When More Work Produces Less Pay
What 79 contributors revealed about the economics of parcel delivery
When I wrote Who Really Pays for Cheap Delivery?, I expected disagreement.
I did not expect 726 page views within its first few days, or the response that followed when I shared it in two Facebook groups used by Evri couriers.
Across the two discussions, I identified 79 contributors. Some simply said they agreed. A few offered the sort of abuse that social media provides free of charge. Around 50, however, offered substantive observations about parcel rates, classifications, workload, costs and the physical demands of the job.
Many had delivered for Hermes and Evri for 10, 20 or even 30 years.
Their comments were not a scientific survey. The contributors selected themselves, their individual claims have not been independently audited and dissatisfied former couriers may have been more likely to respond than satisfied ones.
But the comments did something important. They showed that my experience was not unusual enough to be dismissed as one former courier complaining about his old employer.
More importantly, they revealed why arguments about courier earnings are so difficult to resolve.
There is no single Evri courier rate. There is no single type of Evri courier. There is not even one obvious answer to the question: how much did this courier earn per parcel?
That lack of clarity benefits Evri.
A career in which experience can reduce your rate
One contributor described Evri as the only job he had known where his pay rate went down the longer he stayed.
He was not alone.
A courier with 24 years' service said he left earning less than he had 10 years earlier. Another, after 28 years, estimated that her rate had fallen by nearly 20p. One contributor said she was earning less when she left than when she started more than 20 years before. A courier who departed in early 2025 said he finished worse off than when he began 18 years earlier.
Another said her parcel rate was approximately 10p lower than when she started 25 years ago.
Inflation makes those accounts more striking. Receiving the same number of pence after two decades would already represent a substantial real-terms reduction. Receiving fewer pence means the decline begins before inflation is considered.
More recent accounts followed the same pattern. A courier working for a third-party contractor said her rate had fallen from 75p to 68p within three years while fuel, insurance and vehicle costs had all increased.
One contributor said her standard rate was reduced because she completed her established round quickly.
That is an unusual reward for accumulated competence. Couriers learn their streets, customers, parking places and safe locations. They become faster and more reliable. In most occupations, increased productivity provides an argument for higher pay.
Here, it can demonstrate that the rate has room to fall.
The small-packet racket
The most frequently identified turning point was not a change of ownership or a dramatic corporate announcement. It was a classification.
Couriers repeatedly mentioned the introduction or expansion of the small-packet rate. More deliveries could be placed in lower-paying bands, including items couriers believed would previously have attracted packet or standard payments. Contributors also described more items being treated as postable, even when they could not reasonably be put through a letterbox.
One courier said small packets had “killed” what had previously been acceptable earnings. Another said the change finished him after 13 years. A third estimated that packets had grown from 10 to 15 per cent of his load when he began to around 80 per cent by the time he left.
Another contributor said he once received 75p for a standard parcel, with standards comprising approximately 80 per cent of his load. He later received 56p for a standard while standards had fallen to around 10 per cent of the load. At the same time, he reported leaving the depot at 12.30pm rather than 8am, delivering approximately 100 additional items a day and earning less.
That is why quoting one rate from a courier rate card tells us very little.
For car couriers, deliveries can be paid across several internal bands, including postable, small packet, packet, standard and large or heavy. Each attracts a different payment. The mixture changes by client, round and day.
These internal courier bands must not be confused with Evri's public parcel categories, which describe what customers may send and what they are charged.
A courier may receive more than £2 for a particular large or heavy delivery while most of the van contains packets paying a fraction of that amount. Two couriers can complete the same number of deliveries and receive very different totals.
This is why couriers generally work out their earnings by dividing their basic delivery payments by the number of parcels delivered. The resulting blended average reveals what the load was actually worth.
Cheap delivery does not always require an announced pay cut. The payment for the work can also be reduced by changing how the work is classified.
There is no single type of courier
The workforce is as fragmented as the rate card.
In 2018, an employment tribunal found that the Hermes couriers who brought the case were workers rather than independent contractors. They were entitled to rights including the National Minimum Wage and holiday pay.
In 2019, Hermes and GMB introduced the optional Self-Employed Plus model. SE+ retained self-employed flexibility while providing guaranteed earnings and holiday pay. Later agreements added pension and parental-leave provisions.
Other couriers remained on Lifestyle arrangements. Current GMB information refers separately to SE+, “194s”, worker-status members and Lifestyle couriers. Alongside them are temporary couriers, subcontracted drivers and contractors undertaking Light & Large work.
Even Lifestyle is not one uniform arrangement. An established round holder may receive a lower regular rate than someone brought in when Evri needs urgent additional capacity. A specialist contractor carrying heavier freight may receive a much higher payment per delivery but require a different vehicle and complete fewer drops.
