Choosing a courier you can trust

The pickup boy didn't come again. Eleven parcels are stacked by the door, four of them promised for today, and the number on the courier's website rings out.

Nobody chooses a courier on the thing that actually matters. They compare the headline rate, pick the cheapest, and discover four months later that the rate was never the expensive part. What decides whether a courier works for a small Indian seller is four things: whether they genuinely service the pincodes you ship to, how they charge weight, how long they sit on your COD money, and what an RTO costs you. On a ₹899 kurta with maybe ₹280 of margin, one badly handled return wipes out three good orders, and no rate card difference of ₹8 a parcel comes close to that.

The rate card is the least useful page in the deck

A forward rate of ₹42 per 500g against ₹49 per 500g looks like a 17% saving. On 200 parcels a month that's ₹1,400.

Now put an RTO next to it. A returned parcel usually costs you the forward leg, the return leg, and often the COD handling fee even though no cash was collected. That's three charges on an order that earned nothing, plus the packaging, plus a fortnight of stock sitting in a van instead of on your shelf. Get five more of those a month than you should and the ₹1,400 is gone twice over.

So compare on the things below, in this order, and treat the rate card as a tiebreak.

Your pincodes, not their pincode count

Every courier's homepage says they cover 20,000-something pincodes. That number tells you nothing.

Take your last 200 orders, pull out the pincodes, and count the top 30. That's usually 60% or more of your volume. Then check those 30 against each courier's serviceability tool, one at a time. It's twenty minutes of clicking and it's the single most useful thing in this article.

What you're looking for, per pincode:

• Is it serviceable at all?
• Is it serviceable for COD, which is a separate answer and often a no?
• Is it a reverse pickup pincode, which is a third answer and more often a no?
• Is it surcharged as an ODA or remote location?

A courier that covers 28 of your top 30 with COD beats one that covers 19,000 pincodes nationally. And the two it doesn't cover are what your second courier is for.

Weight slabs, and the rounding nobody mentions

Couriers bill in slabs, and the slab size changes the price more than the rate does.

A 520g parcel on a 500g slab is billed as two slabs. On a 1kg slab it's billed as one. If your typical parcel sits just over a slab boundary, moving to a courier with a different slab structure can cut your cost more than any negotiation will.

Then there's volumetric weight, the same trap as everywhere else: length x breadth x height divided by a divisor the courier sets, and you're billed on whichever is higher. Apparel in a poly mailer is almost never volumetric. Anything in a carton usually is. Find the divisor on your rate card, measure your three commonest boxes, and do the sum before you sign.

Weigh twenty packed parcels on a kitchen scale and write down the distribution. If most of them are 480g to 540g, you have a packaging problem worth solving before you have a courier problem.

COD is a loan you are making to your courier

This is the one that quietly decides how fast you can grow.

When a courier collects ₹899 at a buyer's door, that money is theirs until the remittance cycle pays it to you. On 200 COD orders a month at ₹899, roughly ₹1.8 lakh a month is passing through, and at a week's cycle something like ₹42,000 of your money is sitting in their account at any moment. That's stock you can't buy.

Aggregators commonly run remittance around a week, sometimes with a paid option to get it faster; direct contracts vary and can be quicker at volume. Do not take that from me or from any other blog. Ask for the cycle in writing, in days, and ask what "day" means: the day of delivery, the day of the scan, or the day their finance team runs the batch. Those are three different answers and the gap between them is often four working days.

Then ask the follow-up that gets ignored: what happens to a remittance that's short. You will need this. The routine for catching it is the same one you'd run on any payment rail, and it's in how to know a bank transfer actually arrived. Substitute the courier's remittance statement for the bank statement and nothing else changes.

RTO is where the money actually goes

Return to origin is the defining cost of Indian ecommerce and most first contracts handle it badly. Get answers to all five of these before you ship anything:

1. Do I pay forward and return, or just return? Both is normal. Know which.
2. Is the COD fee charged on an RTO? Sometimes yes, on an order where no cash was ever collected.
3. How many delivery attempts before it returns? Two is common. Three is better. One is a red flag.
4. How long does it sit before it comes back? A parcel held for eleven days is stock you can't sell.
5. Do I get the parcel back sealed? Ask. Then check, because "returned" and "resellable" are different words, particularly for anything cosmetic.

Measure your own RTO rate from your first month and track it monthly by courier and by state. National averages are interesting; yours is the only one you can act on. The techniques for pushing it down live upstream of the courier, in confirmation and deposits, and they're in cash on delivery for small sellers.

Is there a person

At some point a parcel will be stuck at a hub for nine days and the tracking page will say "in transit" the whole time. What you need then is a human.

Ask, before you commit:

• Do I get a named account manager, or a ticket queue?
• What's the response time on a stuck-shipment ticket, in hours?
• Is there a WhatsApp number that a person actually reads?
• What's the escalation path when the first answer is "please wait 48 hours"?

Then test it. Raise a real ticket in week one about a real parcel and time the reply. A courier that answers in three hours during the sales pitch and eleven days in November has told you something important, and you want to find that out in July.

