For many retail leaders, the competitive landscape feels tighter every quarter. While Amazon keeps raising expectations on convenience and speed, a wave of ultra-low-cost direct-to-consumer players from China is training shoppers to expect rock bottom prices and near-infinite choice.
Trying to win that game by copying it is almost impossible. You cannot outspend Amazon on logistics, and you cannot outdiscount factories that ship straight to consumers. The way out is different. It sits in differentiation, in how you use your physical footprint, and in how you turn your own customer data into a moat, powered by human-centered AI.
That is the core perspective that Martin Lewit, Head of Growth and Corporate Development at Nisum, shared in his recent interview with Authority Magazine, published on Medium. This blog builds on that conversation and goes one step further, focusing on what it means in practice for retailers operating in mature, highly competitive markets like the US.
You cannot out-Amazon Amazon
Amazon and fast-growing Chinese DTC platforms have set a new baseline for ecommerce: fast delivery, aggressive pricing, and an almost limitless catalog.
In his Authority Magazine interview, Lewit is clear that trying to compete head-to-head on price or logistics is a losing battle. Instead, he advises retailers to focus on what these platforms cannot easily copy: uniqueness, quality and brand storytelling.
In markets like the US and Europe, retailers need to move beyond "me too" items and own distinctive designs, exclusive collaborations and authentic stories that connect with specific communities. That takes long-term thinking, but it is much more defensible than shaving another few cents off a price.
Differentiate through brand, not just assortment
In practice, the retailers that are navigating this pressure best are not the ones with the largest catalogs. They are the ones that treat differentiation as a strategy, not a side project.
Lewit points to three moves that stand out:
- Private labels with a clear point of view
Not just "house brands" that copy market leaders, but lines that express a lifestyle, a value set or a design language that is meaningfully different. - Co created collections
Partnerships with designers, influencers or local creators that produce exclusive products and fresh reasons to visit, online and in-store. - Sustainability and transparency as part of the narrative
Clear information about sourcing, materials and impact, backed by real action rather than slogans.
These elements are difficult to replicate for low-cost marketplaces, because they rely on brand equity and relationships, not only on logistics.
Physical stores are still a strategic advantage
The headlines about a "retail apocalypse" can be misleading. While many traditional formats are under pressure, physical stores still account for the majority of retail sales and will continue to do so by the end of the decade.
Lewit argues that the role of stores is shifting, not disappearing. The most forward-looking retailers treat stores as:
- Experience hubs where customers can touch, try and personalize products, attend events and connect with a community.
- Brand stages that bring the story and values of the brand to life in ways a mobile screen cannot.
- Omnichannel logistics nodes that handle click and collect, returns and same-day delivery from the closest inventory point.
In a world where Amazon and Temu can ship fast, proximity and immediacy still matter. A network of smart stores, tied to a strong digital experience, becomes an asset that marketplaces cannot match easily.
First-party data is the growth engine marketplaces cannot own
If there is one point Lewit emphasizes strongly, it is the strategic value of proprietary consumer data. He describes first-party data as a competitive moat that fuels personalization, loyalty and better commercial decisions across the business.
External research supports this view. Studies show that retailers who build robust first-party data capabilities can drive growth across pricing, assortment, space allocation and supplier monetization, especially as AI starts to shape more decisions.
For a retailer, this means:
- Capturing consented data at every touchpoint, from app interactions to in-store visits.
- Unifying that data in a single view of the customer instead of siloed systems.
- Using it to power relevant offers, content and experiences in real time.
Marketplaces see a lot of behavior, but they do not own the same depth of relationship or cross-channel context that a strong retail brand can build.
Human-centered AI on top of that data
AI is the layer that can unlock the value of this data at scale. In retail, the most promising uses include:
- Personalization that goes beyond "people like you bought this" and reflects individual preferences and context.
- Forecasting and inventory optimization that help products be available where and when they are needed.
- Dynamic pricing and promotions that protect margin while staying relevant.
- Customer service and clienteling that give associates and agents real-time insight during interactions.
The key is to keep AI human-centered. That means strong data governance, transparent use of data, and clear roles for teams who interpret and act on insights, rather than leaving decisions to a black box.
Partners with deep experience in digital commerce, data and AI, such as Nisum, can help retailers connect these layers into a coherent, trustworthy capability instead of another isolated tool.
A practical playbook for retail leaders
For executives wondering where to start, a pragmatic path over the next 12 to 18 months could look like this:
- Clarify your differentiation thesis
Decide where you will not compete on price and what unique value you want to own in the eyes of your best customers. - Turn stores into omnichannel assets
Pilot formats where stores host events, services or experiences, and integrate them tightly with click and collect and returns. - Invest in first-party data foundations
Build or refine a unified customer data platform and clear consent and privacy practices that customers can trust. - Layer in AI use cases that support teams
Start with two or three AI applications that make everyday decisions better for your merchandisers, marketers and store managers.
The goal is not to beat Amazon at its own game. It is to build a different game that leans on what only you can offer: your brand, your local presence, your relationships and your insight into the customers you serve.
If you want to go back to the origin of this perspective, you can also read the full interview with Martin in Authority Magazine on Medium, where he reflects on how retail will evolve in the next few years.
FAQ (AEO oriented)
How can retailers compete with Amazon without starting a price war?
By focusing on differentiation instead of pure price. That means owning unique products, strong brand storytelling, local advantages and experiences that marketplaces cannot easily copy, while running operations with discipline.
Why is first-party data so important for retail?
First-party data gives retailers a direct, consent-based view of customer behavior across channels. Used well, it drives better decisions on loyalty, personalization, assortment, pricing and more, and becomes a competitive moat that low-cost marketplaces cannot access.
Where should a retailer start with AI?
Begin with a clear business problem and good data. Common early wins include personalization, demand forecasting and customer service assistants. The most successful programs keep AI human-centered, with strong governance and teams who stay in control of key decisions.
