This is the long version, written for readers who want the terms-and-conditions layer. If you only want to know what to install, the short guide answers that in a few minutes and ends with a routine you can follow.
Everything below is the reasoning underneath it: the exact clauses each service publishes, the payout arithmetic, the gift-card mechanics, and the context on who owns what. Every claim traces to a first-party page — a set of terms, a help-centre article, a store listing — and where we could not verify a figure we say so rather than print one.
The advice you see everywhere is to install a coupon extension, a cashback extension and a price tracker, then let them all work at once. That advice is wrong, and the companies themselves are the ones who say so. Every one of the five big cashback extensions names its rivals as a reason your cash back failed to track. Not one of the five publishes a compatibility statement. Rakuten’s help centre is blunt about it: “To make sure you get your Cash Back, please disable or uninstall any competing extensions.”
Coupert is the most candid of the group, and its answer is the one to remember. Its help centre says that using several cash back or rewards programmes can result in credit being assigned to only one of them, and then adds: “Unfortunately, we can’t determine which service will get the credit.” That is a vendor telling you it does not know who wins the race happening inside your own browser.
The one rule that explains all of this
There is a single test for whether two savings tools can coexist: does the tool write, or depend on, an affiliate referral cookie? If it does, it collides with every other tool that does. If it is settled by card-network transaction data, by a receipt, or by a retailer’s own ledger, it does not.
Rakuten describes the collision in its own Browser Extension Terms, in first-party language, and this is the clearest sentence anyone in the industry has written about the mechanic: the extension sets a cookie through its affiliate network to track your shopping session, “which will overwrite any cookies or other similar tracking mechanisms from other affiliate marketing companies related to the Affiliate Store that were already present on your computer or browser, so that you can be credited a Cash Back reward from Rakuten.”
Overwriting is not a bug in this design; it is how the design works. Whoever writes the last cookie before checkout gets paid. Two extensions cannot both be last.
One workaround people suggest does not work either. Using the cashback website and clicking through manually, instead of installing the extension, does not sidestep the conflict — the portal and the extension write the same cookie through the same affiliate network. It is a good idea for other reasons, which we come back to at the end, but it is not a way to run two services at once.
An asymmetry in the documents
This is our observation rather than anyone’s claim, and it is about the published documents only. The override mechanic is disclosed by these companies — but the framing flips depending on which way the money moves. When a rival’s cookie displaces theirs, it appears in the help centre as a problem you should go and fix: Rakuten asks you to disable or uninstall competing extensions, and Capital One Shopping’s terms warn that “engaging with other affiliate services” will interfere with confirming your purchase. When their own cookie displaces something that was already there, it appears in the terms as the mechanism by which you get paid — Rakuten’s sentence above ends “so that you can be credited a Cash Back reward from Rakuten.”
Same act, opposite register. Whoever set the cookie that gets overwritten — a publisher, a review site, a creator whose link you followed — is not the subject of any of these pages, and there is no reason it would be: these are documents written to explain your rewards to you. It is still worth noticing which side of the exchange gets a warning label. And note the sweeping version of this point does not hold: Awin has run a “soft click” safeguard since 2010 that stops extension cookies overriding influencer ones, so the mechanic is not universal across the networks.
Harmful, or merely redundant?
Be precise here, because the internet is not. In the ordinary case, running two cashback extensions is redundant with an unpredictable loser. Two tools race for the cookie, one wins, you get paid once, and nobody can tell you in advance which account the money lands in. That is annoying and unaccountable. It is not theft, and you are not usually left with nothing.
Here is how you do end up with nothing — and to be clear, this is us joining two of their clauses together, not a claim either company makes. Extension B finds and injects a promo code at checkout. Service A wins the attribution race. Service A then voids the cash back, because the code was not one it listed. Both halves are published: Rakuten states that using coupon codes not listed on Rakuten may void Cash Back, and TopCashback lists an off-network code as a reason a purchase does not track. Put a coupon-finding extension and a cashback extension in the same checkout and you have built a machine for triggering that clause.
The nuance that matters practically: exclusivity is decided per transaction, not per lifetime. Using Rakuten at one retailer and TopCashback at another is legitimate, because attribution is settled purchase by purchase. Rakuten still tells you to uninstall rivals, and it is right to, because the failure is trivially easy to trigger by accident.
One service breaks that rule. Capital One Shopping’s terms condition rewards on installation, not use: rewards may not be available if you have installed on your browser or device another extension, toolbar or plug-in offering rewards, coupons, discounts or cash back, since your purchases may be credited to that other application. If Capital One Shopping is your pick, it has to be the only one of its kind on the browser. The same clause also lists clearing cookies, private browsing, VPNs and ad blockers as interference — with a genuinely odd carve-out for Adblock specifically.
