenters, 

 These depots in Texas, New Jersey, Arkansas, and soon New York and California are designed for rapid grocery and household delivery, often using 

 network to move orders at a high speed through major cities.

The Delivery Speed Race Creates New Pressure On Sellers

The bigger story is that both companies are trying to make ultra-fast delivery feel normal.

Amazon has the advantage of a massive logistics network and years of Prime conditioning. Walmart has a major store footprint, strong grocery demand, and physical locations already close to many households.

 That makes this less of a delivery experiment and more of a fight over who becomes the default option when shoppers need something now.

On the one hand, this could make inventory planning more important for sellers using FBA or Walmart Fulfillment Services (WFS). Even when Walmart or Amazon themselves handle delivery, products still need to be in stock and positioned across the right locations to qualify for faster shipping.

, as shoppers are unlikely to order from stores with slow delivery.

 Sellers with weak inventory coverage may struggle to keep pace with the sub-hour delivery expectations that the two retail giants are trying to set.

If Amazon and Walmart train shoppers to expect groceries, toiletries, medicine, and household basics within 30 minutes, slower delivery options may start feeling less competitive, especially for replenishable or urgent-use products.

The pressure will likely show up in a few ways:

Sellers may not control exactly where FBA places their inventory, but they can control whether they send in enough stock for Amazon to distribute it effectively. Keeping healthy inventory levels gives Amazon more room to position products closer to demand, which can help protect delivery speed, visibility, and appeal in categories where speed matters.

At the same time, sellers also need to be careful about becoming too dependent on Amazon’s fulfillment infrastructure alone. Even with strong inventory planning, issues like Amazon misplacing your pallets or receiving delays can still create fulfillment risks.

That makes it important to maintain backup plans where possible, whether through additional 

, alternative fulfillment options, or closer monitoring of inbound shipments.

2. Margins Depend on Inventory Efficiency

Amazon handles FBA fulfillment, but sellers may still face additional storage or inventory management costs if they need to keep more stock available to support faster delivery expectations. If better regional availability helps products sell faster, the sales lift may offset those costs. If demand falls short, however, margin could tighten due to fees tied to overstock or excess inventory.

Not every item needs a 30-minute delivery. Everyday essentials, consumables, pet supplies, grocery-adjacent products, and urgent replacement items are more likely to benefit than slower-consideration purchases.

 to operate, and many past quick-commerce startups 

 with high costs and weak customer loyalty. While Amazon and Walmart have far more scale, 30-minute delivery will likely work best in dense metro areas and high-frequency product categories.

That’s why sellers should treat this as an inventory planning decision, not a push to chase speed at any cost. 

 Amazon or Walmart may control where inventory goes, but sellers still control how much inventory they send in, when they replenish, and whether the expected sales increase justifies the added holding costs (storage fees, tied-up cash, and risk of slow-moving products).

Amazon Expands FBA Liquidations Options for Sellers

 on excess and unfulfillable inventory. Eligible sellers can recover a portion of product value while immediately stopping storage and aged inventory surcharges, making this worth reviewing before Q4 inventory planning ramps up.

Amazon Launches New Business Credit Cards with Cashback Rewards

, giving businesses up to 5% cashback rewards on Amazon, Amazon Web Services, and common operating expenses. The cards also include installment payment options that could help sellers better manage cash flow and operational spending.

Walmart Becomes a Key Battleground for Buy Now, Pay Later (BNPL) Providers

Although Walmart replaced Affirm Holdings with Klarna Group as its official BNPL partner, 

, as more rely on installment payments to manage higher everyday costs and stretched household budgets. Keep an eye on this trend because it could influence how consumers shop and spend heading into the second half of the year.

Etsy is expanding deeper into conversational commerce with its 

. As more ecommerce platforms experiment with AI-assisted product discovery, sellers may need to think beyond traditional marketplace optimization and pay closer attention to how products surface inside AI-driven shopping experiences.

Adapting to Faster Fulfillment and AI-Driven Retail

This week’s retail news shows how retail leaders are doubling down on faster fulfillment, AI automation, and operational visibility to improve execution. The opportunity is not for sellers to control delivery speed directly, but to plan inventory well enough for Amazon or Walmart to distribute stock closer to demand without creating unnecessary holding costs or operational complexity.

For nearly five years, we’ve kept sellers informed, publishing 250 issues packed with policy changes, announcements, and community events. 

 with your team to get these insights delivered bi-weekly.

From Search Bars to Conversations: How Product Data Quality Shapes AI Discoverability

______________________________________________________________________________ 

Consumers shop differently today than they did even a few years ago. Ecommerce once centered around search bars. A shopper entered a few keywords, scrolled through pages of products, compared listings, and gradually worked toward a purchase decision. 

Now, consumers ask conversational questions through large language models (LLMs), like: 

Even platforms powered by conversational AI are changing how products surface during discovery. Recommendation engines and AI shopping assistants are influencing which products shoppers see first and which ones never even enter consideration. 

For brands, this evolution reaches far beyond the traditional merchandising strategies. Product visibility now depends on whether AI assistants can confidently 

 products. That confidence comes from the quality and consistency of the information moving through networks. 

How AI is Redefining Product Discovery: From Keywords to Conversations 

During the 2024 holiday season, AI-referred traffic to retail sites increased more than 1,300% year over year. Now, consumers interact with AI tools the same way they once used traditional search engines.  

Instead of manually comparing dozens of products, shoppers ask direct questions and review a smaller set of curated recommendations. Traditional ecommerce searches rewarded visibility volume. Conversational discovery compresses time spent decision-making and ultimately speeds up sales.  

Products lacking sufficient supporting information may never enter the consideration set. AI systems are determining visibility before shoppers ever begin comparing options. 

Why AI Recommendations Depend on Structured Data 

Traditional ecommerce search environments often rewarded keyword optimization and advertising support. Conversational AI introduces different requirements. 

Recommendation systems interpret intent, compare attributes, and determine whether enough information exists to confidently surface (and recommend) a product. 

Consider a shopper asking: "What’s a lightweight carry-on backpack for a two-day business trip?" 

An AI system must understand the context. It must evaluate dimensions, weight, materials, intended use cases, compatibility information, and category relationships. 

Articles

SupplierWiki curates detailed articles to help suppliers win at retail.

Prepping for Walmart Audits

Overview

Articles