What Does an Undelivered Parcel Really Cost You? Calculate the True Cost of Bad Addresses

A €48 order comes back to your warehouse a week later, stamped "addressee unknown". The customer left one digit out of the house number, doesn't pick up the phone, and only gets in touch to ask where their parcel is. At first glance you've lost the shipping fee. In reality, you've paid a lot more.

Undelivered parcels hide across several lines in your books: shipping, warehouse, customer service, lost revenue. That's why few merchants ever see them as one number. In this article we'll add them up with a worked example, show you how to calculate your own figures from an order export, and walk through specific changes that cut address errors before the holiday peak.

Why parcels come back: three different problems, three different fixes

Before you start calculating, split returned parcels by reason. Each group has a different cause and a different fix:

1. Wrong or incomplete address – missing house or flat number, a typo in the street name, a postcode that belongs to another town. Your checkout form fixes this.
2. Recipient not at home – the address is right, but the courier finds nobody in and the customer never collects the parcel. A valid phone number and clear communication help here.
3. Uncollected parcel at a pickup point or locker – often a cash-on-delivery order the customer changed their mind about. That's a payment policy issue rather than an address issue.

How big is the problem? In an older but still widely quoted PCA Predict study (2018) of more than 300 retailers in the UK, US and Germany, roughly one in twenty parcels failed to arrive on the first attempt (Post & Parcel). And in its usability testing, Baymard Institute found that 9% of manually typed addresses contained a typo.

Worked example: everything you pay for one returned parcel

The figures below are an example for a smaller online store shipping to home addresses. Plug in the rates from your own carrier contract and your own labour costs.

1. Shipping to the customer: €3.90
2. Return shipping to your warehouse: €3.90
3. Box and filler you can't reuse: €0.80
4. Warehouse – receiving the return, checking the goods, restocking (10 minutes at €12/h): €2.00
5. Customer service – tracking the parcel, two emails, one phone call (15 minutes at €12/h): €3.00

Subtotal: €13.60 for a parcel that never arrived. If the customer still wants the goods and you ship again, add another delivery, box and pick-and-pack (€3.90 + €0.80 + €1.00) – and you're at €19.30.

On a €48 order with a 25% margin (€12), one failed delivery wipes out the entire profit on that order – and costs you another €1.60 on top. If the customer cancels in the meantime, you lose the revenue too.

Check the small surcharges in your carrier's price list as well – redelivery attempts, extended storage, return-to-sender fees. Individually they're tiny; over a year they add up.

From one parcel to a full year: a formula for your spreadsheet

One parcel barely hurts. Multiplication does. Use this simple formula:

Annual loss = orders per year × share of parcels undelivered due to the address × cost per undelivered parcel

Example: a store with 3,000 orders a month (36,000 a year), where 1.5% of parcels come back because of the address, at €13.60 each:

36,000 × 0.015 × €13.60 = €7,344 a year.

That's enough to pay for a seasonal warehouse helper over the whole holiday peak, or several months of paid ads. And we haven't counted the customers who never buy again after a bad experience.

How to get your own numbers in one afternoon

You don't need a BI tool – an order export and a spreadsheet will do:

1. Export orders from the last 6–12 months with a status such as "returned", "undelivered" or "returned to sender".
2. Add the return reason to each one from the carrier's tracking: insufficient address, addressee unknown, not at home, not collected.
3. Calculate the share of each group out of all shipped parcels.
4. Read through 20 random addresses from the "address" group. Patterns will jump out: missing house numbers, postcodes from another town, street names typed into the company field.
5. Plug the results into the formula with your own rates.

A common mistake: counting only the parcels the carrier physically returned. Parcels the courier eventually delivered after a phone call to sort out the address also cost you customer service time – they just don't show up in your returns report.

The most common address errors and how your form can catch them

Most errors happen in the last thirty seconds of the purchase – on a phone, in a hurry. A few "wrong vs. right" examples:

1. Wrong: "Station Road" with no number. Right: the form won't continue without a house number, or asks "No house number – is that correct?".
2. Wrong: a postcode from one city and the name of another. Right: the postcode is checked against the town and the mismatch is shown straight away.
3. Wrong: "Staton Rd 124b flat2" typed in one line. Right: the customer picks the address from a suggestion list and it's saved in a clean, consistent format.
4. Wrong: phone number "123". Right: the field checks the format and country code for the delivery country.

A hint right under the field helps too. A wording that works: "Please include your full house and flat number. It helps the courier find you faster."

The most effective fix, though, is not making customers type the address at all. Address autocomplete suggests verified addresses after a few characters and fills in street, number, town and postcode in one go. Baymard reports that 55% of e-commerce sites still don't offer this – even though it speeds up checkout and cuts typos.

Safety nets after the order is placed

Even with a good form, something will occasionally slip through. Add two cheap checks:

1. Address summary in the confirmation email. At the top, not in the footer: "We're shipping your order to: Jane Smith, 124B Station Road, Flat 2, [postcode, town]. If anything's wrong, just reply before 12:00 and we'll fix it before dispatch."
2. Manual review of risky orders before dispatch. Set up a filter for orders without a house number, with a postcode that doesn't match the town, or with a street name shorter than three characters. It takes a few minutes a day and saves several euros on every error you catch.

What to do next

1. Pull returned parcels from the last 6 months out of your export and group them by reason.
2. Calculate the cost of one undelivered parcel using the model above – with your own rates.
3. Multiply by your annual order count and your address-error rate. Show that number to whoever owns the checkout budget.
4. Add a house-number hint and a postcode check to the form, and an address summary to the confirmation email.
5. Set up a risky-address filter before dispatch – ahead of the holiday peak.

If you'd like addresses checked while customers type, have a look at Launify – it suggests and validates delivery addresses right in the checkout form.