The free delivery threshold is one of the strongest levers for basket value in e-commerce, and at the same time a decision most often made by copying competitors. The problem is that someone else’s threshold works on someone else’s basket distribution. This guide shows how to set the threshold from your own data: from the order value histogram, the top-up zone and the margin calculation.
Mechanism
How a free delivery threshold works
Working definition
The top-up zone is the range of basket values just below the threshold from which customers actually add products to avoid the delivery charge. It works when the missing amount is comparable to the delivery price and achievable with one added product.
The mechanism has two sides. Gain: customers with baskets below the threshold add products (order value and margin grow), and some baskets that would have been abandoned because of the delivery cost go on to purchase. Cost: orders that would have exceeded the threshold without any incentive get free delivery as a gift, and its cost comes out of margin. The whole art of setting the threshold is maximising the first side while keeping the second under control. That is why there is no “good threshold” independent of the shop’s data.
Data instead of intuition
Basket histogram: where the threshold really lies
The first step is the order value distribution, not the average. The average basket masks the shape on which everything depends. An example from a stationery shop with a threshold set “by eye” at €49.75 (illustrative data):
Order value distribution · monthstationery shop · threshold: €49.75 · illustrative data
Diagnosis at a glance: baskets cluster in the €20–30 range, while the threshold sits at €49.75. The top-up zone (€40–49.75) covers barely 11% of orders, and a typical €24 basket is more than €25 short: nobody adds that much. This threshold exists mainly on the banner.
The rule for setting the threshold: just above the natural basket cluster, within reach of one added product. For the distribution in the example, the candidate is around €30–32.50: a typical €24 basket is then roughly the price of one product away from the threshold, and the top-up zone covers the densest part of the distribution.
Calculation
Threshold calculation: three scenarios on margin
Threshold candidates are compared using scenarios before anything changes in the shop. For the distribution in the example (1,000 orders per month, illustrative data):
| Threshold scenario | Orders with free delivery | Subsidy cost / month | Top-up zone | Character |
|---|---|---|---|---|
| Threshold €24.75 | 51% | ~€1,775 | 24% of orders | gives delivery away to half of orders, which often exceeded this amount anyway |
| Threshold €32.25 | 30% | ~€1,050 | 27% of orders | the top-up zone hits the basket cluster; typical shortfall: the price of one product |
| Threshold €49.75 (current) | 9% | ~€315 | 11% of orders | cheap because it is dead: almost nobody tops up, negligible effect on the basket |
The cost of subsidies alone does not decide the scenario: it has to be set against the margin from top-ups. If, in the €32.25 scenario, even some customers in the zone add a product worth €6.25–10, the margin from those additions works against the subsidy cost; a test on the live shop with a period comparison settles it. An important detail: customers often top up with the cheapest products, so the calculation is done on the margin of the added products, not on their price (the guide to margin in e-commerce explains why this matters).
Anti-patterns
Four mistakes when setting the threshold
A threshold copied from a competitor
A competitor has a different basket structure, different delivery prices and a different margin. Applied to your distribution, their threshold can be both too expensive and useless at the same time.
A threshold below the median basket
Most orders exceed it without any change in behaviour: subsidy costs rise while the basket stays flat. This is the most expensive version of “free delivery here”.
A threshold out of top-up reach
When the gap to the threshold is larger than the price of a sensible product, the top-up zone is empty. The threshold appears on the banner and does nothing.
No progress counter in the basket
The top-up zone works when the customer sees “€X away from free delivery” alongside product suggestions. Without this information, even a well-calculated threshold works at half capacity.
DataOrganizer
How DataOrganizer calculates baskets and delivery costs
This whole calculation rests on three datasets from the shop, which DataOrganizer keeps together: the distribution of order values, delivery analytics (methods and costs) and margin per product. Setting the threshold stops being guesswork:
Diagnosis of the current threshold and a proposed test:
| Metric | Value |
|---|---|
| Median basket | €24 |
| Basket cluster | €20–30 |
| Orders reaching the €49.75 threshold | 9% |
| Orders in the top-up zone of a €32.25 threshold | 27% |
| Average delivery cost (dominant method) | €3.48 |
The current €49.75 threshold sits more than €25 above the median basket: 9% of orders reach it, and the top-up zone is practically dead. Proposal: test a €32.25 threshold, which puts free delivery within reach of one added product for the densest part of the distribution. To monitor during the test: average order value and margin, the share of orders with free delivery and the total subsidy cost.
Methodological note: all figures in the examples are illustrative. Basket distribution, delivery costs and margin differ in every shop, and the effect of a threshold change is verified by comparing periods before and after (adjusted for seasonality), just like any pricing change.
Key takeaway
The free delivery threshold can be calculated to the last euro. The basket histogram shows where it belongs, the top-up zone shows who will add more, and the margin calculation decides whether it all adds up. All three elements are in the order data.
FAQ
Frequently asked questions about free delivery thresholds
How do you set a free delivery threshold?
From a histogram of your own order values: the threshold goes just above the basket cluster, within reach of one added product, and then scenarios are calculated on margin (subsidy cost versus margin from top-ups). Copying a competitor’s threshold ignores the only data that matters: yours.
What is the top-up zone?
The range of basket values just below the threshold from which customers actually add products. It works when the missing amount is comparable to the delivery price and reachable with one product; a threshold far from the basket cluster has an empty zone.
Does free delivery above a threshold pay off?
It depends on the calculation: subsidies on orders that would have crossed the threshold anyway, versus margin from top-ups and recovered baskets abandoned because of delivery cost. The result changes with the threshold position, so several scenarios are compared on your own distribution, and a test decides.
Which is worse: a threshold too low or too high?
Too low (below the median basket) gives away delivery without changing behaviour: pure cost. Too high leaves the top-up zone empty: no effect, plus a discount for the few largest baskets. The right place is just above the cluster, within top-up reach.
How does DataOrganizer help?
It calculates the basket value distribution, delivery analytics with method costs and margin per product from orders, and the AI assistant turns this into a diagnosis of the current threshold and scenarios for a new one, with a list of metrics to monitor in the test. The decision is based on figures from your own shop.
Cheat sheet for the whole guide
The free delivery threshold in three steps
1. Draw a basket histogram. The cluster of order values shows the only sensible neighbourhood for the threshold; the average will not show it.
2. Set the threshold within top-up reach. Just above the cluster, so that a typical basket is short by roughly the price of one product, with a “how much is missing” counter in the basket.
3. Calculate scenarios on margin and test. Subsidy cost versus margin from top-ups, then a before-and-after period comparison. DataOrganizer keeps baskets, delivery and margin in one model, so the whole calculation fits into a single question to the chat.
DataOrganizer · baskets, delivery and margin together
Connect your shop system: order value distribution, delivery costs and margin per product come together in one calculation. Demo account available instantly; you only pay for connected sources.