The most common reason businesses cannot tell whether their marketing works is that nobody wrote down what “working” would look like before starting. Without that, every report becomes an argument about interpretation, and whoever produces the report wins.
Published: August 2026. Written by Khaled Samy, Founder & CEO of Aelaany, based on how we measure and report results for Egyptian and Gulf clients.
Metrics that feel good and prove nothing
- Impressions and reach. How many people could have seen something. Not how many noticed.
- Followers. Purchasable, and inflated counts actively reduce your reach because they never engage.
- Likes. Weakly correlated with buying. People like things they will never purchase.
- Website visits, on their own. Traffic from the wrong people is a cost, not an achievement.
- Keyword rankings, on their own. Ranking first for something nobody searches is a screenshot, not a result.
These are not useless — they are diagnostic. They explain why a number moved. They are just not the number.
Measurement only pays off when something acts on it — that is the reporting loop inside our paid campaign management.
The four that actually matter
- Qualified enquiries. Calls, messages, forms and visits from people who could plausibly buy. Count them, and count them by source.
- Cost per enquiry. Everything you spent, divided by that count. This is the number that makes channels comparable.
- Conversion rate from enquiry to customer. If this is low, the problem is usually not marketing — it is what happens after the enquiry arrives.
- Customer value. What a customer is worth over the whole relationship. Without this, you cannot tell whether a cost per enquiry is good or ruinous.
Set a baseline before you start
Write down, before any campaign begins: current monthly enquiries, roughly where they come from, current conversion rate, and average customer value. Even rough figures are enough. Without a baseline, three months later nobody can distinguish improvement from seasonality — and both sides will pick the interpretation that suits them.
The tracking that has to exist
None of the above is measurable without a few things in place. In practice this is the layer most Egyptian businesses are missing:
- Analytics installed and actually configured, not just pasted in.
- Form submissions recorded as conversions.
- WhatsApp and call clicks tracked as conversions — critical in Egypt, where most enquiries arrive this way.
- Search Console connected, so organic performance is visible at all.
- One habit that costs nothing: asking every new customer how they found you, and writing the answer down.
That last one catches what analytics cannot — word of mouth, offline signage, someone who saw a post three weeks ago and searched your name today.
How long to wait before judging
- Paid ads: two to four weeks for early signal, longer for stable optimisation.
- Social content: two to three months of consistency before the pattern is readable.
- SEO: three to six months for meaningful movement, longer in competitive markets.
- Email: almost immediate — you know within a couple of sends.
Judging SEO at week three or abandoning a content plan after a month is not discipline; it is guaranteeing the spend was wasted.
The uncomfortable finding
Sometimes the honest answer is that marketing is working and the business is losing the customers afterwards — slow replies, no follow-up, quotes that never get sent. This is a common outcome, and it is not fixed by spending more on ads. It is worth knowing which problem you actually have before buying a solution to the other one.
Frequently Asked Questions
Which marketing metrics prove nothing on their own?
Impressions and reach, followers, likes, website visits on their own, and keyword rankings on their own. These are not useless, they are diagnostic: they explain why a number moved. They are just not the number.
Which four numbers actually matter?
Qualified enquiries counted by source, cost per enquiry, conversion rate from enquiry to customer, and customer value over the whole relationship. Cost per enquiry is the number that makes channels comparable, and without customer value you cannot tell whether a cost per enquiry is good or ruinous.
Why do I need a baseline before starting?
Write down current monthly enquiries, roughly where they come from, current conversion rate, and average customer value before any campaign begins. Even rough figures are enough. Without a baseline, three months later nobody can distinguish improvement from seasonality, and both sides will pick the interpretation that suits them.
What tracking has to be in place?
Analytics installed and actually configured rather than just pasted in, form submissions recorded as conversions, WhatsApp and call clicks tracked as conversions, and Search Console connected. Add one habit that costs nothing: ask every new customer how they found you and write the answer down, because that catches the word of mouth and offline signage analytics cannot.
How long should I wait before judging a channel?
Paid ads need two to four weeks for early signal and longer for stable optimisation. Social content needs two to three months of consistency before the pattern is readable. SEO needs three to six months for meaningful movement, longer in competitive markets. Email is almost immediate, since you know within a couple of sends.
What if the marketing works but sales do not follow?
Sometimes the honest answer is that marketing is working and the business is losing the customers afterwards, through slow replies, no follow-up, and quotes that never get sent. This is a common outcome, and it is not fixed by spending more on ads. It is worth knowing which problem you actually have before buying a solution to the other one.
More in: Analytics & Measurement