The stack: what actually adds up
With the mechanism in hand you can build a stack honestly. One layer is exclusive. Two layers add to it. One layer is an alternative to it that people mistake for an addition.
Layer 1 — know whether the price is any good. Additive.
The largest saving available is usually not a percentage; it is not buying at a bad price. Keepa (4,000,000 Chrome users, 4.7★ from 5,100 ratings) and The Camelizer (800,000 users, 4.3★) both draw an Amazon price-history chart and will alert you on a drop. You consult them before you buy. They are not actors at checkout, so they are not in the attribution race.
There is a second, better reason they are safe on Amazon: Amazon is essentially absent from the cashback layer.There is no standing Rakuten store page for Amazon at all — the URL redirects to a 404. Rakuten’s Amazon offer is a customer-acquisition promotion rather than a partnership: it is a limited-time $5 offer restricted to members who have not yet made a first purchase through Rakuten, and it states outright that Rakuten is not affiliated with, sponsored by or endorsed by Amazon. TopCashback covers Amazon sub-brands only — Devices, Prime, Subscriptions — with the note that you will only earn cash back in selected departments. So on most Amazon purchases there is no attribution to cannibalise in the first place.
One caveat worth knowing, and it is a distinction between the extension and the website. The Camelizer’s developer states on its own Firefox listing that the extension “contains no affiliate tags / ads / monetization / analytics of any kind,” adding that camelcamelcamel the site is supported by those and that this is what funds the free service. The Chrome listing says nothing on the point, and camelcamelcamel.com does use Amazon Associates — its About page says it receives a commission on purchases made through its Amazon links. The practical upshot: reading the chart in the extension popup is inert; clicking through from the website is not. Look at the chart, then start your session where you meant to start it.
Layer 2 — the retailer’s own price adjustment. Additive.
A price adjustment is settled by the retailer against your receipt, with no cookie anywhere in the process, so it does not compete with anything. Target will match the price if you buy a qualifying item and then find the identical item for less at Target.com, in Target stores, or with a Target Circle deal, and matches may be requested within 14 days of purchase. Note the limit: Target no longer matches competitors, and the lower price has to be validated by a team member on Target.com or in Target’s digital ad.
This is the layer that pairs with the price charts. A price-history alert two days after you bought something is not a wasted notification — it is the trigger to go and file a price adjustment.
Layer 3 — cash back. Pick exactly one.
This is the exclusive layer, for all the reasons above. Choose one, uninstall the others, and read the section below on which one is worth the slot.
Layer 4 — card-linked offers. Additive, with one exception you must know.
Amex Offers and Chase Offers sit in your banking app rather than your browser, and they are settled a different way. Cardlytics, which powers a large part of this market, describes settling on bank transaction data: with insight into that data it reports its exact impact on sales, with no need to estimate performance or rely on complex attribution. It never affirmatively says it uses no cookies, so we will not put that in its mouth — but a reward settled on the transaction is not competing for a referral click.
The third-party exclusion clauses people worry about concern the payment path, not the referral path. Amex’s offer terms exclude purchases made via payment facilitators, resellers, delivery services or other intermediaries — that rules out ordering through a delivery marketplace or paying with a buy-now-pay-later provider, and says nothing at all about browser extensions. And the strongest evidence is negative: “browser extension”, “coupon”, “portal” and “affiliate” do not appear in Amex’s US Offers terms or on Chase’s Offers education page, and TopCashback never mentions card-linked offers either. Neither side of this market treats the other as a threat. (Chase Offers runs on Figg, now Chase Media Solutions, after JPMorgan acquired Figg in 2022 — not on Cardlytics. There is no consolidated public Chase Offers terms page, so treat clause screenshots circulating in forums with suspicion.)
The exception is Bank of America. Its deals pagenow mixes true card-linked deals with link-based ones, and describes a “Shop Now Deal” as a deal that requires shopping directly through the link provided in the offer terms. The same page notes that affiliate offers may take up to 75 days once the transaction posts, and that an ad blocker in your browser may result in the reward not being issued. A “Shop Now” deal is click-settled, which means it is in exactly the same race as your cashback extension. For those deals, pick one link and leave the extension dormant. It is the only substantiated card-linked conflict we found.
Cash-back credit cards themselves are issuer-funded on the transaction and do not conflict with anything here. One caveat carried over from above: card offer terms exclude buy-now-pay-later, so do not swap the card for Klarna at the last step of a purchase you are trying to earn on.
