Algeria AgencyDigital marketing
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.
The article accompanying this one ends on a sentence it does not develop: if you can name the businesses you want as customers, advertising is rarely the right tool. This page is what that sentence implies, in practice.
The situation is common here and badly served by imported methods. Many Algerian businesses have a possible market of twenty to two hundred accounts — manufacturers, distributors, clinics, schools — whose list fits on one page. Buying impressions to reach a population you could write to by hand is an expensive detour.
This article describes the detour avoided: building the list, finding the right person, writing the first message, following up a finite number of times, recognising a real reply, holding a sustainable posting cadence, and tracking twenty conversations without buying a tool for it.
One figure appears, and it removes a step the foreign manuals insist on: in a country where the overwhelming majority of registered operators are natural persons, hunting for the decision maker in an org chart is a search with no object.
Twenty names replace an audience
The difference between the two approaches is not budgetary, it is logical. An advertising campaign buys chances to be seen by people described by attributes; a named approach addresses businesses you chose one at a time, because you know why that one.
The second is only possible under one condition, and it is the only question to ask before starting: can you write the list? If the possible market runs to thousands of anonymous businesses, advertising does work you cannot do by hand. If it fits on a page, the reverse is true.
What makes this method uncomfortable is that it produces nothing for two weeks and does not delegate easily. A campaign shows figures from day one; twenty messages written one at a time show nothing until somebody replies, and the temptation to conclude too early is the main failure mode.
You also have to accept the arithmetic. Out of twenty businesses correctly chosen and correctly approached, a handful reply, one or two open a discussion, and the commercial result sometimes arrives months later. That is not a poor return: it is the normal return, and it is excellent against the cost.
Finally, this method has a property advertising does not: it leaves something behind. A campaign stopped leaves nothing; twenty conversations leave twenty people who know who you are, and half of what gets signed the following year comes from there.
Building the list
The list is not found, it is built, and it is built from what you already know. Start with your current customers and write, for each, why they bought. The businesses that resemble those reasons are your list, and that criterion beats any abstract segmentation.
The sources after that are ordinary and free. Trade directories and sector association sites; exhibitor lists from your sector’s fairs, which are public and have the advantage of naming the person who attended; published tenders, which reveal which businesses buy what; and your customers’ customers.
Set the size at twenty to begin with, and resist putting a hundred. Twenty is what one person can seriously approach in a month while writing something specific to each; a hundred produces a generic message sent a hundred times, which is exactly what this method exists to avoid.
For each line note three things and no more: the business name, why it is on the list, and what you know about it that is true and recent. The third column is what will make the message writable; a line without it is a line you have not yet chosen.
Remove without regret the businesses whose third column you cannot fill. A list of twelve names you know beats a list of forty of which twenty-eight are trading names read somewhere.
The right person is often the only person
Imported methods devote a whole step to identifying the decision maker: map the org chart, separate the user from the influencer and the buyer, work up to the level that signs. That is serious work in a market of large structures.
It is not the situation here, and the figure says so unambiguously. Of the operators registered with the commercial register in Algeria, the great majority are natural persons rather than companies. For those businesses there is no org chart to climb: the person who decides is the person who replies, and it is often the same one who opens the post.
The practical consequence is a shift of effort. The time the manuals spend on mapping transfers entirely to what you are going to write, which is the variable that actually matters at this scale. Spend five minutes on who runs it; spend thirty on the three sentences you address to them.
For the minority that genuinely are structured companies — industrial groups, large clinics, institutions — the opposite rule applies and it must be respected: writing to the wrong person in a hierarchy does not produce a refusal, it produces a silence indistinguishable from every other silence.
Finally, check the person exists as you understood them, and check it somewhere other than where you found them. A profile that has not moved in four years, a role that changed, a business that closed: three minutes avoid a message that discredits you before its second line.
CNRC, position as at 10 December 2025
The first message: what it may not contain
The first message is short and is judged on what it does not contain. It does not contain your brochure, your list of services, a request for a meeting, and above all not the sentence saying you have "a solution that might interest them".
It contains three things. Why you are writing to them specifically — that is the third column of your list, and it is what separates this message from the ten others they receive. What you do, in one sentence and in ordinary language. And a question that is easy to answer yes or no.
The closing question is the technical part. A meeting request asks the person to decide to give you an hour before knowing who you are, which is a lot to ask of a stranger; a closed question about their own situation takes ten seconds and opens the conversation where it should open, which is at their end.
