Shoppers will pay a return fee that makes sense for the situation. The one they resent is the fee that lands on everyone. In the consumer research IMRG and nShift published in July 2026, 47.6% of UK shoppers said a fee is fair for frequent returners and 26.2% said the same for international orders, while 26.3% actively avoid retailers that charge for returns at all. The policy that keeps both groups is conditional, and until now a conditional policy meant a change request every time the condition changed.
This month we added a rule engine to nShift Returns that sets return fees and return periods from the order itself: free returns above a basket value, a longer window for orders placed during a campaign, no fee when the customer takes store credit, different terms for a customer group or a store. You build the rules under Settings in the portal your team already uses, and you change them yourself.
The rule engine in nShift Returns, released earlier this month, sets return, claim and exchange fees and periods automatically from conditions on the order: order value, order date, order ID, refund method, or a custom field you send with the order such as customer group, product category or store. Each rule applies to the markets and flow you choose, and when two rules match, the one higher in the list wins.
Why one number per market stopped being enough
A market policy in a returns portal is one fee and one window for every order in that market. It can't tell a €30 order from a €300 one, a Black Friday order from a January one, or a loyalty member from a first-time buyer. So most retailers have picked one of two blunt settings, a fee for everyone or free returns for everyone, and accepted the cost that comes with it: lost orders on one side, subsidized returns on the other.
Most shoppers still want free returns: 48.3% of consumers in the IMRG and nShift survey rated them very important when choosing where to shop, and another 37.4% called them quite important. At the same time, shoppers draw a line between fees that fit a situation and fees that fall on everyone, and 21.5% say charging is never fair. The full breakdown by scenario is in our summary of the findings.
47.6%
of UK shoppers think a return fee is fair for frequent returners
IMRG and nShift, The Ecommerce Returns Opportunity, 1,000 UK consumers, July 2026
26.3%
actively avoid retailers that charge for returns
Same survey
21.5%
say charging for returns is never fair
Same survey
Put those two findings side by side and you get a policy with conditions in it: a fee that applies when the order is small and disappears when it is large, a window that stretches for gifts bought in December, terms a customer group has earned. Most returns portals hold one setting per market, so every one of those exceptions went through a ticket, or through a manual refund adjustment after the fact.
How a rule works in nShift Returns
A rule is a set of conditions and a set of actions. When every condition matches the order, the actions apply. The simplest rule in the Help Center reads: if Order total is greater than or equal to 100, then Return fee is 0.
A condition is a field, an operator and a value. The fields a rule can check today:
- Order total, a number, as it was imported into nShift.
- Order create date, a date, also as imported.
- Order external ID, text, so a single order or a numbering range can be matched.
- Refund method, text, such as Store credit.
- Custom fields you send with the order. If your ecommerce platform or CRM already passes a customer group, a product category or a store ID, those fields can be added to your account and used as conditions.
Operators are equal to and not equal to; greater than and less than, each with an "or equal to" variant; contains and does not contain for text; and "in list", which matches any one of several values, so one condition can cover a set of stores or customer groups.
An action sets one field to one value, replacing the market's standard setting for the orders that match. A rule can carry several actions, though it cannot set the same field twice.
| Action | Flow | What it sets |
| Return fee | Returns | The fee deducted from the refund |
| Return shipping fee | Returns | The shipping cost charged for the return |
| Return period | Returns | Days the customer has to start a return |
| Claim fee | Claims | The fee deducted from the refund |
| Claim period | Claims | Days the customer has to submit a claim |
| Exchange fee | Exchanges | The fee deducted from the refund |
| Exchange period | Exchanges | Days the customer has to start an exchange |
Each rule applies to the markets you select, one or several, and to one flow: Returns, Claims or Exchanges. Markets already separate countries by currency and policy; rules separate orders inside a market, and one rule can cover several markets, so a campaign window for the Netherlands, Belgium and Germany is one rule rather than three.
Rules are evaluated from the top of the list down. If two matching rules set the same field, the rule at the top wins. Ordering the list is the part that takes some thought: once a loyalty rule, a campaign rule and a basket-value rule share a market, their order decides which fee a member's Black Friday order gets. Keep the exception above the general case.