Discounted gift cards: an alternative, not an addition
This is the part most guides get wrong, including an earlier draft of this one. The claim you will read is that paying with a gift card voids your cash back. That is not true as a flat rule. It is decided merchant by merchant, and every provider but one delegates the decision to the retailer.
Rakuten’s binding terms and conditions put it in capitals, and the modal verb is the whole story: purchases with gift cards may not qualify for cash back if excluded in the terms of offer by an affiliate store. Its help centre separates the two mechanics cleanly — some stores exclude cash back on purchases paid for with gift cards, and separately, most gift cards are excluded from earning cash back when you buy them, though some do earn. Capital One Shopping is the same shape: its help centre says retailers may not offer rewards if a gift card, coupon or retailer credit balance is used. Coupert too: it lists gift cards and store credit as a decline reason in cases where the merchant does not support cash back for such purchases.
Karma is the sole unconditional exclusion. Its termssay purchases made using Coins, gift cards, store credit, loyalty points or any other points programme — or any cash back or coupon tool — may not be considered qualifying purchases, with no merchant-discretion caveat attached. Read that clause twice, because the tail of it is remarkable: by Karma’s own terms, running Karma alongside Rakuten disqualifies the purchase.
The two words to search for
Here is the single most useful habit in this article. Open the store’s cash back terms and search them for the phrase “purchase or redemption”. That phrase is the tell that both mechanics are excluded — buying the card andspending it. Where only “purchase” appears, paying with a card you already hold is fine.
It shows up across unaffiliated networks, which is the proof that the exclusion is authored at the merchant level and inherited by whoever lists the store. Rakuten’s H&M and CVS terms both exclude “the purchase or redemption of gift cards”. TopCashback’s Walgreens terms exclude “purchase or redemption of a Walgreens gift card”. But Best Buy on Rakuten excludes only “the purchase of Gift Cards”, with no redemption clause at all — so at Best Buy, paying with a gift card is fine. Rakuten even tells you where to look: exclusions live behind the See all Terms & Exclusions link on the cash back activation page.
When a store does exclude redemption, there are two very different remedies. Most are pro-rata— the wording is “portion of order paid for with…”, and you keep cash back on the cash-paid remainder, as with CVS Extrabucks, Petco Vital Care Rewards and Giftcards.com rewards. A few void the whole order. The strongest such language we found is Kohl’s via TopCashback: using your Kohl’s Cash “will invalidate your Cash Back.” Not reduce it, not prorate it — invalidate it. Jurisdiction matters too: Rakuten Canada is worded more absolutely than Rakuten US, stating it cannot give cash back for the portion of a purchase made with gift cards or certificates.
And the reverse mechanic is not a universal law either. TopCashback actively sells gift cards and pays cash back on them, at rates from 0.5% to 20.4% — 4.65% on a Marriott gift card, for instance — and Rakuten pays 1% on Giftcards.com. “Most gift cards are excluded” is Rakuten’s own wording about its own programme, not a rule for the category.
The bigger risk nobody mentions — and the fix
The redemption question is a minority risk. The larger one is last-click collision. If you buy a discounted card from a resale marketplace in the middle of your shopping session, the last rewards site you clicked was not your cashback extension — and now you have handed the attribution away for reasons that have nothing to do with gift cards at all.
The fix is embarrassingly simple and it is the most actionable advice on this page. Buy the discounted gift card in a completely separate session, ideally days earlier. Then start a clean click-through from your cashback extension and pay with the card already in hand. That removes the collision risk entirely and leaves only the merchant-level redemption question, which you can check in advance with the two words above.
A note on where the discounts actually are, because this corner of the market moved recently. Raise’s resale marketplace was renamed The Gift Card Exchange and now lives at gcx.app, advertising savings of up to 30%; raise.com itself is now a pay-with-gift-card rewards app that its own FAQ says operates separately from that marketplace. CardCash still runs a peer-to-peer resale marketplace and publishes only per-brand figures rather than an overall range — the useful pattern being that discounts run inversely to how universally spendable a card is, so a Walmart card barely moves while a single-restaurant card moves a lot. GiftCardGranny is now a cashback programme itself, describing buying a cash back gift card and earning money on it — which puts it squarely in the last-click collision described above, not in a separate layer. And the deepest verified discount in this entire space is not a resale spread at all: Costco’s restaurant gift cards are described on its own page as typically offering a 20% to 25% discount, which is manufacturer-funded bulk pricing and therefore steadier than a marketplace spread. You do need a membership to shop Costco.com at member pricing, and the resulting card is still subject to the redemption question at the restaurant.
