The Marketing Metrics That Actually Matter

A monthly report lands in your inbox. Impressions up 240%. Reach up 180%. Followers up from 4,100 to 6,700. Engagement rate above industry average, with a chart to prove it. Then you check your accounts and revenue is within a few hundred dollars of where it was last quarter.

This happens constantly, and usually nobody is lying. Every number in that report is true. They're simply measuring activity rather than outcome, and activity is much easier to grow than revenue.

The numbers that look like progress

None of these are worthless. They're all diagnostic, meaning they help explain why something is or isn't working. The mistake is treating them as the result rather than as an input to the result.

• Follower count: measures how many people once tapped a button. It says nothing about whether any of them are in your market, and a follower who found you through a giveaway is worth roughly nothing.
• Impressions and reach: measures how many times something appeared on a screen, including screens scrolled past in under a second. It's the easiest metric to inflate, because more budget buys more of it automatically.
• Raw pageviews: useful only when you know which pages. Ten thousand views on a blog post about industry news and 200 views on your pricing page is a worse month than the reverse.
• Engagement rate without context: a post about your office dog will outperform your service announcement every time. That's a real result for the dog and no result for the business.

The tell is simple. If a number can go up while your revenue goes down, it can't be the number you steer by.

The four that track with money

These are harder to collect and less flattering in a slide deck, which is precisely why they're worth insisting on.

• Qualified leads, not form submissions. A qualified lead is someone in your service area, with a budget in your range, who wants something you sell. If 60 forms came in and 11 of them met that bar, the number is 11. Agencies report 60 because it's larger.
• Cost per acquisition by channel. Total spend on a channel divided by the customers it produced, not the leads. If Google Ads cost $2,400 last month and closed six customers, your CPA on that channel is $400. That single number decides whether the channel stays funded.
• Conversion rate at each stage. Visitor to enquiry, enquiry to qualified, qualified to quote, quote to sale. Four percentages instead of one, because they tell you where the loss is. Plenty of businesses with a traffic problem have a quote-to-sale problem.
• Customer lifetime value. What a customer is worth across the whole relationship, not on their first invoice. A first order of $400 that repeats twice a year for three years is a $2,400 customer, and you can afford to pay far more to win one than the first invoice suggests.

If a number can rise while your revenue falls, it isn't a result. It's an explanation waiting to be attached to one.

Why lifetime value changes every other decision

Say you're running two channels. Search ads cost you $180 to acquire a customer. Social ads cost $60. On cost per acquisition alone, social wins by a wide margin and you'd shift budget across.

Then you look at what each customer is worth over two years. The search customer, who came in looking for exactly what you sell, averages $1,100. The social customer, who saw an offer and impulse-bought, averages $190 and rarely returns. Search returns roughly six dollars for every one spent. Social returns just over three, and that's before you account for the support hours. The cheaper channel was the worse one, and no report built on cost per acquisition alone would have shown you that.

Lifetime value is also what tells you how much you're allowed to spend. A business that knows its customers are worth $1,100 can outbid a competitor guessing from a $400 first order, and will win the auction every time without overpaying.

Setting up reporting that survives a bad month

None of this requires enterprise software. It requires three habits that most small businesses skip.

• Ask every enquiry how they found you and record the answer in the same field every time. Analytics attribution breaks constantly, and a one-line question on your form or your first phone call is more accurate than most tracking setups.
• Track enquiries through to closed sale in one place, even a spreadsheet, so you can connect revenue back to the channel that produced it. Without this link, every other metric is guesswork.
• Review on a quarter, not a month. Marketing has too much noise at 30 days, especially in a business closing fewer than 20 deals a month. One good week can make a bad quarter look fine, and one slow month can bury a channel that was about to work.

When you next receive a report, the question to put to whoever wrote it is what this cost per customer, and what those customers are worth. If the answer arrives with numbers attached, you're working with someone measuring the right thing. If the answer redirects to reach and engagement, you now know what the reach and engagement were covering for.