Where your first hundred customers actually came from, and how to check

Before you read any further, write down where you think your last ten sales came from. One line, no hedging, put a percentage next to each source. Now go and look. If this is the first time you have checked, expect the ranking to be right at the top and wrong somewhere in the middle, and expect the wrong part to involve the channel you spend the most time on.

Why the guess goes wrong in the same direction every time

Attention is not evidence. The platform where people reply, argue and quote you is the one that feels like it is working, because working is what a busy comment section feels like. Meanwhile a newsletter with four hundred readers and no visible reaction quietly sends people who arrive already convinced. One of those produces the sensation of momentum and the other produces the money.

Recency does the rest. A good week on one platform rewrites your memory of the whole quarter. Ask someone in July where their sales came from and you will get an accurate description of June. This is not carelessness, it is how memory works, and it is exactly the kind of thing a written record is for.

The genuinely hard part is that no analytics tool sees everything. Someone watches a video on their phone, remembers your name three days later, types it into a search bar and buys. Every honest report will file that as direct traffic. If you treat direct as noise, you will underrate the channel that actually did the persuading, so hold that thought before you conclude anything.

Two ways of answering the same question

The audit is not about better tooling. It is about swapping one unit of measurement for another. Clicks and views tell you where people are. Buyers tell you where people trust you enough to enter a card number, and those are frequently different places.

The audit, in about forty minutes

Do this in one sitting with a notebook open. You are producing a single ranked list, not a dashboard habit, and you should finish with something written down that you can compare against next quarter.

1. 1Open Analytics in your store.fan dashboard and set the range to cover everything from your first sale to today, not the last thirty days.
2. 2Write down the traffic sources in order of buyers. Ignore the views column entirely on this pass, however painful that is.
3. 3Now write the same list ordered by revenue. Where a source moves up or down between the two lists, that source is selling a different product from the one you assumed.
4. 4Go product by product. A source that sends fifty views and two buyers to a 90 course is not the same asset as one that sends five hundred views and two buyers to a 9 download.
5. 5Export the customer list to CSV and sort by the date each person first appeared. Line the clusters up against your posting calendar; the spikes usually have a specific piece of work behind them.
6. 6Mark every person who has bought more than once. Which source did those people originally arrive from? That is the number that matters most and the one nobody looks at.
7. 7Write one sentence at the bottom: "Next month I do more of X and less of Y." Date it and put it somewhere you will find it again.

What the answer often looks like

Take a plausible shape. A hundred buyers over five months. Short video sent about seventy per cent of the views and produced perhaps a quarter of the buyers, almost all of them on the cheapest product. The email list sent under a tenth of the views and produced a third of the revenue, because those people bought the course. Direct traffic was a fifth. A single answer left in a forum thread in March was still sending three or four buyers a month in July, and nobody had thought about that thread since March.

Nothing there tells you to abandon short video. It tells you what each channel is for. Video introduces you and sells the small thing; email sells the expensive thing; the forum post is an asset you built once and forgot. The obvious next move is not more of everything, it is one more forum-shaped asset and one more reason to be on the email list. These figures are an illustration, not a benchmark, and yours will be shaped differently, which is precisely why guessing does not work.

The mistake this audit is designed to prevent

The classic error is killing the quiet channel. It sends the fewest people, gets the least engagement and takes an hour a week, so it goes first when you are tired. Then revenue drops the following month and nothing on the busy channel explains why. The reverse mistake is just as common: reading two months of data as if it were destiny and dropping everything that has not paid off yet, when the real problem is that some channels take a season to compound. Look at the whole history, not the last thirty days, and read the analytics overview so you know what each figure is counting before you act on it.

You do not need more channels. You need to know which of the ones you already have is doing the work, and that is a forty-minute question with a written answer at the end of it. The customer list keeps the record whether or not you ever look at it, which means the audit you run next January is only as good as the account you open now. Plans and what each one includes are on the pricing page.