Algeria AgencyIndustries
Logistics: you are graded by somebody who is not your customer
The merchant signs the contract. The recipient, who bought nothing from you, writes the review and refuses the parcel.
A carrier has a peculiarity few trades share: the person who decides its reputation is not the person paying it. The merchant signs the contract; the recipient, who has no commercial relationship with you at all, writes the review.
That recipient did not choose you, does not know your rates, is unaware of your constraints, and has no reason to give you the benefit of the doubt. They judge you on a single interaction, at their door, on a day when they were expecting something else.
Nearly everything that goes wrong in this sector follows from that asymmetry. Everything is optimised for the customer who pays — the rate, the invoicing, the volume — and the one who grades you is neglected, even though they decide whether the merchant keeps you.
This article is written from your side of the problem we described elsewhere from the merchant’s. It starts with the judge who is not your customer, and it ends with what we refuse to write on your behalf.
You are graded by somebody who is not your customer
Your commercial customer is the merchant. They compare rates, delivery times and remittance terms, and they keep you or leave you on numbers.
But they do not build those numbers alone: they build them out of what their own customers tell them about you. A merchant receiving three complaints a week about your courier changes carrier before their costs have moved at all.
The consequence is that your most effective commercial action is not addressed to your customer. It is addressed to the recipient: being reachable, announcing a window, warning of a delay, presenting yourself properly at a door.
That is spending which does not show up in a rate card and which decides whether a contract is renewed. It is also why the carriers gaining market share are not always the cheapest.
One last, more strategic consequence: today’s recipient is tomorrow’s merchant. Many online sellers choose their carrier by remembering how they were delivered to personally.
The promise about time is your product
What you sell is not a journey, it is a promise about a duration. And that promise rests on a chain several of whose links are not yours: handover of the parcel, the address supplied, the recipient’s availability, the road.
From which follows a rule few carriers apply: announce a time you keep nine times out of ten, not one you keep half the time under the best conditions.
The arithmetic is easy to do and painful to accept. A short time announced and missed produces a call, a complaint, sometimes a refusal; a longer time announced and kept produces nothing at all, and "nothing at all" is the best possible outcome in this trade.
You also have to distinguish transit time from total time. A parcel handed over at four in the afternoon does not leave the same day, and the recipient counts from their order, not from your collection. Writing that distinction down prevents a misunderstanding that is invariably blamed on you.
Finally, announce a time by zone rather than a national one. A single figure for the whole country is wrong everywhere: too long where you are fast, too short where you are not, and it loses you customers at both ends.
The address: the failure point nobody budgets for
The commonest cause of a failed delivery is neither the road nor the courier: it is an address that does not let anybody find a door.
That is not carelessness on the recipients’ part. A large share of real addresses is described by landmarks rather than by a street and a number, and a form designed elsewhere demanding "street, number, postcode" mechanically produces unusable addresses.
The fix is within your reach and not the merchant’s: give them an address format that matches the country. Province, municipality, neighbourhood, landmark, and a free field for the directions people give naturally.
The second fix is the phone number, which is not secondary data but the real address. A parcel with an approximate landmark and a number that answers arrives; a parcel with a perfect address and a phone switched off does not go out.
The third is what the courier does when they cannot find it. With no written rule everybody improvises: some call three times, others leave immediately. A uniform rule — two calls, a message, a re-attempt window — removes a large share of failures without a single extra kilometre.
Tracking: what the recipient actually wants to know
Most tracking systems answer a question nobody asks: where the parcel is. The recipient does not want to know where it is, they want to know when it will arrive and whether they have to stay in.
A status of "in processing" is therefore empty information, and it paradoxically produces more calls than no tracking at all, because it suggests information exists and is being withheld.
Useful information comes down to two elements: the expected delivery day, and a window on the day itself. The second is what separates a service people appreciate from one they endure. That still assumes you know which of your parcels is in trouble, and the silent parcel is the one tracking does not show.
The third element, rarer still, is announcing the delay. A message saying "tomorrow instead of today" before the recipient has noticed turns an incident into information, and it is the same rule as in every other trade in this series.
A practical note: those three elements beat a detailed ten-stage trace. Granularity impresses the merchant who buys and does not interest the recipient who judges, and this chapter is about the latter.
Proof of delivery, and what it has to contain
Proof of delivery is the document that settles disputes, and in many networks it is reduced to a status ticked in an application.
