The report says it worked. Your bank account disagrees.

You get a PDF at the start of every month. Impressions up. Reach up. Engagement up. A graph with a line going the right way.

And yet nothing in the business feels different. That is not paranoia, and it does not necessarily mean anyone lied to you.

The metrics that are true and useless

• Impressions and reach. How many times something appeared. Appearing is not a result. These go up when you spend more, essentially always.
• Engagement rate. Meaningful for content strategy. Meaningless as a business outcome. A post can be loved by people who will never buy.
• Followers. Only matter if you can reach them later, which depends on a platform that changes the rules.
• Website sessions. Useful as a denominator, not as a headline. Traffic with no enquiry is a cost.
• Keyword rankings. Position three for a term nobody searches is a screenshot, not a customer.

None of these are fake. They are simply upstream of the thing you care about, and they move even when the thing you care about does not.

The four that settle it

• Enquiries. Calls, forms, messages. The raw count of people who raised a hand.
• Qualified enquiries. The ones who were actually in your area, in budget and wanting what you sell.
• Cost per qualified enquiry. Total spend, including the agency fee, divided by that second number. Excluding the fee is the oldest trick in the file.
• Jobs won and their value. Which requires your side of the data, which is why this one is usually missing.

Three questions worth asking

Ask them calmly. A competent agency will have answers and will not mind.

• Is the agency fee included in the cost-per-lead figure?
• How many of last month's enquiries were qualified, and how do you define that?
• Which channel produced the enquiries that became paying work?

If the answer to the third is that it cannot be tracked, that is sometimes genuinely true, especially for phone-heavy trades. It should still come with a plan to make it less true.

Apply this to us too

We send reports. You should read ours the same way, and we would rather you asked the awkward question in month two than quietly concluded in month nine that it was not working.

The point of a report is not to demonstrate effort. It is to decide what to do next month. If it does not change a decision, it is a newsletter about yourself.