Algeria AgencyDigital marketing
LinkedIn ads: you are not buying an audience, you are buying names
Here the public that could buy from you can be counted. And what can be counted can be contacted, usually without paying.
Like the article on Snapchat, this one exists first to decide whether to go there. The difference is the grounds for refusing: there it was the age of your buyers, here it is the size of the list of people who could buy something from you.
On the five consumer platforms in this series you buy attention in bulk and hope buyers are somewhere inside it. Here, for an Algerian business, the population able to decide to buy from you is often small enough to be written on one sheet of paper.
That turns the whole question around. You are no longer asking what a click costs: you are asking whether you can name forty people. And if you can, the rest of this article will explain why advertising is rarely the right instrument for reaching them.
There is, on the other hand, one use of this platform that works solidly for a great many Algerian businesses, and it is not selling. We give it a whole section, because an article omitting it would be dishonest by omission.
A scale with nothing in common with the rest of this series
Start by putting the orders of magnitude side by side, because that explains everything after it.
According to figures published by DataReportal for Algeria, Facebook has 25.6 million users, TikTok 21.1 million and Instagram 12 million.
Those three numbers describe consumer platforms, and they are dated and checkable. The professional population reachable here is several orders of magnitude below any of those bars, and we will quote no figure for it because none exists that is dated and checkable.
The gap in scale is the subject, not the missing measurement. On the platforms above, your potential audience exceeds anything your budget could ever touch. Here your potential audience is probably smaller than what your budget could touch — and that is an entirely different situation.
When an audience is smaller than the budget, advertising stops being a way of going out to find people and becomes a very expensive way of repeating a message to people you could name. The next section draws the consequence.
DataReportal, Digital 2026 Algeria — October 2025 figures
The question that settles it: can you name them?
Take a sheet of paper and write down the names of the Algerian businesses that could buy what you sell, this year. Not the sectors: the businesses, by name.
Three possible outcomes, and each leads somewhere different. If you write fewer than twenty, you do not have a marketing problem but a sales problem: those twenty are handled one at a time, by a human being, and no platform will improve on that.
If you write between twenty and two hundred, that is the commonest case for a service business in Algeria. It is also the case where this platform can help — but not in the way it will be sold to you, and the next section explains why.
If you write several thousand, you are probably selling to very small businesses or to sole traders, and you are in fact in a disguised consumer market. The five previous articles in this series concern you more than this one.
The exercise takes an hour and it decides the spend. A supplier proposing a monthly budget without having made you do it is selling you a channel, not an answer to your situation.
If you can name them, advertising is rarely the right instrument
Here is the sentence that costs us the most contracts on this page: when the list fits on one sheet, the correct instrument is not an advertising budget, it is a series of messages written by hand.
The reason is arithmetic. A campaign paid to reach two hundred specific people costs a great deal per person, arrives in a format those people recognise as advertising, and does not tell you which of them read it.
A connection request followed by a short message, sent to forty people over two weeks, costs nothing, arrives as a conversation, and gives you either a reply or a silence — two useful pieces of information, where an impression is not one.
What that assumes has to be said too: time, patience, and a message that does not read as automated. It is sales work rather than communications work, and that is why an agency would rather sell you the other thing.
Advertising becomes relevant again in one precise case: when you want your name to be already familiar at the moment you write. That is preparation rather than acquisition, and it is not judged on clicks.
The most expensive click in the pillar, and what it buys
This platform is markedly more expensive per click than the other five, everywhere in the world, and it is worth understanding why before deciding that is outrageous or justified.
You are not paying for attention: you are paying for declared targeting. People have written their own job title, employer and sector, which makes it possible to aim at "purchasing manager in manufacturing" with a precision no other platform offers.
That targeting is worth its price only if the role you are aiming at is the one that decides at your customers. If your real buyer is the owner of a twelve-person business who never updates a profile, you are paying a lot for a precision that corresponds to nothing.
We will quote no cost per click for Algeria: no dated, checkable series exists, and the foreign benchmarks we know describe markets whose advertiser structure has nothing in common with this one. The article on advertising budgets explains that rule at length.
The practical consequence is to set a ceiling before starting, and to set it low. A test on this platform is judged in weeks and on a very small number of conversations obtained, never on volume.