Things that quietly conflict
The newsletter welcome code.This is the most counterintuitive item on the page. That 10%-off code you get for handing over your email address is, by definition, a code your cashback service did not list — and every provider quoted above voids cash back on off-network codes. A 10% code that kills 4% cash back is still a win; a free-shipping code that kills it is probably not. Decide deliberately rather than by reflex, and note that Rakuten describes the retailer’s own sitewide and free-shipping codes as compatible while warning that a random code can change attribution or void cash back, and advising against testing a lot of codes right before checkout.
Microsoft Edge Cashback is a competitor, and Microsoft says so itself. Its rebates terms list as ineligibility reasons: “You received cashback or points from a third-party reward program for the same purchase”; “You used a different link, site, or tool that links to the participating merchant after activating the Microsoft Cashback offer”; “You have adware, malware, shopping tools or browser extensions that may affect tracking”; and “Most offers do not provide Microsoft Cashback rewards on purchases that use coupon codes.” Rewards are capped at up to US$200 per calendar month and up to US$1,000 per calendar year. Microsoft has also discontinued the Coupons feature in Edge, in its words to simplify its offerings, while Cashback, Price Comparison and Price History remain. Edge’s price history and price comparison are fine to leave on. Edge Cashback is a fifth cashback service, and it counts as your one pick.
Chrome’s built-in price tracking is described by Google purely as price history and drop alerts, with no mention of commissions anywhere. Be precise about what that is: an absence of disclosure, not a denial. The real cost is elsewhere — it requires signing in to Chrome and syncing your history.
Slickdeals conflicts, because it runs its own cash back: it states that some stores offer it a commission, part of which it shares with you as Cashback Rewards Points. Honey’s Droplist conflicts by association, since it ships inside an affiliate cashback extension. Ibotta splits down the middle — its in-store half, settled by receipt scan or a linked store loyalty account, adds on top of anything, while its online half is just another cashback extension competing for the same cookie.
Store loyalty points: earn alongside, but never pay with them. Karma’s clause disqualifies purchases made using loyalty points or any other points programme, so spending points is a live risk. The other half of that advice — that merely earning points is fine — rests on the absence of exclusionary language rather than on anyone confirming it, and we would rather say that than pretend it is verified.
Seven rules that decide whether you get paid
- Pick one cashback tool and uninstall the rest— not disable, uninstall. Capital One Shopping’s clause triggers on installation alone, and a disabled extension is still an installed one.
- Only use coupon codes your chosen service shows you. An outside code can void the cash back you just activated, by every provider quoted above.
- Activate before you start shopping, not at checkout. This is the opposite of the folk advice, and Rakuten is explicit: “Always activate Cash Back before you add anything to your cart” — because some stores will not link a purchase if items were in the cart first.
- One tab, one session, one device. Finish in the tab the service opened for you. TopCashback lists buying in a different tab as a reason a purchase does not track; switching devices mid-checkout breaks it too.
- Turn the ad blocker off for checkout. TopCashback needs it off to track; Capital One Shopping lists ad blockers as interference — with a carve-out for Adblock specifically, which is as strange as it sounds.
- There is no universal cookie advice. TopCashback recommends clearing cookies and cache before you shop. Capital One Shopping lists clearing cookies as something that will interfere with confirming your purchase. They contradict each other; follow whichever one you chose.
- Different services at different shops is fine — attribution is per transaction. The exception, again, is Capital One Shopping, whose clause is about what is installed rather than what you used.
Which one to keep
Rakuten is the pick for most people in the US. It pays real cash rather than credit, it has the lowest minimum of the five at $5.01, we found no expiry clause, its usage guidance is the clearest of the group, and it is the only one that says it will notify you when a rival extension has killed your attribution so that you can reactivate. The honest cost is the payout calendar: quarterly, on 15 February, 15 May, 15 August and 15 November, so money can sit for roughly three months.
TopCashback is the runner-up, and the UK pick — it publishes the most mechanically honest decline reasons of anyone here, including the flat statement that the retailer has already paid your cash back to another website when it was not the last site you clicked. Two honest caveats. Its minimum payout, its payout methods and its dormancy terms are all unverified, so we are not going to quote numbers we could not confirm — which makes this a provisional recommendation. And it says plainly that not every TopCashback store is featured in the browser extension, so coverage in the extension is narrower than coverage on the site. On the UK point: rakuten.co.uk simply redirects to rakuten.com, so there is no separate Rakuten UK cash back property to compare against.
Capital One Shopping only if gift-card credit is genuinely fine with you. Its minimum is a clean $5.00, but read what you are earning. Its terms say Shopping Rewards “are not money or cash obligations, do not have any monetary value, and may never be redeemed for cash”. Plenty of roundups file it under cashback; it is gift-card credit. Add the installation-based exclusivity clause and it is the one tool here that demands the browser to itself.
