Which Courier Delivers Best Where in Bangladesh

Published rates, checked July 2026 by Banikh against the couriers' own calculators. Rates change, so confirm before you book.

Two things are worth noticing before the table. Steadfast prices by route rather than by weight band, which makes its outside-Dhaka rate straightforward to predict. RedX charges no cash-on-delivery fee inside Dhaka City, which matters more than the delivery charge on a high-value order, because the cash-on-delivery fee is a percentage of the order and the delivery charge is not.

Usually not. The number that matters is landed cost per delivered parcel, not the sticker rate per parcel shipped.

The arithmetic is simple and almost nobody does it. Take the delivery charge, add the cash-on-delivery percentage on the order value, then divide the whole cost of shipping a batch by the number of parcels in that batch that were actually accepted and paid for. The parcels that came back still cost you money; they just produced none.

A courier five taka cheaper per parcel that delivers a few percentage points less often in your main district is more expensive, and the gap widens as your average order value rises, because a returned parcel takes the whole order value with it and not just the delivery charge.

This is why a price table is a starting point rather than an answer. It tells you the numerator. Your own delivery data is the only place to get the denominator.

There is no single answer, and the honest version is that the best courier for Dhaka changes between thanas.

Dhaka is not one delivery market. Coverage density, rider familiarity and hub proximity vary sharply across the city, and a courier that is excellent in one thana can be ordinary two thanas away. A brand shipping mostly to Dhanmondi and a brand shipping mostly to Uttara are not solving the same problem, even though both would read the same city-level comparison.

The same applies outside Dhaka with more force, because coverage there is genuinely uneven rather than merely variable. Some couriers reach a district town well and its surrounding upazilas poorly.

Take thirty days of shipments, group them by courier and destination zone, and calculate the delivery success rate for each pair. That table is your map.

You already have every input. Courier panels expose the status of every consignment, and your own order records hold the destination. The work is joining them and being disciplined about the grouping.

Group by destination zone rather than by district. District is too coarse to act on in Dhaka and often too coarse outside it. Keep the window rolling rather than fixed, because a hub that has just lost riders will show up in the last thirty days and not in a figure you calculated in March.

Set a minimum volume before you trust a cell. Ten parcels through one courier into one zone is not evidence of anything. Below that threshold, fall back to the district-level number rather than acting on noise.

Then state your objective explicitly, because it changes the answer: fewest returns, fastest delivery, lowest cost, or quickest cash collection. The same batch of orders routed against a different objective produces a different courier split, and a brand that has never written its objective down is usually optimising for whichever courier someone signed up with first.

Often enough that an annual review is worthless and a live one is not.

Courier performance in a zone moves with things you cannot see from outside: hub staffing, rider turnover, a new depot, a route reassignment. None of it is announced. It shows up in your delivery rate a few weeks after it happens.

This is the argument for rolling data rather than a one-off comparison. A choice made once and never revisited quietly decays, and the cost of that decay never appears as a line item. It appears as a return rate that drifted up and nobody could explain.

Several, if you ship across more than one kind of area. One, if your volume is concentrated and small.

Multi-courier routing only pays when the differences between zones are larger than the overhead of running more than one account. If nearly all your orders go to two thanas in Dhaka, that overhead is real and the gain is small. If you ship nationwide, the gain is where most of your margin is hiding.

The practical obstacle is not the decision, it is the typing. Running three couriers by hand means three panels and manual entry for every order, which is exactly the friction that pushes people back to a single default. Omnistra integrates six Bangladeshi couriers natively, RedX, Pathao, Steadfast, eCourier, Paperfly and Delivery Tiger, and picks per order against the objective the brand sets, which removes the typing rather than the decision.

Whichever way you do it, the principle holds on its own: the courier is a decision, not a habit.