An honest, side-by-side comparison of Walmart Connect, eBay Ads, TikTok Shop and Meta Ads for Amazon sellers wanting to diversify in 2026.
Beyond Amazon
Walmart Connect suits brands already selling well on Walmart Marketplace and wanting cheap, high intent reach. eBay Ads works best for categories with strong secondhand or collector demand. TikTok Shop rewards brands with genuinely engaging short form content, not just a product feed. Meta Ads is the strongest tool for driving new to brand customers back to Amazon. Our own accounts average 9.24× ROAS and 81.5% new to brand on Meta. None of them replace Amazon. All of them reduce how exposed you are to a single platform’s algorithm, fees and policy changes.
What is on this page
Why look beyond Amazon at all
Amazon is still where most of our clients make most of their money, and this article isn’t an argument against that. It’s an argument against having 100% of your revenue and 100% of your customer data owned by one platform you don’t control. Referral fees, ad costs and competition on Amazon have all risen steadily for years. Diversifying isn’t about abandoning Amazon. It’s about building a second and third channel so a policy change, a suspended listing, or a fee increase can’t take out your entire business in one afternoon.
Walmart Connect
Walmart’s retail media network, built the same way Amazon Ads is: Sponsored Products, Sponsored Brands and display, running inside Walmart’s own marketplace. If you already sell on Walmart Marketplace, this is close to a no brainer: the shopper intent is just as high as Amazon’s, and competition is meaningfully thinner because far fewer sellers and agencies are running sophisticated campaigns there yet. The catch is Walmart Marketplace itself. Approval is more selective than Amazon’s, and inventory expectations are strict. This is a channel for brands that are already operationally solid, not a first step for someone still finding their feet.
Structurally, campaign setup will feel familiar if you already run Amazon Ads: keyword and item targeting, automatic and manual campaigns, a search term report equivalent, and a similar auction logic weighing bid against relevance. The biggest practical difference is fulfilment. Walmart Fulfillment Services (WFS) exists but has a smaller footprint than Amazon’s FBA network, so delivery speed promises matter more here as a conversion factor, because Walmart shoppers are used to comparing against Amazon’s two day standard. If your fulfilment can’t compete on speed, expect a lower conversion rate than the identical listing gets on Amazon, whatever your ad spend.
eBay Ads (Promoted Listings)
eBay’s ad product is simpler than Amazon’s, largely Promoted Listings Standard, a cost per sale model rather than cost per click, which genuinely lowers the risk of wasted spend. It performs best in categories where eBay’s audience still over indexes: collectibles, refurbished and used goods, auto parts, and anything with a treasure hunt buying mindset. For a standard new FMCG product, eBay is rarely worth serious budget. For the right category, it’s close to free money because so few competitors are running it well.
There’s also Promoted Listings Advanced, a cost per click option closer to Amazon Sponsored Products, useful once you’ve validated demand through the standard cost per sale product and want more control over placement. eBay’s audience skews toward deal seeking, comparison shopping buyers, which is why price competitiveness matters even more here than it does on Amazon. A listing priced above the visible market average will struggle regardless of ad spend, because eBay shoppers are actively primed to compare before clicking.
TikTok Shop
The fastest growing of the four, and the least mature in terms of attribution. TikTok’s own reporting and third party measurement still don’t fully agree with each other, so treat any ROAS number from TikTok Shop as directionally useful, not gospel. It rewards genuinely entertaining, native feeling content over polished product ads; a listing with strong creator or UGC style video content will outperform a static product shot by a wide margin. Best suited to categories with strong visual or demonstration appeal, beauty, home gadgets, food, fashion, and to brands willing to invest in content, not just budget.
Three formats do most of the work: In Feed Shopping Ads (boosting a video into the main feed with a product card attached), LIVE Shopping Ads (promoting an ongoing live stream), and Product Showcase Ads. Creator partnerships, paying or gifting product to creators who post organically, then boosting the best performing videos with ad spend, consistently outperform brand made content on this platform. Budget for creator relationships as a real line item, not an afterthought, if you’re serious about testing TikTok Shop properly.
Meta Ads, the strongest bridge back to Amazon
This is where we’ve built the deepest data set, because Meta Ads is where “beyond Amazon” spend most directly supports Amazon sales. Run correctly, driving traffic to your Amazon listing via the Attribution API, or to a Meta storefront that mirrors it, Meta becomes a new to brand acquisition engine rather than a separate sales channel competing for the same existing customers.
New to brand (NTB) is the number that matters most here. It tells you what percentage of purchases came from customers who hadn’t bought your brand on Amazon in the past 12 months. An 81.5% NTB share means Meta is doing exactly the job it should: bringing in customers Amazon’s own ads, by definition, can’t reach, because Amazon Sponsored Ads only shows to people already searching or browsing on Amazon. We cover why NTB is the metric worth reporting, and the metrics that quietly aren’t, in the only Amazon DSP metric worth reporting, and Amazon’s own equivalent programme for off platform traffic in the Brand Referral Bonus, and why most tests fail.
Want to know which platform fits your brand first?
We’ll score your category against our own data, free, before you spend a penny testing it yourself.
