The pattern I see every week
A founder pings me with the same sentence, reworded: "We need more traffic."
Sometimes it's "we need to crack LinkedIn," sometimes "we need to figure out paid," sometimes "we need a podcast." Underneath every version is the same instinct — growth is flat, so push more demand into the top of the funnel.
Three months later the channel is humming, the pipeline number on the dashboard is bigger, and revenue hasn't moved. The founder is now paying for two channels that don't convert instead of one.
Goldratt applies to marketing
Eliyahu Goldratt wrote The Goal in 1984 about factory throughput. The central insight: every system has exactly one constraint. Throughput is determined by the constraint, not the sum of all the parts. Until you identify and exploit the constraint, adding capacity anywhere else does nothing.
Marketing has the same property. Every funnel has exactly one bottleneck. If it's activation — the percentage of signups who reach the "aha moment" — then adding traffic accelerates nothing. You are filling a bucket with a hole in it, and paying more per gallon.
The mistake isn't thinking about channels. Channels matter. The mistake is defaulting to channel-first instead of bottleneck-first.
The 10-minute reverse-funnel math
You don't need an analytics platform to run this. You need a spreadsheet and your last 90 days of numbers.
Start from your revenue target. Work backward:
- Revenue target ÷ average contract value = customers needed
- Customers needed ÷ close rate = SQLs needed
- SQLs needed ÷ SQL rate = MQLs needed
- MQLs needed ÷ page conversion rate = traffic needed
- Traffic needed vs. traffic you have = gap
Most founders find the math explodes before it reaches traffic. The close rate is fine. The SQL-to-MQL conversion is fine. But activation — the percentage of trial signups who reach first value — is 14%. Wes Bush's threshold is 30%. You are below the floor.
When that happens, adding traffic is the wrong answer. The right answer is to fix activation and rerun the math.
The five most common misdiagnosed bottlenecks
Positioning. The product is right. The message is for the wrong person. Traffic lands on a page, doesn't recognize itself in the copy, and bounces. No amount of targeting fixes this.
Activation. Users sign up and don't reach value. The product's "aha moment" is too far from the front door. The fix is product work, not marketing work.
Page conversion. You have the right traffic and the right positioning, but the landing page doesn't convert. This is the one bottleneck that is purely marketing's problem.
Retention. Users activate, then churn before the renewal. Acquisition is working. The product isn't holding. More acquisition makes this worse faster.
ICP mismatch. You built for one customer and are marketing to another. The funnel looks broken at every stage because none of the stages were designed for the people coming through.
The diagnostic you can run today
Run the free AEO audit → to see where AI engines place you in your category. If they're sending you the wrong traffic — wrong intent, wrong buyer type — that's an ICP mismatch signal showing up in the earliest possible stage of your funnel.
The ten-minute reverse-funnel math tells you where to focus. The AEO audit tells you whether the traffic you're attracting is the traffic worth having.
Frequently asked
How do I know if my growth problem is a channel problem or a funnel problem?
Run the reverse-funnel math. Start from revenue target, work backward through close rate, SQL rate, MQL rate, and traffic. If the math blows up at activation or page conversion before it blows up at traffic, it's a funnel problem.
What's the activation threshold below which I should stop acquisition spend?
Wes Bush's rule: if activation is under 30%, no amount of acquisition spend fixes it. Your product isn't converting the users you already have.
What are the five most common misdiagnosed growth bottlenecks?
Positioning (wrong message for the right audience), activation (users who sign up don't reach value), page conversion (traffic doesn't convert to trials), retention (users churn before they pay), and ICP mismatch (you're acquiring the wrong users entirely).