Session 7: Reading the Numbers

Reach is how many different people saw your content. Impressions is how many times it was displayed — the same person seeing it three times counts three impressions. Impressions above reach means repetition, which is useful for remembering but eventually becomes annoying; a ratio far above three usually means the audience is too small or the campaign has run too long.

Engagement is what people did: likes, comments, saves, shares, clicks. The raw number is meaningless without context — what matters is engagement relative to reach, because 50 engagements from 500 people reached is a completely different result from 50 engagements from 20,000. Divide engagements by reach and you have a rate you can compare between posts, which is the only comparison worth making.

Then rank the engagement types by value. Saves and shares mean the content was worth keeping or passing on, and they are strong distribution signals. Comments mean it started a conversation and are also strong. Clicks mean someone wanted more, which is the closest thing to buying intent. Likes are the cheapest action a person can take and the weakest signal of all — which is why an account can have many likes and no customers.

Click-through rate (CTR) is clicks divided by impressions, expressed as a percentage. It answers one question: of everyone who saw this, how many wanted to know more? A low CTR is not a problem with your product — it means the creative or the first line did not earn the click. People saw it and moved on, so change what they saw, not what you sell.

Benchmarks vary by platform and industry, so comparing yourself to a global average is not useful. Compare your own posts to each other: if your average CTR is 1% and one post reached 3%, that post's opening and creative are worth repeating. Your own history is the only honest benchmark you have, and it takes about twenty posts before it becomes reliable.

One warning: a high CTR is not automatically good. Clickbait produces clicks from people who will not buy, and they arrive at your WhatsApp and leave, costing you the click and the conversation. The number to trust is not CTR alone but clicks that turn into enquiries, which is the next stage in the funnel.

A lead is someone who showed interest and gave you a way to reach them — a message, a form, a phone number. A conversion is the action you actually wanted: a sale, a booking, a registration. They are different numbers, and the ratio between them tells you how well you sell, not how well you market.

Cost per result is what you spent divided by the results you got — cost per click, cost per lead, cost per sale. This is the number that decides everything, and it only means something compared to your margin. If a product earns you ₦4,000 profit and a sale costs ₦1,200 to acquire, the campaign works. If acquisition costs ₦5,000, no amount of optimisation makes it profitable — the price, the offer or the product has to change.

That last point is the one most beginners resist, because it feels like giving up. It is not: it is the difference between marketing and gambling. A campaign that cannot reach a sustainable cost per result is telling you something true about the offer, and continuing to spend against it is how a small budget disappears entirely. Set the affordable cost per result before you spend, from the margin, and judge against that number.

Every campaign is a funnel: impressions → clicks → enquiries → sales. Read it in order and the problem names itself. Many impressions and few clicks means the creative or hook failed. Many clicks and few enquiries means the destination or ask failed — the advert and the profile said different things, or nobody answered. Many enquiries and few sales means price, delivery, trust or follow-up failed, and that is a selling problem rather than a marketing one.

Working backwards is the discipline. Start from the result you wanted and walk back to the first stage where the number collapsed. Beginners do the opposite: they see low sales and change the advert, when the advert was fine and the WhatsApp replies were taking a day. Fixing the wrong stage is the most expensive mistake in paid promotion, and it is entirely avoidable by reading the numbers in order.

Then change one thing at a time and give it enough time. If you change the creative, the audience and the offer in the same week, you will not know what worked, and you will have spent money to learn nothing. One variable, a full week, then read again.

Small numbers lie. A post with 40 impressions that produced 3 clicks has a 7.5% CTR, which means nothing — with numbers that small, one person behaving differently changes everything. As a rough rule, do not draw conclusions from fewer than a few hundred impressions or a couple of dozen clicks; below that, you are reading weather rather than a pattern.

Nor does one good result prove anything. Posts perform differently by day, time, mood and luck, so a single outstanding post may be a fluke. The reliable evidence is a pattern across several similar posts — if your how-to posts consistently outsave your product posts across a month, that is a finding. If one how-to post did well once, it is an anecdote.

Finally, watch for vanity metrics: follower count, likes, reach. They are pleasant to watch and they do not pay for anything. The metrics that matter are enquiries, sales and cost per result, and a report built on the first group will always look better than one built on the second — which is exactly why people build them that way. Build the honest one.