Which Marketing Channel Makes Money? How to Tell
Which marketing channel makes money? Count paying customers, not signups, and ask three questions of every channel. Then cut the weakest one without delay.
To know which marketing channel makes money, count the paying customers each one produced, not the visits or signups. Then ask three questions of every channel: how many, at what rate, and at what value. A channel is only good if it holds up on all three.
How to know which marketing channel makes money: stop counting signups
Let's start with the mistake. It's the most common one and the most costly.
You count signups. Signups are easy. They respond fast. They make every channel look decent.
That's exactly the problem. Channels differ hugely in the kind of person they send. A source delivering a crowd of free-tier tourists will beat a source delivering a few serious buyers on every signup chart you'll ever build. You'll move your evenings, confidently, in the wrong direction. Everything on the screen will be green.
So define your outcome as far down as your business allows. A paid subscription. A completed purchase. A closed deal.
If your sales cycle is too slow to steer by that, pick the furthest milestone that reliably predicts money. Booked a call. Upgraded. Activated. Then be honest with yourself that it's a stand-in.
Write the definition in one sentence. If you'd word it differently on a different day, you don't have a measurement system yet. You have a mood.
Which marketing channel makes money: the three questions, in order
Once the outcome is fixed, put every channel through the same three questions. Order matters.
- How many? Outcomes produced.
- At what rate? The share of visitors who turn into that outcome.
- At what value? The revenue, and eventually how long those customers stay.
Each question alone is a trap. Here's how they fail.
| What you see | What it usually means | What to do |
|---|---|---|
| High volume, terrible rate | Expensive noise. Lots of people, few buyers. | Stop celebrating the traffic. Test whether the audience is wrong. |
| Superb rate, trivial volume | A nice fact you can't build on yet. | Find out if it can be scaled, or leave it alone. |
| Good volume and rate, customers leave quickly | A machine for manufacturing support work. | Worst of the three for one person. Check value before you feed it. |
| Modest on all three, but holds up | Quietly good. This marketing channel makes money without making noise. | Protect it. It's probably not exciting, which is why it survives. |
Read that third row again. For a solo operator, a customer who leaves in two months and writes in twice is a bill, not a win. The cheapest channel to acquire from is frequently not the cheapest to serve.
Free guide · 26 pagesStop paying for what does not work.The Attribution Playbook: how to find which marketing produces money.Get the free guide →Why the average lies to you
You may already have a number for what a customer costs you. One figure, averaged across everything you do.
It's useless. It's a blend. Two channels can produce the same number of customers at wildly different cost and wildly different quality, and the blend reports a comfortable middle that describes neither of them.
The fix is to split it. Take each channel on its own. And if you keep watching past the sale, you'll see which sources produce customers who stay. That's the punchline most people never reach, because they stop at the sale.
Cutting is the fastest return you have
Everyone wants the winner. Finding it feels like the point.
It isn't. The bigger, quicker gain is finding the loser.
You almost certainly spend evenings on at least one thing that produces nothing. It survives because it has a plausible story and a sunk cost, and you're its advocate. Nobody can prove it doesn't work, so it stays.
Evidence ends that. The evenings go straight into something already working. You don't need a new idea or a new channel. You need the willingness to look.
If this exercise has never made you stop doing anything, it hasn't paid for itself. It's a comfort object. How to actually pull the trigger is in how to kill a marketing channel.
Be careful with slow channels
There's one place where this goes wrong.
Some marketing pays back in days. Some pays back in quarters. Content, a podcast, showing up in a community, search. Judge a slow channel monthly against a fast one and it loses every time, and it deserves to lose none of them.
So measure the lag. For customers credited to a slow channel, how long passed between the first touch and the money? Once you know that number, you know how long a fair trial is. Decide it in advance.
That is the only reliable defence against cutting something in month four that pays back in month seven.
