In short
- Poland had 67,060 self-service postal machines at the end of 2025, according to UKE, which reported 29.7% annual growth in the device count. That is a stock of infrastructure, not a measurement of free compartments, bookable services or parcels delivered that day.
- UKE reports nearly 1.40 billion courier items in 2025, up 14.3%. Courier items include more than locker deliveries, so dividing them by all machines does not give locker utilisation.
- For a webshop, check the exact network, postcode and service on both the dispatch and receipt side before using a national locker count in a fulfilment plan.
- AirPost is a peer-to-peer fulfilment network: online shops have their orders packed and dropped at a parcel point by verified people near their customers, who earn a fee per parcel from home. No warehouse, no minimum order volume.
Poland’s locker numbers make an attractive expansion slide. A dense network seems to promise easy dispatch and convenient collection everywhere. But a machine count does not tell you whether the locker beside your packing partner accepts your label, whether the customer’s selected compartment is available or whether your product fits. The useful question is how to turn official infrastructure and parcel-flow statistics into a test that your shop can actually run.
What did the Polish regulator actually count in 2025?
The UKE report on the state of the postal market in 2025, published on 29 May 2026, reports 67,060 self-service postal kiosks at year-end. It separately reports nearly 1.40 billion courier items and more than 37 courier items per person. The machine count grew by a reported 29.7%, while courier volume grew by 14.3%. These are different indicators with different units.
Keep the measurement date attached to the infrastructure count. End-of-year stock includes devices added during that year, so those devices were not all available for the entire annual parcel flow. Keep courier items separate from all postal services too. A broad postal total can include letters and other services that do not represent locker demand. Choosing a convenient larger number would make your apparent throughput less meaningful.
| Official observation | What it describes | What it cannot establish |
|---|---|---|
| 67,060 machines, end of 2025 | National stock of self-service postal devices | Free compartments at your customer’s location |
| 29.7% reported machine growth | Change in infrastructure stock | Increase in usable capacity for your label |
| Nearly 1.40 billion courier items | Annual courier activity across delivery modes | Number of parcels collected from lockers |
| 14.3% courier-volume growth | Annual growth of that service segment | Growth of your shop’s Polish orders |
Are machines, compartments and pickup points interchangeable?
No. A machine contains multiple compartments, while a staffed pickup point is another type of location. UKE’s detailed commentary on the 2024 courier market reported 51,811 machines containing 5,145,436 compartments at the end of that year. Dividing those two observations gives about 99.3 compartments per machine. That describes the national average configuration, not the dimensions or free space of any individual machine.
A compartment can only accept an appropriate parcel while it is available. Counting doors therefore does not establish how many large boxes the network can accommodate. A pickup-point total also cannot be added to a compartment total: the units are different. When a presentation combines them into one network number, ask whether the total means locations, individual storage spaces or service access points.
The older detailed machine figure and the newer report’s percentage are also separate published data vintages. Do not force a growth rate by joining them without checking the newer report’s baseline. For a current trend statement, use the rate reported in the current annual publication. For the compartment calculation above, keep both inputs inside the older detailed observation. This avoids presenting a difference between releases as a new operational finding.
Why does parcel volume per locker fail as a utilisation measure?
The numerator must describe the traffic that actually uses the denominator. All courier items are too broad for locker utilisation, because some are delivered elsewhere. The detailed 2024 commentary identifies 626.8 million courier parcels received through machines. That is a more relevant locker-flow observation, but dividing it by an end-year machine or compartment stock still does not reveal how full lockers were during busy hours.
To measure occupancy, you would need information about occupied space and available space over the same period, with dimensions and dwell time taken into account. An annual count does not provide that. A compartment reused quickly can handle several parcels over time; a compartment occupied for longer can handle fewer. Neither behaviour can be reconstructed from the year-end number of machines alone.
You can still make a carefully limited comparison of growth rates. Within the 2025 overview, courier volume grew more slowly than machine stock. That observation does not prove falling locker occupancy or an oversupplied network. The courier numerator is broader, devices have different capacities and locations, and the timing of new installations matters. Treat the comparison as a reason to investigate, rather than as a calculated measure of spare room.
