"10% instant discount on selected bank cards" is the most effective sentence in Indian e-commerce. It's on the banner, on the product page, next to the price, and in the notification that brought you there.

It's also the most common way people end up choosing the more expensive store — because the discount applies to whatever base price the store set, and stores know perfectly well that you're looking at the percentage rather than the price.

None of this makes bank offers bad. They're frequently the largest single saving available on a big purchase. The problem is applying them in the wrong order, and that's fixable in about two minutes.

The three types, and why they're not equivalent

Instant discount

Taken off at checkout, immediately, in rupees. Typically capped somewhere between a few hundred and a few thousand. This is the real thing — money that doesn't leave your account.

Cashback

Credited later, sometimes weeks later, occasionally as reward points rather than rupees. Worth genuinely less than an instant discount of the same headline size, for three reasons: you've paid the full amount in the meantime, the credit can fail to arrive and require follow-up, and points are worth whatever the bank decides they're worth when you redeem them.

A rough mental adjustment: treat cashback as worth around three-quarters of an equivalent instant discount, and points as worth less still unless you have a specific redemption you value.

No-cost EMI

This is not a discount at all, and the naming is doing a lot of work.

It converts the price into instalments. The interest doesn't vanish — it's either absorbed into the product price or presented as a processing fee. The genuinely useful thing it provides is cash-flow smoothing, which has real value if paying ₹60,000 at once is difficult. What it does not provide is a lower total cost.

Two specific cautions. First, no-cost EMI frequently disqualifies you from the instant discount at the same store, so choosing it can cost you a real saving. Second, if you cancel or return the item, unwinding the EMI is slower and messier than reversing a normal payment.

Where the caps hide

This is the part that changes decisions, and it's usually in small text.

Almost every instant-discount offer has a maximum rupee value. "10% instant discount up to ₹2,000" means exactly that: on anything above ₹20,000, you get ₹2,000 and no more.

Which produces this, on a ₹90,000 laptop:

The percentage was accurate at ₹20,000 and becomes increasingly decorative above it. On expensive purchases — exactly where you'd think the offer matters most — it matters least in percentage terms.

Other conditions worth reading before the offer influences anything:

The calculation that matters

Only one number decides anything: what leaves your account.

For each store, compute: base price − instant discount (capped) − coupon + delivery. Then compare those totals. Nothing else.

A concrete example of how the intuitive answer goes wrong:

Store B is ₹500 cheaper. Store A has the banner, the badge, and the percentage. Most people buy from Store A, and they're not being careless — they're responding to the only comparison that was presented to them.

The fix is mechanical: get the base prices first, then apply offers. Never the other way around.

The right order to apply everything

  1. Fix the exact product and variant. Before any prices. This prevents an offer from quietly changing what you're buying.
  2. Find the lowest base price across all stores for that exact variant. This is the foundation everything else multiplies.
  3. Check which cards each store is partnered with this week, and the caps in rupees.
  4. Apply store coupons and any wallet or gift-card balance.
  5. Compute the final payable at the two or three cheapest stores.
  6. Break ties on seller rating and return policy, not on ₹100.

Step two is the one people skip when a banner has already convinced them where to shop. It's also the step with the largest impact, because base-price gaps between stores are frequently bigger than the capped discount.

Stacking: what actually combines

Usually combinable: bank instant discount + platform coupon + wallet balance or gift card + exchange value.

Usually not combinable: bank instant discount + no-cost EMI; two bank offers on the same transaction; the same offer twice in one sale period.

Worth testing rather than assuming: many platforms show the final total at the payment step before you confirm. Reaching that screen and reading the number costs nothing and answers the question definitively for your specific combination. Just don't confirm until you've done the same at the other store.

The offers that aren't from banks

Bank offers are the loudest layer, but three others sit alongside them and behave differently.

Platform coupons. Applied in the cart, usually stackable with a bank offer, and frequently category-specific. Worth a ten-second check of the coupon section before paying — many platforms show applicable codes right there, which is where nearly all of the genuine coupon value is. Hunting external coupon sites for the long tail rarely pays for the time.

Exchange bonuses. On phones and appliances, an extra amount on top of the base exchange value, often during sale periods. This one is easy to undervalue because it arrives as a reduction rather than a discount, but on a phone purchase it's frequently larger than the bank offer.

Wallet and gift-card bonuses. Load ₹10,000, get ₹300 extra. Real, but it locks money into one platform, which quietly undermines the habit of comparing across stores — you'll be reluctant to buy elsewhere even when elsewhere is cheaper. Worth it only if you're certain of spending the balance there anyway.

Cashback through third-party portals. A percentage returned for routing your purchase through an intermediary. The rates look attractive; the practical issues are tracking failures, long payout delays, and minimum withdrawal thresholds. Treat any such cashback as a bonus if it arrives, never as part of the price comparison.

Should you get a card for the offers?

Sometimes, with two caveats worth being honest about.

The arithmetic works if you shop online frequently, always clear the balance in full, and the card has no annual fee or a fee you'd cover through the discounts. Co-branded cards tied to a platform you genuinely use can be worth several thousand rupees a year for a heavy shopper.

It stops working the moment you carry a balance. Interest rates on Indian credit cards are high enough that one month of revolving debt erases a year of instant discounts — comfortably. If there's any doubt about clearing it monthly, the offers are not worth the card.

The subtler cost is behavioural: a card that gives 10% off at one platform is a card that gently discourages you from checking other platforms. Which is precisely the habit that costs more than the discount saves.

The trap of letting the offer choose the store

This is the summary of everything above, and it's worth stating on its own.

Bank offers rotate monthly. Base-price gaps between stores exist every single day, on every product, and they're often larger than the capped offer. If you let the offer decide where to shop, you've optimised the smaller, temporary variable and ignored the bigger, permanent one.

Compare the base prices. Then apply your card to whichever store won, or to the one where the combination lands lowest. That order costs you nothing and it's the difference between using bank offers and being used by them.

Making step two quick

Getting base prices across every major store is exactly what 7Compare does in one search — the exact model and variant matched across nine Indian retailers, live, sorted lowest first. That gives you the foundation number in seconds, and then the card offers apply to the best starting point rather than a convenient one.

The offers themselves still need your eyes, because they depend on your card, your bank and the week. But the sequence matters more than the arithmetic: price first, offer second. Do it in that order and the banners become useful rather than persuasive.

One last habit worth building, since it takes no effort: before confirming any purchase above a few thousand rupees, look at the final total on the payment screen and ask whether it matches what you expected. Offers fail silently more often than you'd think — a cap applied differently, a card that didn't qualify, a coupon that dropped out when the cart changed. The number on that screen is the only one that's real, and catching a failed offer takes two seconds there versus a support conversation afterwards.