Most sellers meet the incident process at the worst possible moment: after a buyer has raised something, while a deadline runs, without knowing what the stages are or what a good response looks like.
It is worth understanding before you need it, because the escalation path ends somewhere genuinely damaging — at the seller account, not at the individual order.
What raises an incident
An incident begins when a buyer reports a deviation from what was contracted. In practice that is almost always one of:
- Late delivery, or non-delivery
- Specification mismatch between what was ordered and what arrived
- Quantity or documentation discrepancies
- Quality failures found on inspection or in use
Note what these have in common: they are all contract performance issues, and they are all things a buyer experiences rather than things the portal detects. Your rating and your delivery discipline are what keep incidents from being raised in the first place.
The escalation path
Left alone, an incident does not stay an incident.
1. Incident raised. The buyer records the deviation. At this stage it is a request for explanation, and it is the cheapest point at which to resolve it.
2. Show-cause notice. If the incident is unanswered or the response is inadequate, it escalates to a formal notice asking you to show cause why action should not be taken. The tone and the stakes both change here.
3. Action against the account. Depending on severity and history, this can include suspension of your ability to receive new orders — a temporary moratorium — alongside other consequences.
4. Appeal. There is a route to contest outcomes, but it is a slower and more formal process than simply answering well at stage one.
The asymmetry is the point. Answering properly at stage one costs you an afternoon and some document retrieval. Getting to stage three costs you every order you would have received while the account is frozen.
Why one incident can freeze a healthy account
This is the part sellers underestimate, so it deserves stating plainly.
A moratorium applies at the seller level, not the order level.
That means a single unresolved incident on a single small order can stop your entire account from receiving new orders, including in categories and from buyers entirely unrelated to the complaint. A seller with ninety-nine clean deliveries and one ignored incident is a seller who cannot trade.
It is also why the incident that gets ignored is so often a trivial one. Nobody ignores a large dispute. People ignore a ₹40,000 order with a documentation discrepancy, because it seems too small to matter, and that is the one that freezes the account.
What a good response looks like
Incidents are closed by evidence against the specific allegation. They are not closed by apology, explanation of context, or assurances about your general reliability.
Structure the response as:
- The specific allegation, restated so it is clear what you are answering.
- The factual account, with dates.
- The evidence, attached and referenced — dispatch proof, transporter documents, delivery acknowledgement, test certificates, inspection reports, the correspondence trail.
- What you propose, where something is genuinely owed — replacement, collection, credit — with a date.
Two failure modes to avoid:
- The generic apology. “We regret the inconvenience and assure you of our best services” answers nothing and closes nothing.
- Arguing the wrong point. If the allegation is late delivery, evidence about product quality is irrelevant however good it is. Answer what was actually asked.
Where the buyer is factually right, saying so early and proposing a concrete remedy resolves incidents far faster than contesting them.
Read the current policy, not a summary
The specific timelines, severity bands and consequences are defined in GeM’s own incident management policy, and they are revised. Anything you read in an article — including this one — is a description of how the process works in shape, not a substitute for the current document.
Before you respond to a notice, check the policy text in force and the deadline stated in the notice you actually received.
Prevention, which is mostly boring
Sellers who rarely see incidents tend to do the same unremarkable things:
- Confirm before accepting. Check that you can actually meet the delivery date and the exact specification before accepting the order, not after.
- Follow through on CRAC. Consignee Receipt and Acceptance is where an order becomes clean. Chase it rather than assuming silence means acceptance.
- Keep dispatch evidence by default. Photographs, transporter documents and delivery acknowledgements retrieved months later are the difference between a closed incident and an escalated one.
- Answer the buyer before the portal has to. Most incidents were preceded by an email the seller did not reply to.
Monitor weekly
For clients on retainer we check incidents weekly, and the reason is not diligence theatre — it is that notice periods run whether or not you have noticed. An incident raised on a Monday and read a fortnight later has already consumed most of the window in which a good response was easy.
If your account is generating orders you cannot personally track, someone needs a standing weekly check. That is a fifteen-minute task that protects everything else.
If you are already at show-cause
Do not draft it alone under time pressure if the stakes are material. Get the exact allegation, gather the evidence that speaks to it, and respond within the stated window even if the evidence is incomplete — a substantive partial response beats a missed deadline.
Our order, payment and incident management service covers response drafting and weekly monitoring, and the first consultation is free. If you have a live notice, say so when you get in touch — those are the ones where a day matters.