Most Marketing Agencies Optimize for Spend. The Best Ones Optimize for Profit.

Most performance marketing agencies are good at spending money. The best ones are good at making it.

That distinction sounds obvious until you sit across the table from a brand doing $500K a month on Shopify, running a 4x reported ROAS in the ad account, and quietly bleeding cash. Revenue is up. The dashboard looks healthy. The agency is celebrating. And the business is in worse shape than it was six months ago.

This happens because most agencies optimize for platform metrics. They report on ROAS, CPM, CTR, and attributed revenue. They build decks around those numbers. Their incentives are tied to those numbers. And none of those numbers tell you whether the business is actually profitable.

The agencies worth working with think like operators. They care about contribution margin. They care about cash flow timing. They care about which revenue is actually worth acquiring. And they build their strategies around business economics, not dashboard vanity.

ROAS Is Incomplete

ROAS tells you how much attributed revenue came back per dollar of ad spend. That is useful as one input. It is dangerous as a north star.

Here is what ROAS does not account for: shipping costs that vary by geography, fuel surcharges that spike without warning, returns and damage rates that vary by product and distance, marketplace fees that eat 25 to 40 percent of the sale, financing delays that stretch cash conversion cycles by weeks, and operational overhead that scales with order volume regardless of margin.

A brand can report a 4x ROAS in Meta and still lose money on every new customer acquired. The ad account does not know your COGS. It does not know your freight bill. It does not know that the customer it just acquired lives 3,000 miles from your warehouse and the product will touch ten trucks before it arrives.

When diesel hits $6.38 a gallon and fuel surcharges are showing up as line items on every freight invoice, the gap between platform-reported performance and actual business profitability widens fast. A $250 fuel surcharge on a single order does not exist in any dashboard your agency sends you. But it exists on your P&L. And it compounds across hundreds of orders per month.

Marketing Efficiency Ratio matters more for operator-level decisions because it captures total spend against total revenue. But even MER is incomplete without contribution margin analysis layered underneath. The question is never just "did the ads work." The question is "did the business make money."

A good ad account can still create a weak business. I have seen it happen repeatedly across hundreds of accounts managing $250M+ in aggregate spend. The numbers look right in the platform. The P&L tells a different story entirely.

Revenue Quality Matters More Than Revenue Volume

Not all revenue is equal. A dollar of revenue from a direct customer who found you through your website, purchased at full margin, and entered your CRM is fundamentally different from a dollar of revenue through a marketplace.

Consider what marketplace revenue actually costs. The platform owns the customer relationship. You cannot email them. You cannot retarget them. You cannot build a brand relationship that generates repeat purchases. The average order value is typically lower because marketplace shoppers buy single units rather than bundling. A customer who might purchase a $2,400 multi-item order on your site buys a single $1,900 piece on the marketplace instead. That $500 differential multiplied across hundreds of orders per month is revenue you are permanently giving away.

Fulfillment complexity increases because marketplace orders ship to every corner of the country without geographic targeting control. You cannot choose which states to prioritize. You cannot adjust shipping pricing dynamically. The algorithm sends your product wherever demand exists, and you absorb the logistics cost regardless of whether it makes economic sense.

And payouts are delayed. Some marketplaces now hold funds until delivery confirmation, then add a 14-day waiting period, then batch payouts biweekly. A product sold today might not generate receivable cash for six weeks. When you are using revenue from previous sales to fund the next manufacturing order, that cash flow delay is not an accounting inconvenience. It is a growth constraint that forces you into financing arrangements you should not need.

One brand I work with generates $260K a month through a major marketplace. Total cost to sell on that platform last month: $60K in direct fees. Add COGS at 30 percent, import tariffs at 25 to 50 percent depending on the product, and the margin collapses to a number that barely justifies the operational effort. Meanwhile, their direct Shopify channel runs at three times the margin on half the revenue.

The agencies that understand this do not just optimize for revenue growth. They optimize for revenue quality. They ask which channels produce customers worth acquiring. They measure customer lifetime value by acquisition source. And they build strategies that prioritize ownership over volume.

Performance Marketing Should Behave Like Operations

Most media buyers treat advertising as an isolated function. Set the budget. Pick the audience. Launch the creative. Report on results. Repeat.

The best media buyers treat advertising as an extension of operations. They factor in shipping zones when choosing which states to target. They adjust CAC targets by geography because a customer in a neighboring state has a fundamentally different contribution margin than a customer 3,000 miles away. They turn off expensive regions entirely when the freight economics do not justify the acquisition cost.