A status is not proof. What is: the exact time, the name of the person who received it, and their relationship if they are not the recipient — a neighbour, a caretaker, a family member.
That last point is the most often neglected and the most often at issue. A parcel handed to a third party without their name recorded is a parcel the recipient will declare undelivered, and you will have nothing to set against it.
Photographing the delivered parcel has become common practice and it is useful, provided it is bounded: it documents an object and a place, not a person, and it has no business showing an interior or a face.
Finally, that proof has to be available to the merchant without them ringing you. A carrier who answers "who signed" in three minutes saves an hour of conversation a week and earns a reputation for seriousness worth more than the minute spent organising it.
Where people look for you, and on which screen
You have two audiences online and they are not looking for the same thing. The merchant looks for your rates, your coverage and your remittance terms. The recipient looks for a number to find out where their parcel is.
The internet market observatory published by ARPCE counts, for the second quarter of 2025, some 59.10 million internet subscriptions in Algeria, of which 88.71% are mobile and 11.29% fixed.
The consequence for the second audience is direct: tracking has to work on a phone, with no account to create, from a number the recipient received by message. Every extra step turns into an incoming call.
The consequence for the first is different and often ignored: a merchant compares rate cards, and a legible, current rate card published rather than sent on request wins you contracts without a single meeting.
And a consequence common to both: your telephone line has to be reachable at the hours people are waiting for parcels, which is late in the day. A switchboard closing at four in a trade that delivers until six is a contradiction your customers notice before you do.
- Mobile subscriptions88.71%
- Fixed subscriptions11.29%
ARPCE, internet market observatory, second quarter of 2025
Coverage: what to write zone by zone
Announcing national coverage is the sector’s reflex and it is nearly always an expensive simplification. You do not serve every area the same way, and pretending otherwise makes you accept parcels you will deliver badly.
The honest wording is a three-level map: areas you deliver yourself, areas you cover through a partner, and areas you do not serve. All three exist at every carrier and nobody writes them down.
That transparency has an unexpected commercial effect: it reassures the merchant rather than worrying them. A carrier who says what it does not do is believed about what it says it does, and that is exactly what a professional buyer is trying to assess.
Add the delivery time per level, and above all the pickup point where one exists. Collection from a pickup point solves half the address problems described above, and it is often the best option for areas you serve badly.
We will give you no coverage or average-time figures for the sector in Algeria: we know of no dated, checkable source, and the rates circulating in the trade are sales arguments. Your own records are better, and the measurement section says how to take them.
A return is a service, not an incident
In a market dominated by cash on delivery, a return is not an anomaly: it is a normal part of the flow, and a carrier treating it as an incident handles it badly.
The first point is the return time. A refused parcel taking three weeks to come back ties up the merchant’s goods for three weeks, and that cost is perfectly visible to them even when you do not invoice it.
The second is condition. A parcel coming back damaged after travelling twice is a clean loss for the merchant, and it is the kind of case that ends a commercial relationship without discussion.
The third is the reason. A return with no reason recorded deprives your customer of the one piece of information that would let them reduce their refusals — and a carrier who supplies it becomes a partner rather than a supplier.
That last point deserves to be seen as a commercial argument rather than a burden. The reason for refusal is data only you can collect, it costs you a tick box, and it is worth far more to your customer than the discount they will ask for instead.
What can be measured in a network
Four numbers are enough and they are kept by zone, never globally. The first is the first-attempt delivery rate, which is the trade’s central indicator.
The second is the real time between collection and delivery, as a median rather than an average. The average is moved by a few very slow parcels and hides exactly the problem that needs seeing.
The third is the breakdown of failures by cause: unreachable, address not found, absent, refused, postponement requested. That list tells you whether your problem is your network, your address format or your couriers.
The fourth is how long refused parcels take to come back, the least tracked number in the sector and one of the most watched by your merchant customers.
Those four measures share a property: they degrade silently. A network deteriorating produces no alert, it produces a merchant who does not renew, and you find out once the volume has already fallen.
Collecting on delivery and remitting
In this market a carrier does not only deliver parcels: it collects money on somebody else’s behalf. That is a financial activity grafted onto a transport one, and it is judged with the severity money deserves.
Remittance time is therefore a first-rank commercial argument, often more decisive than the rate. A merchant waiting three weeks for their money is financing your cash position out of theirs, and they know it.