Lifestyle couriers may have less reason to join GMB when many of the protections it has negotiated apply to recognised groups rather than to their own arrangements. I have found no published breakdown of GMB membership by courier status. The clearest indication of its current reach came in 2026, when more than 6,000 Evri couriers voted on a pay offer and almost 97 per cent rejected it.
Six thousand is a substantial organised group. It is still only part of a network that Evri says contains more than 30,000 couriers.
Fragmentation makes agreement difficult. Couriers discussing “the rate” may be accurately describing entirely different arrangements. One may quote a blended average of 40p, another 65p and another the rate of more than £2 attached to a heavy item.
They may also have different holiday rights, pension arrangements, minimum-payment protections and route obligations.
All three can be telling the truth without describing the same job.
What my invoices actually show
My own invoices demonstrate how several apparently contradictory descriptions of courier pay can be produced from the same document.
In February 2022, the summary stated that my “average earnings per item” were 88p.
That was technically correct. It was not my underlying parcel rate.
The basic payments attached to 2,075 deliveries totalled £1,081.66, producing a blended delivery rate of approximately 52.1p. The much higher front-page figure included collections, supplements, corrections, five days of holiday pay and other adjustments.
The invoice total described everything paid during the period. It did not describe the average payment for delivering a parcel.
The same distinction appeared in March 2023. My 2,754 deliveries generated £1,458.51 in basic delivery payments, an average of approximately 53p. The invoice showed gross earnings of £1,929.33 because it also included collections, a GMB payment, misband corrections, service payments and other adjustments. After pension deductions, £1,825.30 reached my bank.
The basic delivery pay, gross earnings and bank payment were all accurate. None meant the same thing.
The final comparison came in 2025.
In January, while working under SE+, I completed 2,021 deliveries. The basic delivery payment was £1,134.39, producing a blended average of 56.1p per parcel.
February contained no deliveries at all. I nevertheless received £90.28 in delayed SE+ adjustments relating to previous work.
The remaining SE+ rounds on my March invoice contained 937 deliveries. They generated £531.03 in basic delivery payments, an average of 56.67p per parcel.
My temporary Lifestyle work appeared across the March and April invoices. Those 818 deliveries generated £412.66, producing a blended average of 50.45p per parcel.
The blended average for the Lifestyle work was therefore 6.22p per parcel lower than for my remaining SE+ work, a difference of approximately 11 per cent.
The April part of the round also introduced small packets at 44p. Fifty-seven small packets generated £25.08. Had they attracted the 48p packet rate printed on the same invoice, they would have generated £27.36.
This comparison concerns one round. It does not establish the rate earned by every Lifestyle courier. It reflects that particular route, load and parcel mix.
But it was the rate I had actually received.
When I calculated how little I had earned for each parcel, I decided I would never go back.
The invoice never clearly told me that SE+ payments and Lifestyle deliveries were represented on the same document. I reconstructed the change by matching round numbers and calculating the underlying rates myself.
Evri's invoices are detailed enough to look transparent, yet fragmented enough to prevent the courier's actual delivery rate from being immediately understood.
Volume is not a pay rise
The apparent answer to lower unit rates is greater volume.
Several contributors said they had been told by local management that more parcels allowed them to earn more. Mathematically, that can be true. Economically and physically, it is less convincing.
One courier reported three pay reductions in nine years. She was delivering three times as many parcels for approximately the same income while every significant expense had increased.
Another said she compared her invoices and found that she was paid £5 less despite delivering 500 additional parcels.
Volume can improve earnings when additional deliveries are densely concentrated and properly priced. One contributor made that counterargument directly. His average Evri rate was lower than rates available elsewhere, but he said the Evri round produced more profit per hour because it was efficient.
Another claimed to clear more than £2,500 a month after expenses while working three hours a day, six days a week.
The claim deserves to be examined rather than dismissed.
Six three-hour days produce 18 delivery hours a week, or an average of 78 hours a month. Dividing £2,500 by 78 gives an hourly return of £32.05 after expenses.
If I add another 90 minutes each day for sorting and loading, which the contributor did not include in his claim, the calculation becomes 117 working hours a month and £21.37 an hour after expenses.
At an assumed blended payment of 65p, generating £2,500 before expenses would require 3,846 deliveries a month, approximately 148 a day. Completing them in three hours requires just over 49 parcels an hour, allowing about 73 seconds per address.
But the contributor said £2,500 after expenses. If monthly expenses were £500, the required delivery rate would rise to approximately 59 parcels an hour. At £750 it would be approximately 64. At £1,000 it would be approximately 69.
Those rates allow between 52 and 61 seconds per address. That time must include driving, parking, finding the parcel, walking to the property, scanning or photographing it, knocking, waiting and completing any safe-place procedure.