This is not theoretical. I have a client where California is their third-highest spend state by ad dollars allocated and their worst-performing state by actual profitability. Not by ROAS, which looks fine in the platform. By what they take home after shipping, fuel surcharges, damage risk from multi-touch freight handling, and returns.

The solution is not to turn California off entirely if it drives meaningful revenue. The solution is to set a geography-adjusted MER target. A 25x MER in California may be equivalent to a 15x MER locally once you account for the operational cost differential. Separate the campaigns. Set different efficiency thresholds. Monitor them independently. Run remarketing-only strategies for the most expensive states so your acquisition cost stays low enough to absorb the fulfillment premium.

State-level targeting is one of the most underutilized levers in performance marketing. Most agencies do not even look at state-level performance. They optimize at the campaign level, the ad set level, the creative level. They never zoom out to ask: are we spending our most aggressive dollars acquiring the most expensive customers to fulfill?

Marketing should integrate with finance and logistics. The media buyer should know what it costs to fulfill an order in every region you serve. They should know which states have the highest return rates. They should know where fuel surcharges are eating margin. And they should build campaigns that reflect those realities instead of optimizing nationally and hoping the economics work out.

The Agency Model Is Broken

The standard agency pricing model is percentage of spend. The more you spend, the more the agency makes. This creates an obvious misalignment: the agency is financially incentivized to increase your budget regardless of whether that increase produces profitable growth.

I do not like percentage of spend. It incentivizes an agency to spend, spend, spend. There are guardrails, sure. You can only spend more if you hit certain targets. But the fundamental incentive structure is wrong. The agency gets paid more when you spend more, not when you profit more.

Most agencies are aware of this problem. Few have solved it. The ones who have moved toward performance-aligned models produce dramatically better results. Not because the talent is different, but because the incentives are aligned.

A performance model works like this: define a shared metric, usually MER or contribution margin above a baseline. Set a target. When the business exceeds that target, the agency earns more. When it misses, the agency earns less. Both sides have skin in the game.

This changes behavior in specific, measurable ways. The media buyer who gets paid more when the business is more profitable will proactively cut underperforming channels. They will tell you to pull $30K from a marketplace channel that generates five percent margin and reallocate it to direct acquisition where the economics are three times better. They will not recommend spending more just to justify their retainer.

When an agency operates on a performance model, their team checks accounts on Saturday nights. Not because someone told them to, but because finding an optimization that saves the business money earns them money too. That discretionary effort, the 10pm check-in, the weekend deep dive into click-to-purchase timing, the proactive recommendation to pause an underperforming channel, is the difference between a vendor and a partner.

The shared upside and shared downside creates real alignment. When you have a high-profit month, they do well. When things are down, they feel it too. It is a non-equity partnership that treats the agency as a growth operator rather than an outsourced media buyer.

The Future Belongs to Operators

The performance marketing industry is slowly waking up to the gap between platform metrics and business outcomes. The agencies that survive the next wave of consolidation will be the ones that can sit inside the business economics, not outside them.

This means understanding first-time customer P&L at a channel level. Understanding geography-level profitability. Understanding cash flow timing and how marketplace payment terms affect growth capacity. Understanding that the job is not to manage ad spend. The job is to drive profitable growth.

The best marketers think like operators. The future of performance marketing is efficiency, systems, and profitability. Not scale at any cost.

A profitable business with controlled growth will always beat fast growth with weak economics.

Keep reading

Pieces I've written on related topics that pair well with this one:

• What Scaling Past $1M/Mo on Meta Taught Me About the Algorithm, Lessons from scaling Meta ad spend past $1M/month, creative structure, algorithm behavior, attribution at scale,
• How to Structure a Performance Marketing Agency for Profit, Not Just Revenue, Revenue tells you how big you are. Margin tells you if the business works.
• How to Use Spend Pacing as a Diagnostic Tool, Not Just a Budget Control, Spend pacing is more than budget control. Here's how to use it to diagnose creative fatigue, auction pressure, and delivery problems before they hit R…
• What a Healthy Paid Media Account Looks Like at $10K, $50K, and $200K Monthly Spend, The account structure that works at $10K/month will break at $200K. Here's what healthy paid media looks like at each eCommerce spend stage.
• Geography Is One of the Most Overlooked Levers in Performance Marketing, Most brands optimize advertising nationally while ignoring geography-level profitability.