Aggregator or direct

	Aggregator	Direct with a carrier
Rates	Discounted through pooled volume, good at low scale	Better once you're at real volume, negotiable
Setup	Minutes. No contract, no minimum	Paperwork, a KYC process, sometimes a minimum
Coverage	Several carriers behind one panel, auto-routed	One network, one set of gaps
COD remittance	Set by the aggregator, often about a week	Negotiable, sometimes faster
When a parcel is stuck	You raise it with the aggregator, who raises it with the carrier	You raise it with the carrier
Best for	Under roughly 300 parcels a month, or mixed geography	Concentrated volume in one region, or a fragile product needing a specific network

Most sellers under a few hundred parcels a month should start with an aggregator, and the reason is not the rate. It's that one panel gives you five networks and lets you fail over when one of them stops picking up. The cost is a layer between you and whoever is actually holding your parcel, which hurts exactly when you need it most.

Reading a rate card without getting fooled

Headline rates are quoted per 500g forward, and that is never the number you pay. Build this table for each courier with their real figures. The ones below are shaped like a real card, not quoted from one.

Line	Your number	Example
Forward, first 500g		₹42
Each additional 500g		₹38
Your typical parcel weight		700g, so two slabs
Subtotal		₹80
Fuel surcharge, %		+₹12
COD handling, higher of flat or %		+₹35
GST on the lot		+
Landed cost per COD order		around ₹150
RTO cost, when it happens		forward + return + COD fee
RTO rate		your own measured %
True cost per order, RTO-adjusted		landed + (RTO cost x RTO rate)

That last line is the only comparable number, and almost nobody calculates it. A courier that's ₹6 cheaper per parcel and two percentage points worse on RTO is more expensive, and you will not see that anywhere on their pricing page.

The twenty-parcel test

Before you move your whole volume, run this. It takes a week.

Send twenty parcels through the new courier: five to your best pincodes, five to your worst, five COD, and five to addresses you control. Then deliberately refuse one, so you see the RTO process end to end rather than reading about it.

Record four things for each: days to delivery, number of attempts, whether the tracking updated in real time, and whether the COD money appeared when the cycle said it would. Twenty parcels will tell you more than any sales call, and the one you refused will tell you the most.

Worked example: Nikhil, cotton kurtas, Jaipur

Nikhil sells block-printed cotton kurtas at ₹899, about 240 orders a month, roughly 80% COD, mostly to Rajasthan, Delhi NCR and a long tail across UP and MP. His margin is around ₹280 a piece.

He'd been with one aggregator on the cheapest available routing. His month: 240 orders, 41 RTOs, which is 17%. Each RTO cost him forward plus return plus the COD fee, about ₹210 all in, so ₹8,610 gone. Nine of the returned parcels came back with the poly bag opened and two were unsellable.

He did three things.

1. Checked his top 30 pincodes properly. Six of them were routing to a carrier with a two-attempt policy and a poor record in his own tracking data. He pinned those pincodes to a different carrier in the panel.
2. Split his volume. Metro and NCR to one carrier, tier-3 Rajasthan and MP to another with better rural reach, even though it cost ₹9 more a parcel.
3. Started measuring RTO by state, in a spreadsheet, monthly.

Two months later: 252 orders, 28 RTOs, 11.1%. The extra ₹9 on part of his volume cost him about ₹1,100 a month. The RTO reduction saved him roughly ₹2,700. Small numbers, but he also got back the fortnight of stock that used to be in transit both ways, and on a business holding 300 pieces that's the part that actually changed how he buys fabric.

The thing he didn't expect: two of his six problem pincodes weren't a carrier problem at all. They were one apartment complex where nobody ever picked up an unknown number.

When to run two couriers

Almost everyone should, past about 100 parcels a month. Not for price, for redundancy.

One will have a bad week. A hub will flood, a strike will happen, a festival surge will bury them, and you need somewhere to put Tuesday's parcels. Keep a second account live with a handful of parcels a week going through it so it's warm, the KYC is done, and you're not signing up for the first time in the middle of a crisis.

The natural split is geographic: whoever is strongest in your metro, and whoever is strongest in your long tail. It also gives you real comparative data, which is worth more than any negotiation you'll ever have.

What Sailo does here, and what it doesn't

Every Sailo order has fields for the carrier, the tracking number and a tracking URL, plus a shipped timestamp and a status you move through new, confirmed, shipped, completed, cancelled and refunded. You can create a named delivery method per service, with its own fee and its own free-text estimate, and the choice is stored on the order.

The limitation: none of that is a courier integration. Sailo doesn't fetch rates, print labels, generate manifests, push orders to a panel or read a tracking status back. You copy the AWB in from wherever you booked it. At 240 orders a month that's a couple of minutes a day; at 2,000 it's a job, and you'd want a panel that does it for you.

The bigger limitation for an Indian seller is on the money side. The most effective way to cut RTO is to get more orders prepaid, and the obvious lever here is UPI. Sailo has no UPI rail. There's a bank transfer method where you can write your UPI ID into the instructions, and buyers will pay it perfectly happily, but Sailo won't know they did and can't confirm it. Card payments carry 1–3% of the goods, which is a lot of subscription for a seller whose buyers were going to pay by UPI anyway. Being clear about that is more useful than pretending otherwise.

Do this this week

Pull your top 30 pincodes and check them, one at a time, against two couriers' serviceability tools. Mark which are COD-serviceable and which take reverse pickups.

Then ask both for their remittance cycle in days and their RTO charging rule in writing, and build the RTO-adjusted cost line from the table above. That's your decision, and it takes an afternoon. Before you do any of it, make sure you know what you should be charging buyers in the first place, because the two numbers have to agree: how to work out delivery charges covers that.