In a public-page snapshot of Walmart, Target and Macy's across Rakuten, TopCashback and BeFrugal, the displayed offers differed by portal and product category. The useful finding is that a headline rate is not a basket-wide promise. We could not establish an identical eligible purchase across all nine observations, so this study does not declare a highest-paying portal.
Snapshot: October 23, 2026, 12:56–1:02 a.m. Pacific Daylight Time. These are historical labels retrieved from public pages, which may be cached, not guaranteed current account offers or tested payouts.
A same-store comparison sounds simple: open three portals, find the same retailer and choose the largest number. Our small observation exercise shows why the next step—matching the offer to the actual purchase—matters. Several displayed numbers applied to selected categories, while other pages exposed a coupon-associated cash back label without resolving all eligibility.
We recorded nine portal/retailer observations and made the source data available. This is an initial offer-transparency study, not a statistically representative rate survey, an identical-basket experiment or a personal shopping test.
Cashback rate study: the same three stores across three portals
| Retailer | Rakuten | TopCashback | BeFrugal |
|---|---|---|---|
| Walmart | No Cash Back in the retailer heading. | Up to 8% for Flash/Extra Savings Deals; selected ordinary categories 3%; media, gaming and electronics 2%. | Up to 18% on a Fall Rollbacks coupon listing. Full category eligibility unresolved. |
| Target | 1% in the title and visible product labels; whole-basket eligibility not established. | 2% in four listed groups: clothing/shoes/accessories, baby gear/furniture, health/beauty and home/outdoor living. | 0.1% cash back label on a visible clearance coupon listing; separate card discounts excluded from our rate. |
| Macy's | 8% on visible sale listings, shown as increased from 2%. | 15% Lancôme; 10% online purchases, subject to exclusions; 3% marketplace, mattresses, furniture and fine jewelry. | 9% on a Biggest Fall Sale listing. Full item/customer eligibility unresolved. |
Download the nine-observation CSV, including sources, scope and evidence locations. The CSV preserves what we recorded; it is not updated when a provider changes its offers. We did not add referral bonuses, card rewards or coupon-discount percentages to the displayed cash back rate.
Three findings that help a shopper
1. A retailer can have very different visible offers across portals
Walmart illustrates the problem clearly: the retrieved Rakuten heading said no cash back, while other portals displayed conditional rewards. This does not establish that a particular Walmart basket qualifies elsewhere, nor does it predict what a signed-in account will show tomorrow.
The practical response is to check multiple eligible routes before shopping, then choose one. It is not to add all displayed rewards together. Our two-portal compatibility guide explains why competing shopping referrals are not a reliable stack.
2. One retailer page can contain several materially different rates
The Macy's TopCashback table separated a brand-specific rate, ordinary online purchases subject to exclusions and several special categories. Using the largest number for a mattress or marketplace purchase would misread that table.
For any real order, match the specific item and seller to the correct category. If the page does not make that match clear, record it as unresolved rather than multiplying the maximum rate by the whole basket.
3. Similar-looking labels do not prove a controlled comparison
A coupon-associated cash back label and a detailed category table provide different amounts of context. Even the same retailer and timestamp do not control for customer status, personalization, purchase channel, coupon restrictions or the exact product.
That is why we did not average the rates, calculate a winner count or publish a portal ranking from these observations. Those numbers would suggest comparability that the sample does not establish.
How we collected the sample
- Selected the same three U.S. retailers. Walmart, Target and Macy's form a convenience sample, not a representative selection of all shopping.
- Consulted three official portal pages per retailer. The nine cells link directly to those sources.
- Transcribed the displayed rate and context. We used the primary retailer heading, category table or identified coupon listing, not unrelated store recommendations in page navigation.
- Recorded uncertainty. Unresolved category or account eligibility remains explicit.
- Kept historical data. The CSV records the research date, collection window, source URL, short evidence excerpt and location within the retrieved text.
Collection took place during a six-minute research window on October 23, 2026: 07:56–08:02 UTC, equivalent to 12:56–1:02 a.m. PDT. The public pages were returned by a web retrieval service and can be cached. The retrieval time is not a provider's offer-publication timestamp and does not establish simultaneous live rates in authenticated accounts.
No accounts were opened, orders placed or withdrawals requested. The source labels and conditions are factual observations of retrieved material; the shopping interpretation is editorial analysis prepared with AI assistance.
What this study cannot tell you
- Which portal consistently pays the most over a year.
- Whether all nine cells apply to one identical product, basket, customer or checkout route.
- Whether a current signed-in or personalized offer matches this public snapshot.
- How often purchases track, claims succeed or rewards are ultimately paid.
- How much money an average shopper earns.
A stronger follow-up would choose a fixed product and customer scenario, save the full applicable terms from each portal in the same window, and separate rate availability from tracking and payment tests. Until that work exists, the honest conclusion is conditional: compare offers for your purchase rather than treat this historical table as a current shopping recommendation.
How to use the finding before checkout
First compare the price of the actual item, including delivery. Then establish the eligible subtotal and category for the portal offer. A lower purchase price can beat a larger conditional reward, and a permitted coupon can matter more than a headline percentage.
For a hypothetical example, a 10% reward on a $100 eligible subtotal is $10 if approved. A different seller charging $88 with no reward is still cheaper than paying $100 and later receiving $10. These are invented figures, separate from the observed rates above.
Finally, compare redemption requirements. A rate is not a payout schedule. Read our portal comparison, Rakuten review and TopCashback review for minimums and timing. Use the earnings estimator only after choosing appropriate assumptions.
Frequently asked questions
Are these rates available now?
The table is a historical snapshot. Open the linked official source and verify the actual offer for your account and item before shopping.
Does the largest recorded rate identify the best portal?
No. Different category and promotion scopes prevent that conclusion. Unresolved eligibility is part of the finding, not something we replaced with an assumed universal rate.
Why publish a study without a winner?
The useful evidence is how offer labels differ and what a reader must verify. The source data supports that narrow conclusion without pretending it proves payment performance or a permanent ranking.
Sources and methodology
All nine primary sources appear in the table and downloadable CSV. Research date: October 23, 2026. AI assistance was used to organize and interpret the public-page observations. No firsthand purchase or payout results, statistical representativeness or independently measured success rates are claimed. Read our editorial policy and affiliate disclosure.