Does a large receiving network make dispatch equally convenient?
Check both ends. A customer may collect from a machine even when the merchant handed the parcel to a courier or a staffed point. The receiving channel does not identify the dispatch channel. UKE’s 2024 courier report overview also explains its survey coverage: the participating operators represented 99.7% of courier volume. That is useful coverage information, but it still does not turn one channel’s observation into another channel’s availability.
For your packing partner, the practical sequence is label creation, preparation, handover and evidence that the parcel entered the carrier network. For your customer, it is selecting a compatible destination, receiving the notification and collecting the parcel. Test both sequences with the actual service. A nearby machine that serves another network does not solve a handover problem, and a carrier logo alone does not identify every product that can use it.
This distinction matters when you delegate preparation. AirPost connects your shop with people who hold a small batch of stock, pack orders and take the parcels to a parcel point. The carrier performs the delivery after handover. A person living near a machine has not thereby acquired access to every locker service. Agree a real drop-off routine using a service you can book, and confirm the parcel format before moving stock.
How should a foreign webshop test one Polish postcode?
Choose one parcel profile
Select actual products and measure the packed dimensions and weight. Keep multi-box orders outside the first comparison if they would blur the result.
Verify the dispatch option
Check which location accepts the actual label and what evidence of handover is available. Agree who performs that trip.
Verify the receiving option
Use the selected carrier’s current location selector and service conditions. Do not substitute a national network count for address-level availability.
Record exceptions
Log redirection, failed handover, customer collection issues and any replacement shipment separately. Do not invent a national failure rate from a small trial.
Review before adding stock
Compare the measured preparation cost and routine with your current arrangement. Scale the test only when the actual process supports it.
Use the existing Polish pickup-point and home-delivery comparison when you need to compare service types. For carrier differences, see InPost versus DPD in Poland. Those questions are distinct from this infrastructure analysis. Check the current booking before relying on a previously observed tariff or network description.
What cost should you attach to the trial rather than the national count?
Budget the preparation you actually expect to use. A national network does not set your carrier label price or your packing partner’s fee. Keep the latter separate from any Polish-zloty carrier quote, and avoid silently treating a euro figure as a zloty amount. The following is a euro planning component based on AirPost constants, not a Polish carrier tariff or a foreign-exchange conversion.
For an illustrative 30-parcel test, the platform component is €0.85 per shipment and the recommended small-parcel preparation fee is €2.25. Checked on 19 October 2026, their sum is €3.10 per parcel, or €93.00 for the test. The sender sets their own fee. Add the live label amount in its displayed currency, materials, stock transfer and other applicable costs before comparing a complete landed arrangement.
Ask for an ID check before handing stock to a person and agree what work the fee covers. The webshop fulfilment page explains the small-batch model. Your decision should follow the trial’s actual cost and handover evidence. More national machines can make a market worth investigating; they do not supply the missing terms of your individual arrangement.
Frequently asked questions
How big is the Polish parcel locker market?
UKE counted 67,060 self-service postal devices at the end of 2025. Describe that as an infrastructure count; it is not a measure of all courier revenue or available locker space.
Does every courier parcel in Poland go to a locker?
No. Total courier volume includes other delivery channels. The separate detailed 2024 observation identifies parcels received through machines, which is a different measure.
Can I calculate locker occupancy from parcels per machine?
Not from annual volumes and a year-end machine count alone. You also need matching coverage, compartment availability over time, dimensions and dwell time.
Are pickup points included in a machine total?
Do not assume so. A staffed point, a machine and a compartment are different units. Read the definition of each published network count before combining figures.
Does a locker near my packing partner prove I can use it?
No. Verify that the actual network, label product, parcel size and handover process are compatible. The same check applies to the customer’s selected collection location.
Sources
Test a specific packing and handover arrangement with a small stock batch before increasing your Polish inventory.
Explore a local packing test