Flipkart seller analytics and optimization tools.

The PLA report family alone has six shapes, and they do not announce themselves. The preamble rows above the header vary by report type, conversion rate appears on two different scales depending on which export you pulled, and a blank Expected ROI column does not mean zero — it means the campaign is bidding on CPC rather than targeting a return, which is a completely different thing to read.

Get any of those wrong and the numbers still render. That is the danger: a misread Flipkart export does not error, it produces a plausible dashboard with the wrong answer in it, and nothing tells you.

These modules were built against real exports from more than one seller account for exactly that reason. The parsing rules come from what the files actually contain, not from what the column names suggest they should.

Five report types, each with its own module: search terms, campaign orders, placements, products and keywords. Between them they answer where the ad money went, which queries triggered it, and which placements and products returned anything.

The search term report is the one to start with — as on Amazon, it holds both the spend that produced nothing and the queries worth targeting deliberately, and it is the report most likely to pay for the ten minutes it takes.

Earn More is Flipkart telling you what it thinks you should fix — pricing that has fallen out of the buy-box range, listings missing attributes that drive discovery, assortment gaps against demand it can see and you cannot. It is genuinely useful and almost universally ignored, because the export arrives as an undifferentiated list with no sense of which rows are worth anything.

This module ranks those opportunities by what acting on them is actually worth, so the export becomes a work queue rather than a wall of equally-weighted suggestions. In practice a small number of listings carry most of the available upside, and the point of the ranking is to find them without reading the whole file.

It is downloaded from Growth → Nxt Insights → Earn More, where you pick a date range before requesting the listings report — a path worth writing down, because it is not where most sellers look for it.

Flipkart's fee structure differs from Amazon's in ways that matter: commission by category, a fixed fee, collection fee, shipping charged on a slab, and GST on top of the fees themselves.

The settlement calculator works out the net figure per order for a real price and category. Feed that into the margin and break-even calculators and you have the number to price and bid against — the same arithmetic as Amazon, different inputs.

Most sellers reading this already run Amazon ads, and the instinct is to carry the mental model across. It mostly transfers, with three exceptions that change what the numbers mean.

First, campaigns come in two kinds. An ROI campaign has a return target; a CPC campaign has a cost-per-click bid and no target at all. The Expected ROI column is blank on the second kind — not zero, blank — and reading that as a failed target inverts the conclusion on every CPC campaign you have.

Second, on ROI campaigns the target lives on the ad group rather than the campaign, so it can look missing at campaign level even when one is set. Third, PLA and PCA are separate ad products with separate report exports, and pulling the wrong one gives you a file that parses cleanly and describes campaigns you were not asking about.

ACOS, ROAS, TACoS and break-even do not change between Flipkart, Amazon and Myntra — only the fee inputs and the report column names do. All four calculators are free, need no account, and accept any currency.

Flipkart's ROI target on a PLA campaign is the same idea as ROAS stated the other way round, which is worth knowing when you set one: a target ROI of 4 is a 25% ACOS.