Your Agency Earns More When You Spend More. Guess What They Recommend.

Look at the last twelve months of proposals from your agency. Separate the ones recommending you increase spend from the ones recommending you decrease it. The ratio will tell you everything you need to know about the incentive structure underneath.

If your agency charges a percentage of ad spend — sometimes called ”% of media,” sometimes buried inside “performance-based fees” — their revenue goes up every time your spend goes up. It goes up whether that additional spend produces profitable orders or not. It goes up whether the new campaign is a good idea or a bad one.

Percentage-of-spend is not a fringe pricing model; it’s the default across the industry. Public breakdowns of agency fee structures all describe it as one of the two or three dominant models — typically quoted at 10–20% of monthly ad spend, often bundled with flat retainers starting around $1,500 a month, and in some cases rising to 25% or more of spend at smaller account sizes. None of these structures pay the agency to reduce your spend. All of them pay the agency to grow it.

This is not a moral failing on the part of your agency. It’s a structural one. People respond to the incentives they’re paid on. If spend up means paycheck up, then spend up is what gets recommended.

How the conflict actually shows up

The conflict rarely looks like a cartoon villain. It shows up in the soft places:

A new campaign launched “to test incremental reach” that happens to add $4,000 a month in spend. An audience expansion that “surfaces new shoppers” but mostly captures people who were going to buy from you anyway. A display layer “for brand momentum” that your own analytics can’t attribute a single purchase to. A “premium placement” bid strategy that increases your cost-per-click by 20% in exchange for a softer click.

Each decision, on its own, is defensible. You’ll get a slide explaining it. But the direction of drift is always the same. And you don’t have the time, or the specialist knowledge, to audit every one of them.

The tell

The quickest tell is this: ask your agency to reduce your spend by 30% next month, holding revenue flat. Watch what happens.

If the answer comes back confident and specific — which campaigns they’d pause, which keywords they’d trim, which SKUs they’d deprioritise — you’re probably in decent hands. If the answer is a paragraph about “full-funnel impact” and “momentum” and “the risk of losing learnings,” you’ve just surfaced the incentive problem.

A genuinely aligned operator can always tell you where the least-productive spend is. They have to, because they think about it all day.

How kPixies priced around this

We don’t earn a percentage of your ad spend. We don’t earn a monthly retainer. We don’t earn a setup fee. We don’t earn more when you spend more. Below a 5X ad-attributable ROAS, we invoice $0 — our time, the tracking work, the campaign builds, the reporting, all of it runs at our cost until the account clears the floor. Above 5X, our fee scales 1:1 with ROAS and caps at 16% of attributable revenue: at 5X we earn 5%, at 10X we earn 10%, at 16X+ we earn 16% and no more.

Below 5X, we work for free to get you above it. Above 5X, we earn a share of the revenue that clears the bar — and only that share, capped.

Read that carefully. The structure means that every dollar of ad spend that doesn’t produce at least 5X in attributable revenue is a dollar working against our own paycheck. We are, financially, the most motivated person in the room to kill unproductive spend.

What this does to the day-to-day

It changes what we recommend. It changes what we pause. It changes what we refuse to launch in the first place. If a campaign idea doesn’t plausibly clear 5X inside a reasonable learning window, we won’t waste your budget or our time on it. We’ll tell you why, and we’ll move on.

It also changes how we scale. We don’t push spend for the sake of it. When the math supports more volume, more volume gets unlocked. When it doesn’t, we hold. That discipline is the entire reason the 5X floor holds over time.

Who this works for

The retailers this model fits are the ones who’ve been burned by the opposite. Usually two or three agency relationships deep. Usually doing between $1M and $5M in online revenue. Usually with a margin profile that can’t absorb another year of “let’s test more spend.”

If that’s you, the form below takes about two minutes. It’s a call request, not a commitment. We reply within one business day. When we decline, it’s usually because the margin, product mix, or store-level performance doesn’t give us a real path to 5X — and we’ll tell you that directly.

We’d rather tell you no now than invoice you into next year.