My own highest sustained rate was approximately 40 parcels an hour, achieved when most deliveries could be left in porches without a prolonged wait. Because the load consisted heavily of lower-rated items, my average payment was only around 40p. Forty deliveries generated approximately £16 before fuel, insurance, vehicle wear, sorting time and tax.
Speed had not made the work especially lucrative because the parcel classifications reduced the value of each delivery.
None of this proves that the contributor misstated his earnings. His route may be exceptionally dense, his parcel mix may attract higher rates or his description of three hours may omit part of the working day.
It does show why an exceptional monthly figure cannot represent the whole network without further information.
It also raises a question about compliance. More than 50 deliveries an hour may be possible for short periods on a particularly dense round. It may also be achieved by leaving parcels and moving on without waiting for the customer, or by taking other shortcuts around the delivery process. I am not suggesting that this contributor does so.
The point is that parcels per hour tells us nothing about route density, customer contact, safe-place compliance or delivery quality.
When payment is attached to each completed delivery, observing every part of the process takes time but attracts no additional payment. The courier who waits, checks and complies may earn less per hour than one who records the parcel as delivered and immediately moves on.
A system built around volume should not be surprised when volume begins to compete with service.
The human limit
Volume also assumes that a human body is infinitely scalable.
One courier wrote that she could deliver 250 parcels a day at 38 but could not do so at 57. Another, now 67, said that after one round was removed she sometimes received only around 70 parcels, too few to make continuing worthwhile.
Others remained because the work fitted around family responsibilities, they were approaching retirement or alternatives were limited. One said she had stayed for more than 20 years and was “holding on like a thread.” Another described a daily grind sustained by the false hope that something might change.
Several contributors also described what happened when couriers were injured or unwell. One said that after a road collision, the first concern expressed was whether the parcels could still be delivered. A driver from another company described calling the depot while awaiting an ambulance and being asked how many stops remained.
These are individual accounts and should be treated as such. But they sit alongside repeated descriptions of couriers feeling worthless, ignored or known only as delivery capacity.
One courier with 30 years' service said parcel volumes kept rising, depot departures became progressively later and senior management would not have known who she was.
Another described a combined loss of around £1,200 a month for herself and her husband before she retired and he found different work. After a combined 54 years delivering, she said they were better off after leaving.
Long-serving couriers possess the local knowledge and reliability that the network needs. Yet maintaining their earnings increasingly depends on a volume of physical work that becomes harder to sustain as they age.
The system benefits from their experience while gradually making it more difficult for them to remain.
What the response establishes
The comments do not establish the average earnings of an Evri courier. They do not prove that every rate has fallen in every depot, or that no courier can construct a profitable round. They cannot by themselves isolate the effect of private-equity ownership from the wider price competition affecting parcel delivery.
What they do establish is that my experience was shared by enough long-serving couriers to deserve serious examination.
Across two groups, contributors independently described the same mechanisms:
lower unit rates;
increasing use of packet, small-packet and postable classifications;
disappearing bonuses and premium payments;
greater volume masking weaker payment per item;
rising costs transferred to self-employed couriers;
later depot departures extending the working day;
experienced couriers leaving because the arithmetic no longer worked.
Evri can make bold claims about courier earnings because almost any chosen figure may be true for somebody within such a fragmented network.
It can quote an hourly average based on delivery time only, without making the treatment of sorting time and expenses immediately obvious. It can point to union-backed holiday and pension protections that do not apply equally to every courier. It can cite higher individual parcel rates without revealing the mixture of lower-rated items in the average vehicle.
Evri does not need to invent a number. It merely needs to select which couriers, parcels, hours and protections are allowed to represent the whole network.
The original article asked who really pays for cheap delivery.
The couriers supplied the answer.
Customers see an inexpensive service. Retailers see a distribution network. Investors see scale, efficiency and value.
Couriers see another packet added at a lower rate.
That is where part of the saving is made.
If you enjoyed this article, you may also find this one interesting:
It looks beyond Evri to examine the wider financial model behind many corporate acquisitions.
A leveraged buyout can generate substantial returns for investors while leaving the acquired company carrying the debt. Employees, suppliers and customers may then bear risks that do not appear in the headline account of success.
When is a leveraged buyout really a success?
https://www.thegoodmethod.co.uk/good-thoughts/when-is-a-leveraged-buyout-really-a-success
About the author
Paul Clark is the author of The GOOD Book: A Behavioural Operating System for Escaping Debt and Rebuilding Control and the creator of The GOOD Method. If you enjoyed this article, explore more of his writing in GOOD Thoughts, listen to The GOOD Conversation podcast, or join the GOOD Community to receive new articles, podcast episodes, resources and forthcoming videos by email.