What needs writing down fits in three lines: how often you remit, within what time after delivery, and what deductions apply. The third line is the one missing everywhere and the one that produces disputes.
The direction of the market is dated: over one year the number of online merchants grew 26%, card transactions on the internet 38%, and the amount paid online 179%, while the card base grew only 9%.
Read it as a trend that concerns you directly: every prepaid order is a parcel you do not collect for, do not remit, and which is far less likely to be refused at the door. That is a shared interest with your customers, and few carriers put it that way.
GIE Monétique, 2025 annual balance sheet
What to check before signing with a supplier
The first question sorts everybody: ask what they would do for the recipient, who is not your customer. If they talk only about your image with merchants, they are working on half the problem.
The second is about tracking: will they accept it working with no account creation, on a phone, from a number received by message? Every added step becomes an incoming call you will pay for in time.
The third is about what gets published: your rate card and your three-level coverage map. A supplier who prefers "request a quote" loses you the merchants who compare, which is all of them.
The fourth is about data: a tracking system handles names, addresses and numbers of people who are not your customers. Ask where it is hosted, what is retained, and for how long.
The fifth is a test: ask them to write your standard delay message. If they produce an apology they have not understood; the right message states the new date, does not justify itself, and offers an action.
What we do, and what we will refuse to do
What we will refuse: announcing uniform national coverage when it is not uniform. That would make you accept parcels you will deliver badly, and the bad reviews that followed would cost more than the contracts won.
We will also refuse to write a delivery time your own records do not support. In this trade a stated time is a contractual promise disguised as a sales argument, and you will be carrying it, not us.
A limit of competence, stated plainly: we are not operators and we will not design your rounds. The first-attempt delivery rate is corrected by organisation and by address format, and we will not pretend to correct it with a page.
What we do: a published, legible rate card; a three-level coverage map; tracking that answers the recipient’s only two questions — when, and do I have to stay in; a standard delay message; and the three remittance lines, deductions included.
And what you should do without us this month: record your first-attempt delivery rate by zone, and the breakdown of your failures by cause. Two columns, four weeks. It is free, and it will tell you whether your problem is your network, your address format or your couriers — three problems, only one of which is solved with us.
Frequently asked questions
Should we announce a short delivery time to win contracts?
Announce the one you keep nine times out of ten. A missed time produces a call, a complaint and sometimes a refusal; a time kept produces nothing at all, and "nothing at all" is the best outcome available in this trade.
Why do so many parcels fail to arrive?
Most often because of an address that does not let anybody find a door. Give the merchant a format that matches the country — province, municipality, neighbourhood, landmark, free field — and treat the phone number as the real address.
What should our tracking contain?
Two pieces of information: the expected day and a window on the day itself. A status of "in processing" is empty information that produces more calls than no tracking at all.
What should proof of delivery contain?
The exact time, the name of the person who received it, and their relationship if they are not the recipient. A ticked status is not proof, and a parcel handed to an unnamed third party will be declared undelivered.
Should we publish our real coverage?
Yes, at three levels: delivered by you, by a partner, and not served. A carrier who says what it does not do is believed about what it does, which is exactly what a professional buyer is looking for.
Does remittance time really matter?
Often more than the rate. A merchant waiting three weeks for their money is financing your cash position out of theirs. Write the frequency, the time after delivery and the deductions — that third line is the one missing everywhere.
Where we come in
Four weeks of records per zone — first-attempt delivery, real lead time — give you what your customers will check anyway.
- We publish a readable rate grid rather than “price on application”.
- We distinguish three geographic service levels instead of one uniform claim.
- We refuse to write a commitment your records do not support.
We are not operators: rounds, vehicles and dock organisation will not come from us, and we will not advise on them.
Read next
AI in logistics: the list of abnormal parcels, before the customer calls
A network learns about its incidents by telephone. The parcel in trouble is the one that produced no event, and that is exactly what tracking cannot see.The round: building a driver’s day, and what happens when an address fails
You are paid per parcel and you spend per attempt. A driver’s whole day is decided in that gap.The pilot that will never ship: the signs, from week two
A pilot that fails rarely fails at the end. It fails in the second week, quietly, and everybody carries on for three months.
Let us talk about your project
A free audit, no commitment: we look at your online presence and tell you what is holding it back.