What works: a person writing, not a page posting
On this platform, content published by a person travels far better than the same content published by a company page. That is not an aesthetic preference, it is a fact of distribution.
The consequence is uncomfortable for a business used to speaking with one voice: the person who has to write is the director, the engineer, the project manager — somebody with a trade and a name, not a communications department.
What works in that register is equally particular: a problem encountered and how it was solved, a figure from your own operations, a reasoned disagreement with a common practice in your sector. What does not work is the announcement, the congratulation and the press release.
The tempo is modest and that is what makes it sustainable: one post a week, written in twenty minutes by somebody who knows what they are talking about, beats an editorial calendar nobody fills in after six weeks.
And the honest check before starting: will somebody in your business agree to write under their own name, every week, for six months? If the answer is no, that part of the platform will not open, and it is better to know before budgeting for it.
The company page distributes nothing, but it is checked
The previous section says a page’s content travels badly. It should not be concluded that the page is useless, because it serves one precise purpose and serves it alone.
When you write to somebody, that person makes two moves in order: they look at your profile, then they open your company page. The second move produces no statistic and nonetheless decides whether your message gets a reply.
What they are looking for comes to three things: that the business genuinely exists, what exactly it does, and how many people work there. A page with no description, no address and no associated employees answers all three badly, and the message goes out carrying that handicap.
It is the same mechanism as the Instagram grid and the channel on the video platform: a place that brings nobody but decides what happens to the people everything else brings. We meet it on four platforms in this series, and it is neglected every time.
The correction takes an hour and does not recur: a description stating the trade in one sentence, the address, the headcount, and a request to your colleagues to attach their profiles to the page. Do it before writing to anybody.
Language, and what the choice rules out
This is a decision taken once that closes a door, so it deserves a section rather than a line.
French dominates written professional exchange on this platform in Algeria, and it is the default choice for addressing a national market. Arabic is markedly less used in this particular register, which is a fact of usage rather than a judgement.
English is the right choice in one case only: when your customer is not in Algeria. For a business aiming at export, subcontracting or foreign clients, it is the only language that works, and it is also one of the few uses of this platform we recommend without reservation.
The trap is mixing. A profile alternating between the three languages from one post to the next is read by fewer people each time, because distribution rests on the reaction of the first readers and those readers change with every change of language.
The rule is therefore one language per intended audience, held to. If you are aiming at two markets, that requires two accounts or two acknowledged series of posts, and not one account that hesitates.
Hiring is the most robust use of this platform here
We sell advertising, and this is the most useful section in the article: for a great many Algerian businesses, what actually works on this platform is not selling but recruiting.
The reason is simple and can be verified in an evening: the profiles kept up to date here belong mostly to people who are looking for, or would accept, another position. That is what makes declared targeting reliable for a recruiter and fragile for a salesperson.
A business struggling to recruit a technical, accounting or sales profile will find here a pool that neither job boards nor word of mouth provides, at a cost that stays modest compared with a post left vacant for six months.
It also has an indirect effect on sales, and it is the only reputation mechanism we have seen work reliably here: a business whose employees talk about their work under their own names becomes known to its future customers through the same channel that makes it visible to its future employees.
If you came looking only for a sales channel, this section is the one that may make you keep the platform for another reason — and that is a better outcome than a budget spent in the wrong place.
What does not work, and is practised everywhere
There is a widespread method on this platform, taught on training courses, and it produces a negative result: the connection request followed immediately by a sales pitch.
The mechanism is easy to understand from the receiving end. They accepted a connection, which is a courtesy with no commitment, and within the minute they receive a text selling them something. The penalty is not a refusal, it is a permanent filing.
The automated version is worse, and it is recognisable: a message opening with your first name, mentioning your company, and continuing into an offer with no relation to what you do. In a market where everybody knows everybody, that practice costs a reputation for a return close to zero.
What works instead takes longer and returns more: a connection request with no message, then nothing for two weeks, then a short message about one specific thing concerning their business rather than yours.
We refuse to set up automated message sequences on this platform. It is a service requested regularly, it bills well, and it does lasting damage to the name it is sent under.
What can be measured, and over what period
The platform’s figures are even less useful here than elsewhere, because the volumes are small: a rate calculated on two hundred impressions says nothing statistically sound and will nonetheless be presented as a result.