How the four compare, side by side
| Platform | Best for | Min. viable monthly budget | Attribution maturity |
|---|---|---|---|
| Walmart Connect | Brands already live on Walmart Marketplace | ~$1,000+ | Strong reporting, similar to Amazon’s |
| eBay Ads | Collectibles, used goods, auto parts | ~$300+ | Solid, simpler cost per sale model |
| TikTok Shop | Visual/demo categories, content ready brands | ~$1,000+ incl. content | Still maturing |
| Meta Ads | New to brand acquisition for existing Amazon sellers | ~$1,500+ | Strong, via Attribution API |
Diversifying badly is worse than not diversifying at all. A thin test on four platforms at once tells you nothing. One platform, run properly for a full quarter, tells you everything.
Setting up tracking so you can actually compare these platforms
The single biggest mistake in multi platform testing is comparing numbers that aren’t measuring the same thing. Amazon Ads reports ad attributed sales on Amazon. Meta’s Attribution API reports Meta driven purchases on Amazon, using a different attribution window and model than Amazon’s own reporting, so a “9.24× ROAS” on Meta and a “6× ROAS” on Amazon Sponsored Products are not directly comparable numbers, even though they look like the same metric. Before running any of these platforms, agree on one thing: what counts as a conversion, over what window, and whether you’re measuring last click or a more considered model. Where budget allows, run a genuine incrementality test, holding out a matched audience or region from ads entirely for a few weeks, because it’s the only way to know how much of that reported revenue would have happened anyway.
A decision framework, which to test first
- Already selling on Walmart Marketplace? Start with Walmart Connect. You’re leaving cheap, high intent clicks on the table by not running it.
- Category has a strong secondhand, collector or parts angle? Test eBay Promoted Listings before anything else on this list. The cost per sale model makes the downside small.
- Strong visual product and appetite for content production? TikTok Shop, but budget for content, not just media spend. A product feed with no creative strategy underperforms badly here.
- Established on Amazon, want new customers rather than more of the same ones? Meta Ads, aimed specifically at new to brand, is the highest leverage option on this list for most sellers.
The honest caveats
TikTok Shop’s attribution genuinely isn’t as reliable yet as Amazon’s or Meta’s. Budget for some ambiguity in the numbers, and don’t make big decisions off a single week’s data. Walmart Marketplace approval can take weeks and isn’t guaranteed, so don’t plan a launch date around it until you’re actually in. eBay Ads will underperform badly outside the categories it suits. Don’t force it onto a standard FMCG product expecting returns like Amazon’s. And Meta Ads requires your Amazon listing itself to already convert well; sending new to brand traffic to a weak listing just makes the weak listing more expensive to discover.
There’s also a UK specific wrinkle worth being direct about: Walmart Connect and eBay’s strongest opportunity is US focused, since that’s where both marketplaces have the deepest buyer base relative to competition. UK and EU sellers can still run both, but should expect thinner volume than a US brand testing the same platforms. TikTok Shop is genuinely strong in the UK specifically, which makes it one of the more interesting tests for a UK based seller compared with Walmart or eBay. None of this rules any platform out. It just changes how much volume you should realistically expect before deciding whether a test worked.
Not sure which platform actually fits your brand?
We run Walmart Connect, eBay Ads, TikTok Shop and Meta Ads alongside Amazon, happy to give you a straight read on which is worth testing first.
Frequently asked questions
If you’re already selling on Walmart Marketplace, yes, Walmart Connect offers intent similar to Amazon’s at generally lower competition. If you’re not yet approved to sell on Walmart, that’s the actual first step, not the ad platform itself.
Only if your product has strong visual or demonstration appeal and you can invest in content, not just media budget. A generic product feed with no creative strategy tends to underperform on TikTok Shop specifically, more so than on Amazon or Meta.
Meta’s Attribution API tracks purchases that happen on Amazon after a click from a Meta ad, and reports what share of those purchases came from customers who hadn’t bought your brand on Amazon in the prior 12 months. Our accounts average 81.5% new to brand, meaning Meta is reaching people Amazon’s own ads structurally can’t.
There’s no universal number, but as a starting point, most of our brands test one new platform at roughly 10 to 15% of their Amazon ad budget for a full quarter before deciding whether to scale it. Splitting a small budget across three or four new platforms at once produces data too thin to act on.
It varies a lot by agency. Many are Amazon only. We manage Amazon PPC and DSP alongside Walmart Connect, eBay Ads, TikTok Shop and Meta Ads specifically because most of our clients’ growth now comes from tying Amazon to at least one of these, not from Amazon alone.
In our own data, Meta Ads produces the strongest ROAS (9.24× average) specifically because it’s aimed at new to brand customers with genuine buying intent, not because it’s inherently “better” than the others. eBay and Walmart perform very well too, just for different, narrower use cases.
Not directly. Different platforms use different attribution windows and models, so a Meta ROAS figure and an Amazon Sponsored Products ROAS figure are measuring related but not identical things. Agree on one measurement approach before comparing platforms, and run an incrementality test if the budget allows. It’s the only way to see how much revenue would have happened without the ads.
See our real numbers before you commit budget anywhere
$112M managed, 200+ brands, 16 marketplaces, published pricing, no long lock in.
Send the right reports. Get a straight answer back.
We read the account properly and write back with where the money is going, what it is buying, and the three things we would change first. You keep it either way.
No obligation · no card · a person replies, usually the same working day