How to Choose Your First Funnel: Start With Just One
How to choose your first funnel: pick the one sequence that ends nearest the money, keep to five or six steps, and add a second only once it changes a decision.
To choose your first funnel, pick the one sequence that ends in the outcome that matters most to you, keep it to five or six steps, and ignore every other path until this one has changed a decision. One well defined funnel that people actually look at beats six that nobody maintains.
The temptation, once you see how useful a funnel is, is to map everything at once: every path through the product, every entry point, every kind of customer. It feels thorough. In practice it produces a wall of charts and no decisions. This post is about the choice that keeps the work small enough to finish.
How to choose your first funnel: the one nearest the money
The rule from the guide is short. Choose the sequence that ends in the outcome that matters most, and define it properly. In most businesses that outcome is the payment, so the funnel to start with is the one whose last step is a customer paying.
There are good reasons for starting there. A funnel that ends nearest the money measures the thing the business lives on. Any improvement to it has the most direct effect on revenue. And when you bring a loss to a colleague, the case for fixing it is easy to make, because the cost of the leak is plain to see.
It also helps to remember what a funnel is for. Improving a step multiplies against every visitor you will ever acquire. The nearer that step sits to the payment, the more directly that multiplication shows up where it counts.
| Candidate funnel | Where it ends | Why it might start there |
|---|---|---|
| Stranger to paying customer | Paid | Nearest the money, directly tied to revenue |
| Stranger to email address | Gave us their email | Useful, but one step away from the outcome that matters most |
| Stranger to first use | Used it once | Informative, but still before the payment |
The table is not a ranking of what is easy. It is a reminder to ask where each funnel ends, and to prefer the one that ends closest to what you are in business to achieve. If you are unsure, the shorter question is: if this funnel improved, would I notice it in my revenue?
Why one funnel and not six
Instrumentation is easy. Attention is not. That single sentence explains why the advice is to start with one.
Each funnel you add brings its own steps to define, events to track, counts to check, windows to set and segments to read. Every one of those needs maintaining, and every one competes for the same limited attention. Six funnels that nobody looks at weekly tell you less than one that everybody looks at. The danger is not that you will miss something. It is that you will measure many things and change none.
Notice also that a funnel is only useful if it leads to action. A good test of your first funnel is whether it will feed the habit described in how to run a funnel improvement loop. One funnel can sustain that loop. Six usually cannot.
Keep it to five or six steps
When you define the funnel, resist the urge to add more detail than you can use. Five or six well chosen steps describe almost any purchase decision at a resolution a founder can act on.
The guide's reasoning is worth repeating. More steps feel more precise but are worse in practice. Each additional step adds noise, divides your volume into groups too small to read, and adds more to maintain. A twelve-step funnel is not twice as informative as a six-step one. It is harder to read and slower to trust.
If one step turns out to be complicated inside, you can break that single step down later, once you know it is the problem. That is a much better order of work than starting with the detail and hoping the important part shows itself. The steps themselves should be states a person reaches, written as in how to define funnel steps.
A short test for your first choice
Before you commit, put the candidate through four questions. If the answer to any is no, reconsider.
- Does it end in the outcome that matters most to the business?
- Is it a single sequence, one path a person actually walks from start to finish?
- Can it be described in five or six steps without losing the main decision?
- Will anybody look at it regularly, in a rhythm, rather than once?
The last question matters more than it appears. A funnel that is looked at on a set rhythm becomes part of how the business makes decisions. A funnel that is looked at when someone remembers becomes a chart.
Decide what it is measuring before you build it
The guide's checklist includes a line that is easy to skip: decide what outcome this funnel is ultimately measuring. Write it as one sentence. It sounds trivial, and it prevents a quiet drift in which the funnel gradually comes to measure whatever is easiest to count.
A clear outcome also makes the other decisions easier. It tells you which segments matter, because you know what you are trying to explain. It tells you the time window, because you know which transition you are watching. And it tells you when the funnel has done its job.
Expect the funnel to cross team lines
Choosing the funnel nearest the money has a side effect that is easy to miss. The path from a stranger to a payment passes through marketing, sales and product, and usually more than one tool. So the first funnel you define will cross internal boundaries, and that is a feature, not a flaw.
It means that for the first time somebody is looking at the whole path. The steps at the boundaries are frequently the worst performing ones, precisely because they have never had an owner. Do not narrow the funnel to fit one team's territory just to make ownership simpler. A funnel drawn to match the org chart answers a question the customer never asked.
Signs you picked the wrong first funnel
- Nobody can say what outcome it measures.
- It ends well before the money, so improving it would not clearly change revenue.
- It has grown past five or six steps, because several paths were folded together.
- Nobody has looked at it in weeks.
Any of these is a reason to cut it back, not to add more. Return to the single sequence that ends nearest the money and start again from there.
What to do before you change anything
Once the first funnel is defined and the events are in place, resist the urge to start improving on day two. Look at it for a month first. You need a baseline and a sense of ordinary variation, or you will react to a normal Tuesday.
This is also the right moment to make sure the funnel is counted properly. Confirm that each step counts only people who completed the earlier ones, as explained in funnel cohort analysis, and walk the whole thing yourself to check that your own journey appears in order.
The first month's data is the only clean read you will ever get of what the funnel does when nobody is interfering with it. That makes it valuable, and a good reason not to rush.
When to add a second funnel
The guide gives a clear condition: add a second only when the first has changed a decision. Until then, the first is not finished. It has not yet earned its place, and a second would divide attention before the first has paid for it.
| Situation | What to do |
|---|---|
| The first funnel has not yet led to any change | Keep working on it, do not add another |
| The first funnel has changed a decision | It has earned its keep, so a second is reasonable |
| You feel pressure to map everything | Remember that attention is the scarce thing |
The same discipline appears throughout the guide: do the small thing properly, let it prove itself, then extend. It protects you from a common failure in which a founder builds an impressive set of instruments and never uses any of them.
Where the first funnel will take you next
With one funnel in place, the next jobs follow naturally. Find the step that loses the most people, which is the subject of how to find where a funnel loses people. Then work out which of the three causes you are dealing with. You will do all of that more easily on a single funnel you trust than on several you half trust.
Frequently asked questions
Which funnel should I start with?
The one that ends in the outcome that matters most, which is usually a customer paying. Start with the sequence that ends nearest the money.
Why not map every path at once?
Because instrumentation is easy and attention is not. Many funnels that nobody maintains give you less than one that people look at in a regular rhythm.
How many steps should it have?
Five or six. More steps add noise, divide your volume into groups too small to read and add maintenance. You can break down a single complex step later.
When do I add a second funnel?
Once the first has changed a decision. Not before.
The full method, with the checklists, is in The Funnel Playbook, a free 30-page playbook.