The Slow Death of "We Just Need More Ads"

It's 11pm. You're in the ads dashboard again. Six months ago your ROAS was 3.2; tonight it's 2.4, and it has been a slow, almost invisible drift the entire way down. The next move feels obvious, test new creative, refresh the audience, maybe nudge the budget to hit the number.

It isn't.

That instinct is the first step onto a treadmill that speeds up the longer you run on it. Let me name the loop, show you the exact math that makes it a trap, and give you a two-question test to find out whether you're on it, or whether, for your store, more ads really is the right call (sometimes it is).

The treadmill, named

Here's the cycle almost every scaling store runs, whether they've noticed it or not:

1. Ads work. Returns are good, so you scale spend.
2. As spend grows, CAC creeps up, the cheapest, warmest buyers get bought first.
3. Rising CAC compresses your margin.
4. To protect total profit, you scale spend more to push more volume through.
5. Which pushes CAC up again. Back to step 2.

Each lap feels like progress, because revenue genuinely grows. But your profit per order is quietly thinning, and the only lever you're pulling is the one that makes the next lap harder. That's the slow death in the title: not a crash, a grind.

Why the spiral feels so rational

The trap is sneaky precisely because every individual step is locally sensible. More spend made sense when returns were strong. Testing creative is good practice. Pushing volume protects this month's revenue. None of the decisions are dumb in isolation.

The problem isn't any one step. The problem is the loop, and the reason the loop tightens has a name from media buying: diminishing returns. As you scale a channel, the next dollar reaches a colder, more saturated slice of the audience than the last dollar did. Your marginal CAC rises, and your incremental ROAS, the return on the additional spend, which is the only number that should ever drive a scaling decision, falls well below the comfortable blended ROAS you see on the dashboard (Recast). When cold-prospecting frequency climbs past 3–4, you're showing the same people the same ads, paying more to convince fewer (Stella).

The number nobody puts on the dashboard

Here's the metric that reframes the whole thing. Write it down:

What a visitor is worth to you = your AOV × your conversion rate.

Run it with realistic benchmarks. DTC average order value sits around $75 (Triple Whale's 2025 median was $74.12) and a typical store converts around 2% (Triple Whale). So each visitor you buy is worth about $1.50 to you.

Now scale spend. Diminishing returns mean the cost to acquire each additional cold visitor keeps rising, but their value is stuck at $1.50, because your conversion rate didn't change. Every new dollar buys visitors closer to (and eventually past) the line where they cost more than they're worth. You're not buying growth anymore. You're buying volume at a worse and worse margin.

Now look at the other lever. Lift conversion from 2% to 2.5%, half a percentage point, and visitor value jumps from $1.50 to about $1.88. That's a 25% increase in what every click can profitably cost you, across your entire ad budget, this month and every month after, without spending a single dollar more on ads.

That's the whole game in one line: ad spend is a lever you have to keep pulling harder. Conversion rate is a multiplier that pulls every dollar for you, permanently.

Name the waste

There's a blunter way to see it. At a 2% conversion rate, for every $100 you spend on ads, about $98 buys visits that don't convert. That's the waste, spend that produced no sale.

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. So doubling spend on a leaky funnel doesn't just double revenue's potential, it doubles the waste, at an accelerating rate. Fixing conversion is the only move that changes the ratio itself. (I run the full worked example on this in Article 14: scaling spend on a leaky funnel.)

"But sometimes the answer really is more ads"

Yes, and any honest version of this argument has to admit it, because the opposite advice ("never scale, always optimize") is just as wrong.

There's a real scenario where scaling spend is exactly the right move: when your conversion rate is already at or above benchmark, and acquisition is your genuine constraint. If your page converts cold traffic well (say 2.5%+) and your incremental ROAS on new spend is still comfortably above breakeven, then you're not on a treadmill, you're under-fueling a healthy engine, and you should feed it. Holding back spend there is leaving money on the table.

The treadmill is the other case: scaling spend while conversion sits below benchmark. Same action, opposite outcome. The entire skill is knowing which one you're in, so here's how to tell.

The diagnostic that tells them apart

Two questions, about ten minutes:

1. Is your conversion rate at or above benchmark? Cold-heavy pages above ~2.5% lean healthy; well below ~2% lean treadmill (Shopify benchmarks).
2. Does the next increment of spend still pay? Look at incremental (not blended) ROAS as you add budget, and watch cold-prospecting frequency. If marginal ROAS holds and frequency stays under ~3, scaling is working. If marginal ROAS craters and frequency climbs past 3–4, you've hit saturation, more spend is the treadmill.

Healthy engine on both? Spend. Sub-benchmark CR or collapsing marginal returns? More ads will only make you poorer, faster.

What actually breaks the spiral

You break it by switching levers, from the click math to the post-click math. Fixing what happens after the click raises that visitor-value multiplier, which lowers your effective CAC at every spend level at once. It's the exact inverse of the treadmill: instead of each step making the next harder, each point of conversion makes every future ad dollar work harder. It compounds in your favor. (This is why I argue your post-click experience matters more than your ad creative, Article 12.)

Run it on your own numbers

Before your next budget increase, do this:

• Pull your real AOV and conversion rate, and multiply them, that's your visitor value.
• Pull your cost per visit from cold prospecting. If it's near or above your visitor value, more spend is buying losses.
• Model a +0.5pp CR lift and recompute visitor value. That percentage gain applies to all your spend, current and future.

"But I can't lift conversion overnight"

Fair, and true. Conversion work isn't instant the way a budget slider is. But here's the comparison that actually matters: a budget increase produces a temporary bump that decays as the new traffic cools, and you have to keep paying for it every single month. A conversion fix produces a permanent multiplier you pay for once. One is rented growth; the other is owned. Even if the conversion fix takes a few weeks, it then pays back on every dollar of spend from that point on, including all the spend you were going to make anyway. The slow lever wins because it compounds while the fast lever decays.

Why the blended number hides all of this

One reason the treadmill is so hard to see: the dashboard shows you blended ROAS, which averages your warm, cheap conversions together with your cold, expensive ones. The blended number can look "okay" while your marginal spend, the newest, coldest dollars, is already underwater. By the time blended ROAS drops enough to alarm you, you've been losing money on the margin for months. This is why I keep pushing operators toward incremental ROAS: it's the only number that tells you whether the next dollar is worth spending, which is the only decision scaling actually requires.

Key takeaways

• The "we just need more ads" reflex is a treadmill: spend → CAC up → margin down → spend more.
• It tightens because of diminishing returns, the marginal dollar is always the weakest, and frequency >3–4 signals saturation.
• Visitor value = AOV × CR. Ad spend pulls one dollar at a time; conversion is a multiplier on every dollar.
• A +0.5pp CR lift ≈ +25% in what every click can profitably cost, permanently, at zero added ad spend.
• More ads is right when CR ≥ benchmark and marginal ROAS holds. It's the treadmill only when CR is sub-benchmark.

The bottom line

More ads is a lever you pull harder every month for less. Conversion is a multiplier you set once and benefit from on every dollar after. If your ROAS has been drifting down, the question isn't "what's the next creative?", it's "is the next dollar of spend even worth more than the last one?"