The compounding maths nobody wants to look at

Most growth decisions are made on a feeling about volume and almost none on the arithmetic of conversion. That is a mistake, because the arithmetic is brutal and it does not care how you feel about your ad creative. Once you actually multiply the stages out, the case for fixing joints over buying volume becomes impossible to argue with.

The average machine, stage by stage

Take the published B2B SaaS benchmarks as your reference gauges, not your goals. Visitor-to-lead rates vary widely by channel, from roughly 0.7 percent to 2.2 percent, with SEO traffic around 2.1 percent. From there, roughly 40 percent of leads become an MQL, 39 percent of MQLs an SQL, 40 percent of SQLs an opportunity, and 37 percent of opportunities a close. None of those individual numbers looks alarming. Lined up, they are.

Multiply them through and the end-to-end visitor-to-customer rate lands far below one percent, on the order of a few customers for every ten thousand visitors. That is the average machine, and it is the one most operators are quietly running without ever having done the multiplication.

Why a 20 percent fix beats doubling the budget

Here is the part nobody wants to look at. Because the stages multiply, a proportional lift at any one weak joint flows through every stage that follows it. Lift a single weak joint by 20 percent and you lift the whole end-to-end rate by 20 percent, for the cost of fixing one seam. Double your top-of-funnel spend instead, and you have doubled the input to a machine that still leaks the overwhelming majority of it out of the same holes, at twice the cost.

That is the entire economic argument for the joint over the chamber. New top-of-funnel is the most expensive lever you own because it pays full price for traffic that then runs the same gauntlet of leaks. Fixing the steepest-drop joint is the cheapest, because the visitors are already bought and you are simply keeping more of them.

Finding the binding constraint

The practical move is to stop treating all your stage rates as equally worthy of attention. Lay them out, compare each against its benchmark, and find the one furthest below where it should be. That is your binding constraint, and it is the only place worth spending this week.

• Pull your real rate at every joint, not just top and bottom.
• Set each against the benchmark for that stage.
• Rank the gaps and pick the single steepest drop.
• Fix only that joint, then re-measure before touching anything else.

This is theory-of-constraints applied to growth, and it is unfashionable precisely because it tells you to ignore most of your dashboard most of the time. The benchmarks above are the gauges that let you spot the broken joint. They are emphatically not the target; a machine tuned to merely average rates is a machine quietly bleeding out. Treat them as the floor you measure against, find your worst joint, and go to work on that one thing.