On language: write in the one the person uses in their work. For most professional exchanges here that is French; for some sectors and some wilayas it is Arabic, and English is a choice that excludes more people than it impresses. The criterion is not your fluency but theirs.
Write the twenty messages in one sitting, but write them one at a time. The batched sitting is efficient; the template filled in series is not, and it is recognised instantly — the recipient cannot say why, but they know they received something that was not meant for them.
The follow-up, and how many
Silence is the commonest reply and it almost never means refusal. It means the message arrived on a busy Tuesday. That is why follow-up produces half this method’s results, and why not following up amounts to doing the work without taking the benefit.
The rule is a finite number decided in advance: two follow-ups, then stop. The number matters less than the fact that it exists — without it, you follow up three times with the people who reply politely and never with those who said nothing, which is the reverse of what is needed.
The interval widens: one week after the first message, then two weeks after the first follow-up. And each follow-up brings something new rather than repeating the question — an example, an observation, a useful link. A follow-up that only says "just circling back" asks for a reply while offering nothing in exchange.
After the second follow-up, write one last line that closes cleanly: that you will not press further, and that they can write when it becomes relevant. It is the message people reply to most often, and even without a reply it leaves a decent impression instead of awkwardness.
Then note the date and move on. A business that has not answered three messages is not lost: it is simply to be picked up in six months, with a new reason, and that restart works better than a fourth follow-up in the first series.
What a reply is, and what is not one
Three things look like success and are not. A profile view, a reaction to a post, and an accepted connection: all three are countable, visible in notifications, and give the impression that something is progressing when nothing has been said.
A reply is a sentence written by the person to you. It is the only unit that counts in this method, and it includes refusals: "we already work with somebody" is a reply, it is useful, and it gives you back a slot on your list.
The distinction is not pedantic. A method judged on profile views appears to work for two months and produces nothing; the same method judged on replies shows honestly, by the third week, whether the messages are good or need rewriting.
When a reply comes, the only possible mistake is to answer with the brochure. Somebody who writes three lines expects three lines, and an eight-page document in return closes the conversation as surely as silence. Answer briefly, answer what was asked, and propose one next step.
The personal account works, the page verifies
The two objects do not have the same function and confusing them is expensive. A personal account is what writes, what gets read and what gets replied to; a company page is what people check to confirm you exist, and its distribution is close to zero.
The consequence for effort is clear: the posting work is done from the personal account of somebody who runs the business or does the work, not from the page. The page is filled in once — what the business does, where it is, a few pieces of work, a means of contact — and then left alone.
That does not make the page useless. It is checked, systematically, by the person who has just received your message and wants to know who they are dealing with. An empty page, or one abandoned two years ago, is a negative signal at the precise moment somebody is deciding whether to reply.
The human difficulty is real and worth naming: this asks a business owner to write under their own name, which many do not want to do. The honest answer is that the method works less well without it, and that knowing this before starting beats compensating with volume.
Posting without a campaign: the sustainable cadence
Posting regularly is what makes the messages easier to send, because the person you write to has already seen your name. It is not an acquisition channel in itself at this scale; it is what turns a stranger’s message into a message from somebody vaguely known.
The cadence that holds is the one you will still be keeping in six months, and it is lower than the one set on day one. Two posts a month sustained for a year beat three a week for five weeks, and the second rhythm is the one almost everybody chooses.
The most effective subject is the most boring to write: what you did this week and what you learned from it. A problem met, a method that worked, an error corrected. It needs little preparation, never runs out, and demonstrates competence instead of asserting it.
What does not work is recycled content — quotations, sector figures, general advice. It is easy to produce, indistinguishable from what ten thousand other accounts post, and it occupies precisely the space that should hold the one thing only you can write.
Recruitment: the one advertisement worth paying for
The neighbouring article establishes that recruitment is this platform’s most solid paid use here. What needs adding is the operational part, because a job advertisement fails in three predictable ways.
The first is publishing a job description instead of an advertisement. A description lists responsibilities; an advertisement says who it is addressed to, what the person will do in their first month, who they will work with, and what the business offers in return. The difference shows immediately in the quality of applications.
The second is the absence of a salary range. It multiplies irrelevant applications and wastes both parties’ time; leaving it out is not a negotiating position, it is a filter you are declining to use.
The third is response time. Good candidates are in work and decide within days; a process that takes three weeks to give a first answer loses exactly the people the advertisement was paid for. Decide before publishing who answers, within what delay, and where applications arrive.
