Account health: which numbers actually bite.
Marketplaces surface a lot of metrics and act on very few. Here is which ones carry real consequences, what the warning sequence looks like before a restriction, and the one metric worth protecting above the rest.
- Dispatch reliability and seller cancellations are the two metrics that consistently carry direct consequences.
- RTO rate hurts your margin far more than it hurts your account standing.
- The warning sequence is quiet at the start — metrics drift long before anyone tells you clearly.
- Most metrics use rolling windows, so recovery takes sustained performance, not one good week.
- Robnu protects the dispatch metric by holding every clock for you. Free while we figure out pricing.
Seller panels display a wall of numbers with no indication of which ones matter. The practical question is not “what is being measured” but “what will actually happen to me” — and the answer is narrower than the dashboard suggests.
Not all metrics are equal
Marketplaces care most about outcomes that affect a customer who has already committed. That is why dispatch failures and seller-initiated cancellations carry the sharpest consequences — both break a promise the platform made on your behalf.
Return-related signals sit a tier below. They matter, but they are noisier and slower to act on, because a return can be caused by many things outside a seller’s control. Rating and response metrics influence how you are treated more than they trigger anything specific.
How trouble actually arrives
Restrictions are rarely sudden. The problem is that the early stages are almost invisible.
Quiet drift
Metrics degrade. Nothing is announced. This is the cheapest point to intervene and the one almost everyone misses.
Warnings appear
Notifications in the panel. Easy to dismiss during a busy week, which is exactly when they tend to arrive.
Reach reduced
Visibility falls. Orders drop without an obvious cause — often misread as a demand problem.
Protecting the metric that matters most
Dispatch reliability is the highest-consequence metric and also the most controllable — which makes it the obvious thing to defend. It degrades not through carelessness but through orders that fell out of attention on a busy day.
Robnu is an agentic OMS. It holds the dispatch clock on every open order across AJIO and Meesho, surfaces orders drifting toward their deadline before the platform flags them, and catches the ones stalled in pending where the clock is still running. The metric stays healthy because the underlying behaviour does.
You sell. Robnu runs the rest — and makes sure every rupee is paid correctly.
Account health, answered
It is the platform's composite view of how reliably you operate as a seller — measured through dispatch performance, cancellation behaviour, return and quality signals, and customer-facing outcomes. It matters because it influences how much visibility your catalogue receives and, at the extreme, whether you can continue selling.
Dispatch reliability and seller-initiated cancellations are the two that consistently carry direct consequences across marketplaces, because both directly affect a customer who has already paid or committed. Quality and return-related signals matter but usually act more slowly and diffusely. Rating and responsiveness metrics tend to be influential rather than punitive.
Less directly than sellers fear. RTO is a delivery outcome rather than a policy breach, so it is not usually penalised the way a missed dispatch is. It does signal underlying listing or targeting problems, and it destroys margin regardless. Treat it as an economics emergency rather than a compliance one.
Generally: metrics degrade quietly, then notifications or warnings appear in the panel, then visibility or catalogue reach is reduced, then harder restrictions apply. The important point is that the early stages are easy to miss because nothing dramatic happens — the damage accumulates before anyone tells you clearly.
Weekly is enough for most small sellers, and far better than the common pattern of checking only when something goes wrong. The purpose is to catch a trend while it is still a trend. By the time a warning is explicit, you have usually been drifting for some time.
Dispatch reliability. It is the one you control almost entirely, it is measured on every order, and failures compound into both penalties and lost sales through cancellation. Sellers who get dispatch right rarely find themselves in account-health trouble from the other metrics alone.
Yes, but slowly, because most metrics are computed over a rolling window. That means recovery requires a sustained period of good performance rather than a single good week — and it is why letting metrics slide is expensive even when nothing visible has happened yet.
The specific names, thresholds and calculation windows differ, but the underlying concerns are remarkably consistent: did you dispatch on time, did you cancel, did the customer get what they expected. If you sell on more than one platform, the operational discipline transfers even where the metric definitions do not.