Doug Jensen, Global Head of Sales and Sourcing at Pattern: Cross-Border E-Commerce Success Hinges on Being "Slightly Better" Across Every Touchpoint

On September 17, Doug Jensen, Global Head of Sales and Sourcing at Pattern, delivered a speech titled "Case Breakdown of Global Exclusive Buyout Distribution Models: Practical Execution for 3P and Vendor Central Channels" during the first themed panel of the *Accelerate26 Cross-Border Ecosystem Summit · China Edition*, an event hosted by Pattern with Ebrun as its strategic partner. Doug Jensen noted that as traffic entry points and channel playbooks evolve, brands must re-examine core operational components across cross-border e-commerce, including advertising, inventory, fulfillment, returns, promotions, and content. Growth does not stem solely from scaling investment, he argued, but also from identifying previously unseen inefficiencies and leveraging data and technology to execute every operational action with greater precision. Drawing a parallel to Olympic competition, he pointed out that winning a gold medal rarely requires a dominant, landslide victory over rivals; more often than not, outcomes are decided by narrow fractional margins. For cross-border brands, holding even a slight efficiency edge over competitors in keyword bidding, content conversion, inventory turnover, and delivery speed can compound over time, ultimately translating into higher organic rankings, greater traffic, and faster growth. Doug Jensen also shared analytics from a dataset covering more than $1 billion in Amazon ad spend, which found that roughly 27% of ad spend goes to clicks that generate no conversions. In one extreme case, the share of wasted ad spend reached approximately 78%. For many brands, this means that before increasing advertising budgets, the first priority is to identify exactly where capital is being misallocated.

The following is an edited transcript of Doug Jensen’s remarks:

(This article is compiled from the speaker's on-site presentation, with edits and condensations made without altering the core intent.)

01 First Identify Overlooked Waste: Ad Spend, Platform Fees, and Return Costs

Traffic entry points are shifting, and channel operating strategies are changing right alongside them. As we work to improve business performance, we first need to ground decisions in tangible information and data to pinpoint opportunities for efficiency gains. Pattern’s model is straightforward: when brands partner with us, we purchase their inventory outright and sell as much of it as possible across marketplaces. Beyond that, we aim to help brands identify untapped efficiency opportunities across their cross-border e-commerce operations.

Let’s start with advertising. We analyzed data across multiple Amazon seller storefronts, covering more than $1 billion in total ad spend, and found that roughly 27% of that spend is wasted. By “wasted,” we mean spend directed at clicks that did not drive any conversion—brands paid for ad placements but received no corresponding business outcome.

To address this, brands need to break down ad spend in granular detail, assessing budget allocation, return on ad spend (ROAS), and cost per click (CPC) across Sponsored Display, Sponsored Brands, and Sponsored Products campaigns, while also analyzing performance differences between branded and non-branded search terms.

Non-branded terms are particularly critical. If a consumer already knows a brand’s name and actively searches for it, they already have a high intent to click and purchase. Non-branded terms, by contrast, help brands reach new consumers, making them the area where investing extra time, effort, and budget into optimization delivers the greatest return.

Through funnel analysis, we can drill down into even more granular segments of ad performance to identify waste across different ad formats. In one extreme example, roughly 78% of total ad spend qualified as ineffective investment. For brands, the priority is not simply increasing budgets, but first identifying where waste occurs, which investments need to be eliminated, and how to extract greater value from every ad dollar spent.

Beyond advertising, platform operations hold many other easily overlooked costs. Amazon has FBA, Walmart has WFS, and TikTok is building out its own logistics service network. While these platform logistics networks offer convenience for brands, they also carry hidden costs including inventory fees, storage fees, chargeback fees, and Inbound Placement Fees. Some brands continue paying these fees without full visibility into exactly where their money is going.

Brands need to audit every step of the fulfillment process, including which warehouses inventory is stored in, whether transfers are required between fulfillment centers, whether inventory levels are appropriately calibrated, and whether there is room to reduce transportation and storage costs.

All of these factors ultimately impact cash flow. When large volumes of inventory are concentrated in a single location, inventory turnover can slow, locking up capital in storage and logistics. Improving operational efficiency therefore does not just mean cutting costs—it means strengthening cash flow.

Returns represent another common pain point. Consumers may purchase multiple sizes of the same item, test them at home, and then return some or all of their order. Every step of the purchase, shipping, and return process incurs additional costs.

While returns are unavoidable, brands can reduce return rates through clearer consumer communication, more accurate user persona targeting, and structured analysis of return reasons. For already returned items, teams can minimize value loss by determining disposition based on product condition: options include repackaging, refurbishment, resale, clearance, donation, or recycling.

02 Turning Fulfillment from a Cost Center into a Scalable Operational Capability

When Pattern purchases inventory outright and it arrives at our warehouses, we leverage technology systems to manage inbounding, quality checks, stocking, sorting, and outbound shipping.

When cargo arrives at the warehouse, we generate inbound shipment labels for every carton to track the entire inbound process. After warehouse staff scan the labels, the system guides teams through product quality inspections, information entry, labeling, and prep work including multi-packs and bundled assortments.

