Amazon vs Flipkart for sellers: the arithmetic, not the feature list

Every version of this question gets answered with a feature list. Amazon has better tools. Flipkart has less competition. Amazon has more buyers in metros. Flipkart owns tier two. All of it is true and none of it tells you where your particular product makes money.

We run both for brands. The answer is decided by three things, and none of them is on the feature list.

The fee stacks, side by side

Both platforms take a percentage, a flat fee per order, a shipping charge and GST on all of it. What differs is where the thresholds sit.

Two things worth pulling out of that table.

Both platforms now advertise zero percentage fees on large parts of the catalogue. Amazon says zero referral fee on over 12.5 crore products. Flipkart runs 0% commission on fashion and on eligible items below 1,000 rupees. The headline rate is no longer where the two separate.

Amazon's flat fees start lower and its shipping fee starts higher. A closing fee from 2 rupees against a weight handling fee from 37 rupees per item tells you where the cost actually sits on a low ticket order. Flipkart's free local and zonal shipping under 500 grams is the mirror image of that, and it is the single biggest structural difference for a light, cheap product.

Check both current rate cards before you plan around any of this. Amazon revised closing fees effective 7 September 2026, and Flipkart revises by sub-category regularly. A number from a blog, including this one, is a starting point rather than your number.

The first thing that decides it: your price point

Flat fees do not scale with price. A fixed fee of 20 rupees is 10% of a 200 rupee order and 0.4% of a 5,000 rupee order.

That single fact reorganises the whole comparison. On low ticket products the flat fees and the shipping dominate, and small differences in the flat fee schedule decide profitability. On high ticket products the percentage dominates and the flat fee is noise.

So the honest version of "which is better" starts with a question back: what does your product sell for. Below a few hundred rupees, compare the flat fee and shipping schedules line by line and largely ignore the percentage. Above a couple of thousand, do the opposite.

The second thing: your category's return rate

This is the number that decides Indian marketplace profitability and it appears on neither fee page.

Returns and RTO in Indian ecommerce commonly run at 20 to 30 percent of orders depending on category, and apparel sits at the bad end. Every returned order has already cost you a forward shipping fee, a reverse leg, and frequently a product that comes back unsellable.

The arithmetic that matters is not your margin per order. It is your margin divided by one minus your return rate, because the orders that stick have to carry the ones that do not.

A category with an 8 percent return rate and a category with a 28 percent return rate are different businesses on the same platform, and the platform comparison barely moves the needle next to that gap. We wrote the full version of this calculation in why your break-even ACoS is about half what your calculator says.

The third thing: which platform your category is actually strong in

Category strength is real and it is specific rather than general. Flipkart historically indexes strongly in electronics and in tier two and tier three demand. Amazon indexes strongly in books, metros, and a wide long tail. Fashion is fought over by both and by Myntra and Meesho underneath them.

The useful test is not which platform is bigger. It is to search your own product category on both, as a buyer, and count how many credible sellers already hold the first screen. Competition density in your category is visible in ten minutes and it predicts your advertising cost better than any published statistic.

So which one?

If you can only run one, run the one where your category is strong and your price point suits the fee structure. That is usually decidable from the two sections above in an afternoon.

If you can run both, run both, and here is the reason that is not a cop out: the second platform costs far less to add than the first. The catalogue work is done, the images are shot, the pricing is worked out, and the operational habits exist. The marginal cost of the second marketplace is mostly listing time.

What we see go wrong is not choosing the wrong platform. It is running both at half attention, with one campaign structure copied across, and concluding that the weaker one does not work when it was never actually set up.

If you are weighing more than these two, the same arithmetic applied across every major Indian marketplace is in which marketplace is most profitable for sellers in India.

Which is better for sellers, Amazon or Flipkart?

Neither, in general. Amazon suits higher ticket items and a broad long tail, Flipkart suits categories where its flat fees and free local shipping under 500 grams favour a lower price point. Decide it on your category, your price point and your return rate rather than on platform size.

Who is more profitable, Flipkart or Amazon?

For the same product, profitability usually lands within a few percentage points of each other, because both fee stacks have converged. The differences that matter are the flat fee schedule at low price points and the shipping thresholds. Your return rate moves profit far more than the choice of platform does.

Should I sell on both Amazon and Flipkart?

Usually yes, once the first one works. The catalogue, imagery and pricing are already done, so the second platform mostly costs listing time. Do not launch both at once from a standing start, because you will not have the attention for either.

Is Flipkart easier than Amazon for a new seller?

Competition in many Flipkart categories is thinner, which makes early visibility easier to buy and to earn. Amazon has better reporting and a more mature advertising console. Neither difference matters much next to whether your listing and your unit economics are right.

We run Amazon and Flipkart for Indian brands, catalogue and campaigns together. If you want the arithmetic run on your own numbers rather than on an example, talk to us.