Karma pays in Coins, and the arithmetic is the story. The minimum is 30,000 Coins, and its terms put PayPal withdrawals at 1,000 Coins = US$1.00 — so $30 before you can take anything out, six times Rakuten’s threshold, with Coins expiring 12 months after they are earned. That combination catches exactly the wrong people: low-volume shoppers, who take longest to reach a high bar and are therefore likeliest to meet the expiry clock first.
Coupert is the weakest of the five on terms — three separate expiry clocks, a 12-month dormancy rule under which rewards may be adjusted, deducted or cleared, and no published minimum payout at all. Credit where it is due, though: it is also the most honest of the five about conflict — its terms state that where multiple coupon services, cash back programmes or similar reward platforms are used for the same purchase, only one may receive credit for the transaction — and its guidance to run only a single savings platform at a time is the correct advice, published by a company with an incentive not to say it.
| Service | Minimum | Paid in | Expiry |
|---|---|---|---|
| Rakuten | $5.01 | Cash — quarterly, on 15 Feb / 15 May / 15 Aug / 15 Nov | No expiry clause found |
| TopCashback | Unverified | Unverified | Unverified |
| Capital One Shopping | $5.00 | E-gift cards only | Balance does not expire, but is forfeited if the account is terminated |
| Karma | 30,000 Coins — $30 | PayPal cash or a gift card | Coins expire 12 months after they are earned |
| Coupert | No published minimum | PayPal or e-gift card | Three clocks — 7-day unclaimed, 12-month gold, 24-month cash back — plus a 12-month dormancy rule |
A word on star ratings, since they are the first thing people compare. Rakuten holds 4.9★ on the Chrome Web Store from 43,600 ratings; TopCashback USA holds 3.7★ from 214. Those counts are not remotely like-for-like, and the gap is not a quality verdict — 214 ratings is a sample that a few bad months of support tickets can move on its own. Take both numbers with the counts attached or not at all.
And the answer for privacy-conscious readers is different. Every tool here asks to watch you shop; the declarations on their store listings vary a lot, and the honest recommendation is to install none of them. Use a cashback website and click through manually. It costs you a few seconds per purchase, it keeps nothing resident in your browser, and it earns the same cash back — while the deals stack above still works, because price history and price adjustments do not need an extension either.
Know who you are installing
Short version, because this is context rather than the point of the article. PayPal Honey is the giant of the category at 12,000,000 Chrome users, and PayPal owns it. In December 2024 the MegaLag investigation alleged that Honey overwrote creators’ and publishers’ affiliate cookies and that it showed users worse coupon codes than were available. PayPal’s statement of 15 January 2026 is a partial admission that disputes only scale: “The code causing this behavior has been identified and no longer has an impact. The code was implemented prior to PayPal’s acquisition, and appears to affect less than 0.1% of Honey’s traffic.” The affiliate networks acted regardless — in January 2026 Rakuten Advertising removed Honey from its network, Awin suspended its payments, and impact.com removed it from its Discovery Marketplace.
What has nothappened matters just as much, because the internet has largely decided otherwise. The litigation is at the pleading stage and Honey has not lost: a motion to compel arbitration was denied on 7 November 2025, the complaint was dismissed with leave to amend on 21 November 2025, and a separate consumer privacy case was dismissed with prejudice on 15 June 2026 — the judge describing the conduct as routine commercial behaviour — and is now on appeal. There is also a correction the category deserves: Awin has operated a “soft click” safeguard since 2010 that stops extension cookies overriding influencer ones, so the sweeping version of the accusation — that every extension in this category is quietly taking money from creators — does not survive contact with how the networks actually work.
The one thing worth checking before you install anything is the privacy declaration on its store listing, because they differ more than you would expect. Honey is the only one of these declaring that it handles financial and payment information. Karma is the only one of the five cashback tools declaring location. TopCashback USA declares the least of any of them — user activity only. We are deliberately not publishing a permissions table alongside that: host permissions are a separate question and we did not verify them.
The cashback extensions, ranked
| Tool | Best for | Pricing |
|---|---|---|
1 Rakuten | Best overall (US) | Free |
2 TopCashback | Best for the UK | Free |
3 Capital One Shopping | Best if gift cards are fine | Free |
4 Karma | Coins, not cash | Free |
5 Coupert | Weakest of the five | Free |
6 PayPal Honey | Listed for context, not recommended | Free |
See the full comparison: Best Cashback Extensions →
Or read the short version: how to actually get the best deal online →