All of this is under Settings, then Rule engine, in the Returns portal: create a rule, name it, pick the markets and the flow, add at least one condition and one action, save. The Help Center article on the rule engine walks through every field and operator.
What a merchant does with it
Each of the situations below is a single rule, built from the conditions above.
Free returns above a basket value
If Order total is greater than or equal to 100, then Return fee is 0. Add a second action, Return shipping fee is 0, if the label should be free as well. Orders below the threshold keep the market's standard fee, untouched. A retailer that wants the threshold to track average order value changes one number in one rule.
A longer window during a campaign
Two conditions on the same field, both of which must match: Order create date is greater than or equal to 20 November, and Order create date is less than or equal to 24 December. The action: Return period is 60. Gifts bought in late November can still come back in January without anyone extending the window by hand, and the rule stops matching on its own once the dates pass. Set the markets on that rule to every market running the campaign and the window is the same everywhere it should be.
No fee when the refund stays in the business
If Refund method is equal to Store credit, then Return fee is 0. The customer who chooses credit keeps the value in your store, and the fee that made sense for a cash refund no longer does. The same shape works for a reduced return shipping fee on store credit, if you would rather soften the fee than remove it.
Different terms for a customer group or a store
This one needs a custom field. Send the customer group, the store ID or the product category with the order, and a rule can read it: Customer group is equal to Member, Return period is 45. Store ID is in list, followed by the stores in question, Return shipping fee is 0. If your CRM already scores customers on their return behavior, send that segment with the order as a custom field and a fee for frequent returners becomes a rule you can switch on.
What the rule engine leaves alone
The market's standard policy stays the default. A rule replaces a setting only for the orders it matches, and removing the rule puts the default back. Nothing about a market has to change to add an exception to it.
Rules read the order as it was imported into nShift: the fields listed above, plus your custom fields. They do not read the reason the customer selects at return time, so a policy that depends on the reason belongs in your return-reason setup.
Claims and exchanges are separate flows with their own fees and periods. A rule written for the Returns flow never touches a claim, which keeps a fee threshold well away from a faulty item.
Statutory rights apply before any rule does. The 14-day withdrawal right and the treatment of faulty goods are the legal baseline of an ecommerce returns policy; the rule engine sets the commercial policy you offer on top of it.
Rules inside the returns flow customers already use
The rule engine is part of nShift Returns, the returns management software that gives customers a branded portal on your own URL, lets them choose how to send the item back, and passes the return to your warehouse and ERP so refunds, exchanges and restocking run without a ticket. The policy the rules produce is the policy the customer sees in that portal, in the market's language and currency.
The results retailers report from that flow come from digitizing the return itself. Since going live on nShift Returns, Hunkemöller has seen a 15% shift from return-to-warehouse to in-store returns, which creates opportunities for assisted service and repurchase in store. Fashion retailer Quiz expects to save up to 30% in staff time through digitalized, paper-free returns. Rules add policy control to that same flow.
If you get stuck on a condition, nShift Companion is built into the Returns portal. Ask it how to set up a rule and it answers in plain language from the Returns documentation. It explains and suggests; you make the change.
Set the first rule this week
Start with the rule your team has debated longest, usually the free-returns threshold or the campaign window, and run it in one market before you extend it to the rest.
If you are not on nShift Returns yet, bring the return policy you want to run and we will show you how it reads as rules.
Frequently asked questions about return rules
Can I charge different return fees for different orders?
Can I give customers a longer return window during a campaign?
Can I waive the return fee when the customer chooses store credit?
What happens when two rules match the same order?
Does a rule change my market's standard return policy?
About the author
Frida Wikingsson
Frida leads the nShift Returns roadmap, from the rules that shape fees and return windows for each order to the tools used by warehouse and support teams every day. She joined nShift in 2021, working across technical support, project delivery and presales solutions architecture before moving into product. That experience gives her a firsthand view of returns, from configuring customer flows to designing the product behind them.