Once packages enter the conveyor system, an API initiates the shipment creation process and assigns a destination fulfillment center. In the sorting zone, the system automatically prints platform carton labels and shipping manifests. Cartons then pass through a visual recognition system for multi-angle scanning, weighing, and dimensioning, before being automatically routed to the appropriate shipping dock.

Because shipments from multiple seller accounts can be consolidated into a single load, we increase vehicle load density, make fuller use of transport capacity, and lower overall shipping costs.

The same logic applies to direct-to-consumer (DTC) fulfillment. After brands send inventory to our warehouses, Pattern completes inbound inspections, inventory putaway, and bin location assignment, then syncs real-time inventory status across all sales channels. When an order is placed, algorithms select the appropriate packaging and fulfillment warehouse, comparing options across more than 40 carrier and service level combinations.

Once the shipping origin and delivery method are confirmed, the system prioritizes picking based on order outbound cut-off times, and maps the most efficient picking routes for warehouse staff. The entire process relies on inventory scanning to ensure order accuracy and full traceability.

Inventory, orders, and fulfillment require real-time coordination. As soon as cargo enters our warehouses, we complete inbound processing as quickly as possible to make units available for sale and delivery, while adjusting inventory allocation across different markets to avoid overstock or stockouts.

Reverse logistics also requires systematic handling. When returns arrive at the warehouse, we inspect and grade every item in line with platform rules and brand-provided guidelines. Assessment criteria vary by product category, and final decisions on whether to refurbish, repackage, resell, or apply other disposition paths are always aligned with brand requirements.

Crucially, returns processing should not be focused solely on clearing returned inventory. Through returns analytics, industry benchmarking, and product-level insights, brands can also leverage return data to optimize product detail pages, identifying flaws in product descriptions, sizing guidance, packaging, or product quality to reduce returns at the source.

The value of fulfillment capabilities extends far beyond simply shipping out orders. Whether brands sell via Amazon, Walmart, TikTok Shop, Shopee, eBay, or DTC channels, they need to build tight alignment across inventory, platform operations, and last-mile delivery. Only when products reach consumers faster and more accurately can brands deliver on the promises they make to customers.

03 Growth Comes from Marginal Gains: Using Data to Reimagine Promotions, Advertising, and Content

Next, let’s talk about growth.

Many teams still rely on highly intuition-based approaches to building promotion strategies: they run sales when holidays arrive, then use past experience to decide whether to offer a 20% or 10% discount, or whether a campaign should run for two or three days.

While this approach may have worked in the past, the age of AI allows us to rapidly analyze massive volumes of data to make far more granular decisions. Promotions do not have to rely on round-number discounts, nor do they need to run for a full 48 hours. Based on data, a brand might find that a 17.5% discount run over 30 hours delivers better results. These differentiated choices can become the source of differentiated profitability.

Consumers also do not always select the lowest-priced option. When consumers trust a brand and recognize its product quality, they are willing to pay a premium. Promotion strategy therefore cannot revolve solely around “how much to cut prices,” but must also account for consumers’ real sensitivity to price, brand reputation, and delivery timelines.

The same holds true for ad bidding. Programmatic bidding requires ongoing optimization at the individual SKU level, and when SKU catalogs are very large, it is impossible for teams to make such granular judgments manually. Only by combining data, mathematical models, and machine learning can brands accurately determine which keywords deserve investment.

Many brands compete fiercely for a small set of high-traffic keywords, which only drives bid costs higher over time. That is not a sustainable path to profitability. Our goal is to use ad investment to help more keywords earn organic rankings—not by pouring in capital blindly, but by using targeted, efficient spend to earn top positions across a wider set of keywords and search phrases.

As products gain visibility for more keywords, overall sales can rise, which in turn signals to platform algorithms to allocate even more traffic. This is a process of compounding advantage. It is just like Olympic competition. A table tennis match is not necessarily won 11-0; the final score might be 11-9. Gold medals are often won by narrow, marginal advantages.

The same is true for cross-border e-commerce. If a brand performs slightly better than competitors across a set of keywords, it can redeploy the budget it saves to target even more keywords; if a product’s conversion rate is slightly higher than competitors, search algorithms may reward it with more traffic. When these small advantages accumulate, they drive sustainable growth.

Content creation also cannot rely on gut feel. Brands need to research competing SKUs, ASINs, product detail pages, and on-shelf content, analyzing which image types competitors use and which content elements correlate with strong conversion performance, to inform their own content design.

This does not mean copying competitors directly. Every brand has its own positioning, and consumer behavior and psychological profiles differ across audiences. The role of data is to help teams understand which content is most likely to drive clicks and conversions, so that creative investment is rooted in analytics rather than one person’s subjective opinion that an idea “feels right.”

Across promotions, advertising, and content, the core required shift is identical: moving from intuition-based decision-making to rapid, granular, data-driven decision-making.

When I asked the audience today whether saving money or making money is more exciting, most people chose making money. But in my view, the two are not separate priorities. Cutting waste in advertising, inventory, fulfillment, and returns frees up resources for growth; using data to optimize promotions, bidding, and content converts those resources into new sales. Whether we are cutting waste or driving revenue, our end goal is to help brands enter more markets faster and achieve greater success.

This article was first published on the official website of Ebrun.