A shopper fills a basket, goes to pay, and a postage charge appears that they had not been thinking about. Some of them pay it. A good number close the tab instead, and a few of those go and buy the same product from a shop that told them the postage cost up front.
The interesting part is that the charge is rarely the real problem. Perfectly reasonable postage, revealed at the last moment, does more damage than higher postage stated up front, because at the final step it does not read as a cost, it reads as a catch.
First, work out what it is actually costing you
Before changing anything, find out whether you have a problem and where it happens. Your analytics can tell you how many people reach the basket, how many reach the checkout, and how many complete. The gap between those steps is the interesting number, and a large drop at the moment postage first appears is a clear signal.
It is worth knowing your average order value at the same time, because almost every decision below depends on it. If you do not know what a typical basket is worth, you cannot sensibly set a threshold, and you will end up picking a round number that feels right and quietly costs you margin.
Ask a couple of recent customers, too. People are surprisingly willing to tell you what nearly stopped them.
Free delivery is not the only answer
The reflex is to make delivery free and absorb it. Sometimes that is right. Often it is expensive and unnecessary, and it is worth knowing the alternatives before committing.
A threshold is the most common approach, and it does two jobs at once. It removes the objection for larger baskets and it gives smaller baskets a reason to grow. Set it a little above your average order value, not at it, so that a typical shopper is a short distance from qualifying rather than already there. Too far above and it stops being motivating and becomes an annoyance.
A choice of speeds works better than most people expect. A cheaper, slower option alongside a faster paid one lets the shopper decide what matters to them, which converts better than deciding on their behalf. Plenty of people will happily wait an extra two days to save a few pounds, and plenty will pay for next day when it is genuinely needed.
Collection is worth offering if you have somewhere for people to come. It costs you almost nothing and it appeals to exactly the local customers who are most likely to come back.
And building it into the price is a legitimate option, though it works best when your prices are not directly comparable to a marketplace listing. If they are, a higher price with free delivery can lose the comparison before the delivery is ever seen.
Say the number earlier
Whatever you decide, the single biggest improvement is usually not the charge but when you mention it.
Postage should be answerable on the product page, before anything is added to a basket. If the item qualifies for free delivery on its own, say so on the item. If it depends on the basket total and the shopper already qualifies, tell them, because it changes how the next click feels. If there is a charge, state it, or state the range if it varies by address.
A banner is worth considering if your terms are simple enough to state in a sentence. “Free UK delivery over £40” in the header of every page removes the question before it forms, and it is one of the highest-return changes available to most shops.
The same logic applies to the basket itself. Showing progress towards a threshold, along the lines of “spend £6 more for free delivery”, is helpful rather than pushy, because it is telling somebody something they would want to know. It also lifts basket sizes more reliably than most discounting.
Dates beat durations
Cost is only half the question. The other half is when it turns up, and this is where most shops are unnecessarily vague.
“Three to five working days” is a hedge that means nothing to somebody wondering whether a birthday present will arrive in time. An estimated delivery date, or a cut-off such as “order by 2pm for dispatch today”, answers the actual question. Certainty is worth money to people. Vagueness is not.
If you cannot be precise, be honest about the range and say what it depends on. That still beats a phrase that could mean anything.
Surcharges and the places you do not ship
Anything that varies by address needs stating before the checkout, not at it. Highlands and islands surcharges, longer times for Northern Ireland, and international rates are all reasonable, and all of them feel like a trick when they appear at the end.
The same goes for anywhere you do not deliver at all. Somebody who has spent ten minutes choosing and is told at the final step that you cannot ship to them will not be back, and they will remember it. Say it early, on the delivery information you have already put on the product page.
If you ship internationally, be clear about duties and import charges. A parcel that arrives with an unexpected bill attached produces a complaint, a refund request, and often a review, none of which were worth the sale.
What to do with the money
If you decide to absorb some delivery cost, the temptation is to take it straight off the bottom line and hope volume makes up the difference. It is worth being a little more deliberate.
Work out what a customer is worth beyond one order. A shop where people buy several times a year can justify carrying more delivery cost than one selling a single expensive item somebody buys once, because the cost is being spread across a relationship rather than a transaction. That calculation, rather than what a competitor is doing, is what should set your threshold.
Watch your margins by product too. Flat rate free delivery over a threshold is simple and clear, which is its main strength, but it treats a light item and a heavy one identically. If some of your range is genuinely expensive to send, it is better to say so on those items than to quietly lose money on every one you sell.
Where to start
Look at the product page for your best seller and count how many clicks it takes to find out what postage costs. If the answer is more than none, that is the change to make first, and it is usually a small one.
After that, find your average order value and check your threshold against it. Those two together account for most of what delivery costs a typical shop, and neither requires a rebuild.
There is a broader piece on the other reassurances that belong on a product page in anxiety busters: what belongs on a product page, and where.