Why Scaling Ad Spend on a Leaky Funnel Makes You Poorer

Here's the math most founders avoid running, because the answer is uncomfortable: at $100k/month in ad spend and a 2% conversion rate, doubling your budget doesn't double your revenue. It doubles your waste. Let me show you, with numbers, and then show you the one move that does the opposite.

The reflex

When growth stalls, the instinct is to spend more on ads. It feels like the lever most under your control, open the dashboard, raise the budget, push more volume. Every founder reaches for it, because it's fast and it's visible. So let's actually do the arithmetic on what it buys, because the arithmetic is not on the reflex's side.

The worked example

Start with a realistic setup, anchored to benchmarks: $100,000/month in ad spend, a 2% conversion rate (the low end of Shopify's 2–3% norm), and an $80 average order value (between Triple Whale's $74 median and Shopify's ~$85). Say that spend buys you 50,000 visits.

Look at the two moves side by side:

• Doubling spend cost you an extra $100,000 to earn an extra $80,000. The marginal return on that new spend is 0.8, you spent a dollar to make eighty cents. You didn't grow; you bought a bigger loss.
• Fixing CR from 2% to 2.5% cost you $0 extra and earned an extra $20,000, and it lifts the return on every future dollar you spend, forever.

*And it's actually worse than the table shows: that second $100k reaches a colder, more saturated audience (diminishing returns), so its real conversion rate drifts below 2%. The marginal dollar is always the weakest, which means doubling spend on a leaky funnel doesn't just double the waste, it buys waste at an accelerating rate.

The waste, named

At a 2% conversion rate, for every $100 you spend, about $98 buys visits that don't convert. That's the leak, money spent on visitors who left without buying. When you double the budget, you double both halves: the converting and the non-converting. Except the new money skews toward the non-converting side, because it's colder.

Fixing conversion is the only move that changes the ratio. Lift CR to 2.5% and you convert more of every single dollar, so waste-per-dollar drops across your entire budget at once, not just on the new spend, on all of it. That's the difference between the two levers in one sentence: more spend buys more of the same leaky outcome; better conversion improves the outcome on money you're already spending.

The structural point

Your conversion rate is the ceiling on your ad-spend efficiency. Below that ceiling, more spend just buys more traffic that hits the same leaky page, at a worse and worse marginal rate, because each additional dollar reaches a colder slice of the audience.

A 0.5-point CR lift (2% → 2.5%) is a 25% relative improvement that's well within reach of focused conversion work. Making your next $100k of ad spend convert as well as your first $100k is, by contrast, nearly impossible, that's simply not how ad auctions work. You can improve the page. You cannot improve the auction. So the lever with real slack is the one almost everyone ignores.

An honest caveat

Two things, so this stays credible. First, real profitability depends on your margins and COGS, these figures illustrate ad efficiency, not net profit, and your actual breakeven depends on your unit economics. Second, a CR lift isn't guaranteed; conversion work is a bet, not a switch. But the relative logic isn't arguable: a fixed conversion rate caps what any amount of spend can return, and extra spend always hits diminishing returns. You can debate the exact numbers; you can't debate the direction.

"But you have to spend to test ads and find winners"

True, and I won't dismiss it, scaling spend funds creative testing, and finding winning creative genuinely matters. So here's the honest distinction (the same one as Article 12's "page then creative").

When your CR is at or above benchmark, scaling spend to discover winning creative is sound, you've got a funnel that banks what the ads bring, so more spend funds real learning. But when CR is below benchmark, you're funding those tests through a broken funnel. A winning ad can still post a losing ROAS, a losing ad might've won behind a better page, and you can't trust the signal you paid for. You're not buying learning; you're buying noise at scale. Fix CR to benchmark first, then scale and test, same budget, clean signal, real winners.

The reframe

Fix the page first, then scale. Not because conversion is more exciting than ads, it isn't, but because the math leaves no other order. Scaling spend on a sub-benchmark funnel is the one move that reliably makes a growing brand poorer: more revenue on the top line, less profit on the bottom. The founder feels busy and the dashboard looks bigger, while the bank account quietly shrinks.

What "fix the page first" looks like here

Not a six-month redesign, that's its own mistake (Article 17). Diagnostic-led conversion work: find the one structural constraint capping CR (Article 7) and fix that. At this stage it's days, not months, and crucially, it's the prerequisite that makes the spend you're already committed to start paying. You don't pause acquisition to do it; you fix the funnel the acquisition is pouring into.

Run it on your own numbers

Before your next budget increase, do this five-minute exercise:

1. Write your real monthly spend, CR, and AOV.
2. Compute current orders and revenue.
3. Now model doubling spend at the same CR, note how marginal ROAS compares to your breakeven.
4. Then model the same spend with +0.5pp CR, note the revenue gain at zero added cost.
5. Compare the two. The honest math almost always favors the page.

Why this is the hardest advice to follow

I'll be honest about why founders resist this, even when the math is unarguable. Spending more on ads feels like growth, the dashboard gets bigger, the team feels busy, there's a visible action you took today. Fixing conversion feels slower and less heroic; there's no budget slider to drag, no instant number to watch climb. So the psychologically satisfying move and the financially correct move point in opposite directions. That gap is exactly why so many growing brands scale themselves into thinner and thinner margins, not because they don't understand the math, but because the wrong move feels better in the moment. Naming that bias is half of resisting it.

The compounding case

Here's the part that makes conversion the better bet over almost any horizon. A budget increase is a cost you pay every month for a result that decays as the traffic cools. A conversion fix is a cost you pay once for a multiplier that applies to every dollar of spend from then on, this month and every month after, including all the spend you were going to make anyway. One is rented; one is owned. Over a year, a single +0.5pp lift quietly improves the return on your entire ad budget, twelve times over. That's why, dollar for dollar, the page beats the budget slider on any timeline longer than a few weeks.

One more honest point: do both, in order

None of this means "never scale spend." Acquisition matters; you have to feed the machine, and a great funnel with no traffic sells nothing. The argument is purely about order. Fix the funnel to benchmark, then scale into it, and you get the best of both: more traffic and more of it converting. Scaling first and fixing later just means you spent a pile of money learning, at full price, what the math would have told you for free. Sequence is the entire lesson: page, then spend. Get that order right and the same ad budget that was making you poorer starts making you money.

Key takeaways

• At a fixed CR, doubling spend doubles the waste, and the marginal dollar is the weakest, so it's worse than linear.
• At 2% CR, ~$98 of every $100 buys non-converting visits; only fixing CR changes that ratio.
• Your CR is the ceiling on ad-spend efficiency, below it, more spend buys more waste.
• A +0.5pp CR lift (≈25% relative) is reachable; making the next $100k convert like the first is not.
• Fix the page first, then scale, scaling a sub-benchmark funnel is how a growing brand gets poorer.

The bottom line

More spend is the most expensive way to grow a leaky funnel, you pay full price for traffic and convert a fraction of it. Conversion is the cheapest, because it makes every dollar you already spend work harder. Run the table on your own numbers before your next budget increase.

--- ---