That is the gap between the bill landing on the table and someone paying it. If the check does not fit in that gap, on a phone, with one hand, while friends are talking, it does not get used no matter how good the ranking underneath is. So every design decision starts from that constraint.

First, the check has to work without an account. A guest can type in a bill and see cards ranked by estimated rupees saved, no sign-up at the counter. A check that asks for signup at the table is a check most people will skip, so the guest path is the real product, not a demo of it. Sign-in exists for saving cards, not for getting the answer.

Second, the answer comes first. The top of the result is one card and one estimated saving in rupees not a list of percentages to compare by hand. Percentages are the trap the product exists to solve: a big percentage with a small cap can save less than a smaller percentage with a bigger cap. So the interface ranks by estimated rupees after the cap, and says so. Early on, one screen sorted by percentage by default while the homepage promised ranking by what you actually save. Those two screens disagreed, and users noticed. The fix was not a new feature it was making every screen tell the same truth.

Third, nothing on the page may ambush the user. Tapping "sort by rupees saved" once sent guests straight to a sign-up page with no warning they asked to re-sort a list and landed somewhere else entirely. It felt like a broken link. Now a gated action says so before you tap it, and only moves you after you choose to go. The same rule applies to sign-in wording everywhere: if one page says "sign in to compare" and the next page compares without sign-in, one of them is lying, and users believe the stricter one.

Fourth, on mobile the answer must never hide behind floating chrome. The site is mobile-first: at a narrow phone width there is no sideways scrolling, the bottom navigation sits where a thumb reaches, and offer terms read cleanly. But a floating install banner once covered the entire top recommendation the one card the page exists to show on every screen. The lesson is blunt: a floating element is fine until it covers the answer, and then it is the worst bug on the page, because the user never sees what they came for.

If you are building a check people use in a hurry, the checklist is short: no account before the answer, the answer first in rupees not percentages, no surprise navigation, and nothing floating over the result. Everything else is polish.

Savings are estimates, not promises, and a listing is not a confirmed redemption. The site is not affiliated with any bank, and this is not advice to get a card or take credit. It is a comparison for discounts you already hold. The comparison is at https://www.kaunsacardpk.com and the offers browser is at https://www.kaunsacardpk.com/explore.