Marketing advice is the cheapest thing in the market. That is exactly why most of it is given with confidence by people who carry none of the consequences, and a small business pays for the wrong advice because it has no margin to absorb a mistake.
I am not going to tell you to listen to me instead of them. Here are five pieces of advice that keep coming back, why each one is wrong, and at the end three questions you can use to filter any advice you are given, including mine.
One: "be on every platform"
This gets said as though it were free. It is not. Every platform costs time, a different kind of content and a different habit of replying, and a business where one person does everything spreads that person across five platforms and ends up weak on all five.
The part nobody mentions is that platform numbers do not compare the way people assume. In DataReportal’s Digital 2026 report for Egypt, internet users at the end of 2025 stood at 98.2 million, a penetration rate of 82.7 per cent, while active social media user identities were 51.6 million in October 2025, around 43.4 per cent of the population. Those are two different measures, and the gap between them is the first thing to notice.
The report puts warnings on its own figures. Its LinkedIn number for Egypt, 15.0 million, is described as registered members rather than monthly active users, and the platform figures are advertising reach estimates that are not directly comparable with each other. If you are choosing a platform because "it has more people", you are comparing things that do not compare.
What works instead: one platform where your customer actually is, until you are good at it. The rest can wait. And if the real question is whether social media is enough at all, that is whether social media is enough for a business.
Two: "the number in the ad dashboard is the truth"
This is the most dangerous one on the list, because it looks scientific. Someone opens the dashboard and says: look, forty conversions. That number is not necessarily forty things that happened.
Google documents this itself. On its page about modeled conversions, Google says modeled conversions are reported in the same Conversions column as observed ones, and that modeling predicts attribution and "doesn’t determine whether or not a conversion happened". The same page says modeled conversions can take up to five days to fully process and stabilise, and that conversion values can see retroactive increases over several days.
Three practical consequences. The number you saw yesterday is not final, so do not decide on a two day report. If the account has very few weekly conversions, Google says it does not report modeled conversions at all where data is too sparse, so your number is structurally incomplete. And most importantly, if someone is selling you a result using that number alone, ask for the number from your side: sales, transfers, invoices.
What works instead: treat the platform number as a direction and your own business number as the verdict. Reconciling the two is monthly work, not a one off.
Three: "drop your price to compete"
This works once and then eats you. Cutting the price does not win a customer, it teaches the market that your price is negotiable, and a customer who arrived on a discount leaves on a bigger one.
There is also simple arithmetic that gets forgotten: a discount comes out of profit, not revenue. Ten per cent off a thirty per cent margin means you gave away a third of your profit to win the deal. Handling a cheaper competitor without cutting your own price is its own subject.
Four: "wait until everything is ready"
Logo, identity, website, brand video, then we start. It sounds professional and it is really a delay. A business waiting to look ready spends its money on appearance before it knows whether anyone wants what it sells.
The order that works is the reverse: sell one thing to one person with what you have, listen to the objection, fix the offer, then spend on appearance. Early trust comes from small promises kept rather than from a finished identity, which is the argument in how a new brand earns trust.
Five: "the person advising you is the person selling the fix"
This one is not advice, it is a pattern. Someone tells you that you need ads, and they run ads. Someone tells you that you need a new identity, and they are a designer. Someone tells you that you need automation, and they sell tools. None of them is necessarily lying. Each is seeing the problem from where they stand.
That applies to me too. When I say start with analysis rather than with ads, that is an opinion from somebody who trains and consults, so take it as a claim worth questioning rather than as a fact. The filter below is what protects you from them and from me.
Three questions that filter any advice
One: what is it based on? A published number, personal experience, or something going around? All three have value, and they have different value, and asking reveals which it is.
Two: does it fit my situation or somebody else’s? Advice that worked for a business with a budget and a team is not automatically advice for a business with one person in it.
Three: if it is wrong, who pays? Advice whose cost falls on you alone deserves more scrutiny than advice from somebody who shares the downside.
And if you want the order I actually work in before any advertising, that is why you start with analysis rather than with ads.
The honest limits
This is an opinion piece, not research. The first three have published figures I can point at. The last two are observations from working with small businesses, and that difference should be visible rather than hidden.
Every piece of advice above also has a case where it is right. Being on several platforms makes sense once you have a team and a working publishing process. A discount is a sound decision when it is temporary and tied to a clear goal such as clearing stock. And waiting until you look ready is correct if you are selling to a government body or a large company that needs documentation. Wrong advice is advice given as a rule for all cases, not advice given as a decision in one case.
On the commercial side of this, when you are the client hearing such advice from an agency, there is a breakdown of what a quote should actually contain from KF Agency in how to write a quote that convinces companies. It is in Arabic, and it is written from the seller’s side, which is precisely why it is useful when you are the one reading a quote.
The simplest test beats any channel matrix: ask the last ten customers who paid you how they found you. The answers show which channels work, and which ones you pay for that nobody came from. That closes Part 5 of my free book Growth Engine, which places this subject inside a complete growth system: Part 5, Digital channels: how you reach them (in Arabic).