A brand doing $40M on Amazon decides to take Walmart seriously. The agency exports the Amazon keyword list, rebuilds roughly the same campaign structure, points it at the same catalogue and sets the same ACoS target.
Six weeks later Walmart is "not working," the budget goes back to Amazon, and the conclusion recorded internally is that the brand's category does not perform there.
Usually none of that is true. The plan was wrong, not the channel.
What transfers
More than people expect.
Category and competitive intelligence. Who you compete with, what claims win, which attributes shoppers filter on. That research is channel-agnostic and you have already paid for it.
Imagery and A+ style content. With reformatting for Walmart's module specs, the visual argument carries over almost entirely. The first three images are doing the same job for the same reason.
Your margin structure and CAC ceiling. The arithmetic that tells you what a customer is worth does not change because the storefront did. Fees differ, so recalculate — but the framework holds.
Broad keyword themes. Not the exact list with its match types and bids, but the themes: what people call the product, which modifiers signal purchase intent, which terms are informational.
What has to be rebuilt
| Amazon behaviour | Walmart Connect behaviour |
|---|---|
| Near-real-time reporting supports daily optimisation | Reporting lags, so weekly is the shortest honest cadence |
| Deep auctions on almost every category term | Thinner auctions on many terms — attention is cheaper, ranking is more volatile |
| Content changes propagate in hours | Item setup is stricter and changes take time to appear |
| Buy box weights price and fulfilment together | Delivery promise and WFS carry more weight than most brands expect |
| Shopper often searching for a specific item | Shopper more often building a basket |
| Four-week test windows produce usable signal | A quarter is the minimum honest test window |
Reporting lag changes the whole operating rhythm
This is the one that trips up good Amazon operators. On Amazon, a bad day is visible the next morning and you adjust. On Walmart, adjusting daily means you are reacting to incomplete data and then reacting again to the correction.
Set a weekly cadence. Resist the reflex. The instinct that makes someone excellent on Amazon is actively counterproductive here for the first quarter.
Item setup is unforgiving
Attributes that Amazon treats as optional are load-bearing on Walmart. Incomplete item setup suppresses organic surfacing regardless of what you spend, and fixing it later means waiting again for propagation.
Do the content and setup pass before any spend. On Amazon you can run ads while you fix listings and the fix arrives quickly. On Walmart that overlap wastes budget.
Fulfilment is a media variable
Delivery promise affects conversion and buy-box weighting materially. A media plan that ignores whether you are on WFS is planning around a conversion rate it cannot control.
That does not mean everyone should be on WFS. It means the decision belongs in the media conversation rather than downstream of it.
Where the opportunity actually is
Thinner auctions on category terms is the headline. For a brand with genuine Walmart distribution that has under-invested on the network, the same creative and the same catalogue can frequently buy attention at a lower cost than Amazon — because fewer sophisticated bidders are competing for it.
That advantage is temporary and it is closing. The brands capturing it now are the ones treating Walmart as a channel with its own plan rather than as an Amazon overflow budget.
A sensible first ninety days
- 01
Read the category on Walmart
Using your own item-level data rather than an Amazon extrapolation. Where does your category actually convert on this network, and against whom?
- 02
Fix item setup and content
Complete attributes, correct categorisation, rebuilt imagery and copy. Then wait for propagation before spending.
- 03
Launch narrow, weekly cadence
A focused keyword set on your strongest SKUs. Weekly reviews, no daily tinkering, targets set with the reporting lag understood.
- 04
Judge at the quarter
Against your TACoS target and organic rank movement, not against a four-week ACoS reading.
The broader point
Every retail media network has its own auction dynamics, reporting behaviour, shopper and operational constraints. Instacart is basket-first and replenishment-driven. Roundel is a planning channel with long lead times. Kroger's value is loyalty-card audiences measured over purchase cycles, not weeks.
Amazon expertise is real and it transfers as judgement, not as a template. The agencies that struggle off-Amazon are usually the ones that never noticed the difference.
Common questions
Is Walmart Connect cheaper than Amazon?
Cost per click is generally lower on many category terms because the auction is thinner. Whether it is cheaper per acquired customer depends entirely on your conversion rate there, which depends on content, buy box and delivery promise.
How much budget does a real Walmart test need?
Enough to gather signal across a full purchase cycle with the reporting lag accounted for. In practice that is usually a quarter, not a month. A four-week test on Walmart tells you almost nothing.
Do we need Walmart Fulfillment Services?
Not strictly, but WFS affects delivery promise and buy-box weighting, which affect conversion, which caps what your ads can achieve. Treat it as part of the media plan rather than a separate logistics decision.
Can the same creative be reused?
Images and A+ style content largely transfer with reformatting. Keyword strategy and campaign structure do not. Copy usually needs a pass because the Walmart shopper is building a basket rather than researching a single item.