Note finally that the advertisement serves twice. The candidates who do not fit this time remain a readable pool for next year, provided they were kept somewhere and received a polite reply — which costs ten minutes and is remembered for a long time.
Tracking twenty conversations in a spreadsheet
At this scale a sales management tool costs more than it returns. Twenty lines are tracked in a spreadsheet, and the spreadsheet has one decisive property: you actually open it, which is not true of software you have to log into.
Six columns are enough: the business, the person, the date of the first message, the date of the last action, the state, and the next action with its date. The last column does all the work — without it a tracking sheet becomes a history, and a history makes nothing happen.
The states fit in four values: to write, written, in discussion, closed. Resist adding more. A seven-level scale requires a fine judgement at every update, and what happens is that the update stops being made.
Open the sheet once a week, at a fixed time, and do the actions whose date has passed. Twenty minutes. That is the entirety of the discipline this method asks for, and it is the one part not to delegate, because whoever does that review is the one who knows where the conversations stand.
How long before judging
The question comes after three weeks, when nothing has yet come of it, and the answer is that it is too early — which is hard to hear and true. The decision cycle of the businesses this method targets is counted in months, not days.
What is judged at three weeks is not the commercial result but the reply rate, and it is judged severely. Out of twenty specific messages sent to well-chosen businesses, zero replies after follow-up means something is wrong in the message or in the list, and that is corrected straight away.
What is judged at three months is the conversion of discussions into proposals. And what is judged at six months is the commercial result, with one caution: the first signatures often come from conversations opened at the start, which means a judgement at six months is really about the first two months’ work.
Keep one additional measure through that period, the best leading indicator available: the number of live conversations. It rises before anything is signed, it falls when the weekly discipline slips, and it warns several weeks in advance of an empty quarter.
What we do, and what we refuse
What we do is bounded. We build the list with you from your current customers, we write the messages, we keep the tracking sheet and the weekly review, and we prepare recruitment advertisements when there are any.
We refuse to send messages from a client’s personal account. That account carries their name, and a conversation opened in their name by somebody else puts them in difficulty the moment the other party replies with something specific. We write the messages, the person sends them, and that slowness is the price of the one thing that makes the method work.
We also refuse to work from a list you have not read line by line. A bought or automatically compiled list produces generic messages addressed to businesses nobody chose, which is an expensive advertising campaign sent by hand.
And we do not promise a number of meetings. What depends on us is the number of specific messages sent, their quality, and the regularity of the follow-ups; what does not is the moment a business needs what you sell. There is also work only you can do, and it decides everything else: write, for each of your last ten customers, why they bought.
Frequently asked questions
How many businesses should the list start with?
Twenty. That is what one person can seriously approach in a month while writing something specific to each. A hundred names produce a generic message sent a hundred times, which is exactly what this method exists to avoid. Remove any line where you cannot write down something true and recent that you know about that business.
How do we find the decision maker?
Most often there is nobody to find. Nine in ten registered operators in Algeria are natural persons: the person who decides is the person who replies. Spend the five minutes needed to confirm who runs it, then put the thirty minutes saved into the three sentences you write them. For genuinely structured companies the opposite rule applies and must be respected.
How many times can we follow up?
Twice, then stop — and the number matters less than having decided it in advance. One week after the message, then two weeks after that. Each follow-up brings something new rather than repeating the question, and the last says you will not press further: that is the one people reply to most often.
Is a profile view a good sign?
It is a sign of nothing, and counting it is the commonest way to believe a method is working for two months while it produces nothing. The only unit that counts is a sentence written by the person to you — including a refusal, which is a useful reply and gives you back a slot on the list.
Should we post from the company page?
No: it distributes almost nothing. The posting work is done from the personal account of somebody who runs the business or does the work. The page has a different, real and passive function — it is checked by the person who has just received your message and wants to confirm you exist, so it should be filled in once and not abandoned.
How long before we know whether this works?
Three weeks to judge the reply rate, which is corrected immediately if it is zero. Three months for discussions converting into proposals. Six months for the commercial result — and that judgement is really about the first two months’ work. Meanwhile track the number of live conversations: it is the only leading indicator available.
Where we come in
The last column carries the next action and its date, and that is what moves anything forward. After three weeks nobody fills it in any more.
- We keep that column current and remind you what falls due.
- We write each message from an account carrying your name, not ours.
- We stop after the third follow-up, whatever the urge to continue.
If you have not read your list line by line we will not use it: an automatically compiled list produces twenty useless messages.
Read next
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.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.