The catch: you'll never see it all
Some of your best marketing is invisible. Nobody clicks a tagged link out of a podcast, a talk, or a conversation. So some channels will look weaker than they are.
A free-text question on your main form fixes much of that. Something like "how did you hear about us?" It's optional, it's messy, and it catches the invisible half better than any tracker, because it asks the one party who knows. Never use it alone. Beside tracked data, it stops you cutting the channel everyone mentions and nothing measures. More on that in how to track untaggable marketing channels.
And treat the numbers as a compass. Use them to compare channels against each other. Don't expect them to reconcile to your bank account.
Free assessmentHow trapped are you, really?Take the Corporate Suffocation Index, the free solopreneur readiness assessment.Take the assessment →What to do on Monday
- Write your outcome in one sentence.
- List every channel you spend time on.
- For each, answer the three questions with whatever you have. Gaps count as findings.
- Pick the weakest. Decide how long it gets before it's cut.
Do that and you've done more than most people with a dashboard.
Where the data comes from
Knowing the questions is one thing. Answering them is another. You need three layers joined up, and most people build only one.
Touchpoints are the anonymous visits. Identity is the moment a visitor gives you an email. Outcome is whether they paid. Your site analytics gives you the first, your payment system gives you the third, and the form in the middle is what joins them.
If the form drops the source when it creates the person, you can still count visits and you can still count sales, and you can't connect them. Every channel then looks the same, because you can't tell which customer came from where. If that sounds familiar, how to track where your leads come from shows the fix.
Read monthly, not daily
This data is noisy. At your volume, a single week means almost nothing.
Check it daily and you'll react to variance. You change something, and now you can't interpret the next reading either. It's whiplash. So set one day a month and give it a fixed agenda.
- What did each channel produce on the three questions?
- What moved, and do I believe it?
- What is the one reallocation I'm making?
- What did last month's change actually do?
Change one significant thing at a time. If you cut a channel, double down on another, rebuild a landing page and change your pricing in the same month, the following month tells you nothing. It feels slower to sequence it. It's dramatically faster, because every change teaches you something. The full routine is in how to review marketing results every month.
Keep a short written record of each decision and the reasoning. A year of those notes is worth more than any dashboard, because it lets you check your own judgement, not just your channels.
Free guide · 26 pagesStop paying for what does not work.The Attribution Playbook: how to find which marketing produces money.Get the free guide →What a good answer looks like
After one pass, you should be able to say something like this out loud. "This channel sends the most visitors, but a small share buy, and they leave fast. That one sends few, but they pay and stay. The third I can't judge yet, because it works slowly."
Three channels, three different verdicts, and a decision for each. That is how you see which marketing channel makes money. Feed one, question one, and give one a fixed trial. That's the output. Not a chart. A sentence you can act on before dinner.
If you can't say that sentence yet, don't worry. The gap is the finding. It tells you which part of your setup to fix next.
Free assessmentHow trapped are you, really?Take the Corporate Suffocation Index, the free solopreneur readiness assessment.Take the assessment →Frequently asked questions
What if I don't have enough customers to compare channels?
Use the furthest milestone that reliably predicts money, such as a booked call, and be clear it's a stand-in. Compare channels on that, and add the self-reported answer from your form as a second opinion.
Should I judge every channel monthly?
Review monthly, but give slow channels a trial length you decided in advance. Content and community can take far longer to pay back than a campaign, so a monthly verdict will wrongly kill them.
Why not just count signups?
Because channels send different quality of person. A channel sending many free users will beat one sending a few buyers on every signup view, and you'll shift your time the wrong way.
Is the cheapest channel the best one?
Not necessarily. The cheapest source to acquire from is often not the cheapest to serve. Look at how long the customers it produces stay before you decide which marketing channel makes money.
The full method, with the checklists, is in The Attribution Playbook, a free 26-page playbook.
How trapped are you, really?
Take the Corporate Suffocation Index, the free solopreneur readiness assessment.