Three things can usefully be counted, and they are counted by hand. The number of conversations genuinely opened, meaning people who replied with something other than a courtesy. The number of meetings obtained. And the number of proposals sent.
The judging period is this channel’s particularity: a decision cycle in business-to-business sales runs to months, and often to quarters. Judging a campaign after three weeks is like opening the oven to see whether the bread is rising.
The consequence is that the length of the test and the number of conversations that would make it conclusive have to be decided in advance. Write both numbers down before spending, because afterwards you will adjust them without meaning to.
And as everywhere in this pillar, the question put to the customer — how did you hear of us — remains the most reliable measurement, and here it is the only one that survives a six-month cycle.
What to check before signing
The first question is this article’s: how many Algerian businesses can buy what I sell, and can you name them to me? A supplier who has not made that list with you cannot know whether the channel makes sense.
The second is about the alternative: why a campaign rather than direct contact with that same list? An honest answer exists in some cases; the absence of an answer is itself the information.
The third is about the writing: who will post, under what name, and how often. If the answer is "the company page", point out that a person’s content travels better, and ask how that will be achieved.
The fourth is about messages: will there be automated sequences? If so, ask under whose name they will go out, and whether that person agrees to their professional reputation being used as the vehicle.
The fifth is the usual test: what the supplier would refuse to do. Here the right answer is "I would refuse if your list fits on one sheet", and it deletes the line they have just proposed.
What we do, and what we will refuse to do
What we will refuse: selling a campaign to a business whose list of possible customers fits on one sheet. Those people are contacted one at a time, free, and saying so costs us the campaign every time we say it.
We will refuse to set up automated message sequences, whatever return is claimed for them. They bill well, they go out under a real person’s name, and that name circulates in a world where everybody knows everybody.
We will also refuse to propose this platform to a business selling to consumers, and to quote a cost per click "typical in Algeria" — no dated series establishes one, and inventing it would be the easiest way to get a budget accepted.
What we do, when the channel makes sense: the named list drawn up before any budget; one language per audience and no hesitation between three; the writing handed to a person in your business rather than to a page; and the length of the test and the number of conclusive conversations fixed in writing before the first spend.
And what you should do without us this week: write the names. One sheet, one hour, the Algerian businesses that can buy what you sell this year. If you have fewer than twenty, you have just saved a budget and discovered what to do instead.
Frequently asked questions
Is this platform right for our business?
Only if you sell to other businesses. Then write the names of those that could buy this year: fewer than twenty and the work is sales, not advertising; several thousand and you are in a disguised consumer market.
Why is the click so expensive here?
Because you are paying for declared targeting — people wrote their own role and employer. It is worth the price only if the role you aim at is the one that decides at your customers; if your buyer is an owner who never updates a profile, the precision matches nothing.
Should we post from the company page?
A person’s content travels markedly better than a page’s, and that is a fact of distribution rather than a taste. So the question to settle before budgeting is: who in your business will write under their own name every week for six months?
Which language should we write in?
French for a national market, English if your customers are not in Algeria. Above all, one per audience: an account alternating languages is read by fewer people at each switch, because distribution depends on the first readers.
Do automated message sequences work?
No, and they cost more than they return. An accepted connection followed by an immediate pitch produces a permanent filing rather than a refusal, and in a market where everybody knows everybody that is paid for in reputation.
After how long should we judge?
In months, often in quarters, because the business-to-business decision cycle is long. Fix in writing, before spending, the length of the test and the number of conversations that would make it conclusive — afterwards you will adjust them without noticing.
Where we come in
How many names you wrote down decides everything else. Under twenty, advertising makes no sense; above it, the question becomes which of the two routes.
- We count with you and settle between named approach and paid distribution.
- We verify those companies actually exist on the platform, one at a time.
- We pick one language per audience, without mixing inside a single message.
If you sell to consumers this platform is not for you, whatever budget is available.
Read next
Approaching twenty named businesses: the method without advertising
If you can name your possible customers, you should not be buying them. How to build the list, what to write, how many times to follow up, and when to stop.What advertising budget to set — and how to know
There is no published rate card for Algeria. Here is what the available data actually lets you conclude.A live campaign: the moves of the first week
The budget is set and the money is going out. What to look at, what to change, and above all what not to touch yet.
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.