How Much Should Ecommerce Brands Spend on Paid Ads?

“How much should we spend on ads?” is the question we hear most from ecommerce founders, and it usually arrives with a bad answer attached: a percentage someone read, or a competitor “doing $50K a month on Meta.”

Here’s the uncomfortable truth: there is no universal ecommerce advertising budget. The number falls out of your unit economics and your stage of growth, not off a rule-of-thumb shelf. Two brands with identical revenue can have wildly different correct budgets: one has 75% margins and strong repeat purchases, the other sells a one-time product at 30% margins.

This guide covers how we set a paid advertising budget for DTC brands: why the percentage-of-revenue rule misleads, the unit-economics method that replaces it, how the answer changes by stage, minimum viable spend per platform, and the line items most founders forget.

Why “spend X% of revenue” is the wrong starting point

The percentage-of-revenue rule survives because it’s easy, not because it’s right. Three structural problems:

It anchors your future to your past. Paid ads are an investment in next quarter’s customers, not a tax on last quarter’s. A brand with strong margins and a proven offer might rationally spend far more than any tidy percentage suggests; a larger brand with broken economics should spend close to nothing until the offer is fixed. The percentage rule gives both the same advice.

It ignores margin structure. An ecommerce marketing budget only makes sense relative to contribution margin. A supplements brand with high gross margins can afford a customer acquisition cost that would bankrupt an electronics reseller. Revenue is the wrong denominator; contribution dollars pay for ads.

It says nothing about allocation. Even a correct total wouldn’t tell you how to split it across Meta, Google, TikTok, and creative production, where the actual leverage sits.

The one legitimate use for the percentage heuristic: a sanity check after the real math. Even then, keep the bands qualitative. Early-stage brands commonly run ad spend at a higher share of revenue than mature ones; they’re buying growth and data, not defending a base. If the bottom-up math lands at 60% of revenue, recheck your assumptions. Use it as a smoke alarm, never as a blueprint.

The unit-economics method: build the budget from the bottom up

The correct ecommerce ad spend calculation has three inputs and one multiplication.

Step 1: Find your allowable CAC

Allowable CAC is the most you can pay to acquire a new customer while hitting your profit goals. It comes from contribution margin and lifetime value, not platform benchmarks.

Start with first-order contribution: AOV minus cost of goods, shipping, fulfillment, payment processing, and returns, the cash one new customer generates on day one. Then decide how much of your customers’ future value you’re willing to spend today. Strong repeat behavior lets you pay more than first-order contribution and recover it over months; a one-time product means allowable CAC has to fit inside the first order.

The payback window is the judgment call. Break-even on first order is the conservative play. Spending against 60 or 90-day LTV is reasonable when your repeat data is real and your cash position can carry the gap. Spending against 12-month LTV is how funded brands grow fast and how bootstrapped brands run out of money.

The same math in platform language: break-even ROAS equals 1 divided by contribution margin. At an illustrative 40% contribution margin, that’s 1 / 0.4, or 2.5. Below it a campaign buys revenue at a loss, which is how brands celebrate a 2x ROAS while quietly shrinking; above it, you choose how much profit to trade for growth.

Step 2: Decide how many new customers you need

This comes from your growth target, not your ad account. If you know next quarter’s revenue goal and your repeat rate, you can back into how many new customers paid media must deliver each month.

Step 3: Multiply

Allowable CAC times target new customers equals your monthly media budget. That’s the whole formula.

Here’s a worked example with deliberately round, illustrative numbers, not benchmarks. Say your AOV is $100 and your first-order contribution after all variable costs is $40. Repeat purchases bring a typical customer’s 90-day contribution to $80, and you decide to spend up to $60 of that acquiring the customer. That’s your allowable CAC. If your growth plan requires 500 new customers a month, your paid media budget is 500 times $60, or $30,000 a month. Tighten the payback to break-even on first order and allowable CAC drops to $40, so the same customer target costs $20,000.

If you don’t know your contribution margin or repeat rate, you can’t set a budget. That’s not a flaw in the method; it’s the method telling you what homework comes first.

A scoping note: this is the paid acquisition budget. Retention (email, SMS, loyalty) is a separate, smaller line that defends the LTV your allowable CAC leans on. Fund it; keeping a customer is cheaper than the ads that replace one.

Your stage changes the answer

The formula gives you a ceiling. Your stage says how close to it to operate.

Validation stage: budget for signal, not scale

Until you’ve proven paid ads acquire customers at or under allowable CAC, the budget question is: what’s the minimum spend that produces a trustworthy answer? You need enough conversions across enough creative tests to separate signal from noise, and enough runtime to see whether early results hold. The goal of a validation budget is a decision, not revenue. If even minimum viable spend (below) strains the business, you’re not ready for paid yet.

Growth stage: scale while marginal CAC holds

Once acquisition works at your allowable CAC, the budget question inverts from “how much should we spend?” to “why aren’t we spending more?” As long as the marginal customer (the next one, not the average one) still comes in under your allowable CAC, capping spend at a comfortable number leaves profitable growth on the table. Watch marginal cost, not blended averages: blended CAC can look healthy while the last 20% of spend loses money.

Mature stage: spend to the point of diminishing returns

Every account eventually hits the point where marginal CAC meets allowable CAC. That intersection is your budget. Not a percentage, not a round number, the point where the next dollar stops paying for itself. Mature brands manage to that frontier, pushing it outward with better creative and new channels, pulling back when auctions or fatigue move it inward.

Minimum viable budgets: what each platform actually requires

There’s a floor below which a digital advertising budget doesn’t underspend, it malfunctions. Ad platforms run on machine learning that needs conversion volume to optimize delivery. Meta’s guidance points to roughly 50 conversion events per ad set per week to exit learning; Google and TikTok have equivalent learning periods. Feed the system too little data and it never stabilizes; costs stay erratic and teach you nothing.

This is why the classic small-brand mistake, $10 a day across five campaigns, fails so reliably. Each campaign gets a trickle of data and none exits learning. If your CAC is likely around $50, a $10 daily budget means one conversion every five days. No algorithm can optimize on that.

So how much should you start testing with? Enough daily budget, concentrated in one or two campaigns on a single platform, for the algorithm to exit learning at your expected CAC within a few weeks. Fund fewer things properly; a focused Meta ads budget on one prospecting campaign teaches you more in 30 days than a fragmented account does in a year.

Allocating across Meta, Google, and TikTok

Once the total is set, advertising budget allocation across platforms should follow marginal return, not a fixed split. Anyone handing you a “60/30/10” formula is selling simplicity, not performance. The platforms do have distinct jobs for ecommerce, though, and the jobs shape where budget starts.

Meta is the demand creation workhorse. For most DTC brands, Meta (Facebook and Instagram) is where new customers first meet the product, so it typically carries the largest share of spend and most of the creative testing. It finds people who didn’t know they wanted what you sell.

Google captures the demand you and others create. A Google Ads budget for ecommerce works differently: Shopping and search harvest existing intent, and branded search protects the demand your Meta spend generates. Google rarely creates DTC growth on its own, but underfund it and you pay to create demand that competitors collect at the point of search.

TikTok earns budget in creative-led categories. A TikTok ads budget makes the most sense where the product demonstrates well and the category rewards native, fast-moving creative: beauty, food, gadgets, pet products. A pet food brand, for instance, has endless organic-feeling content angles, exactly what the auction rewards. TikTok punishes recycled, polished Meta ads, so only fund it if you’ll make creative for the platform.

Let marketing efficiency ratio (MER) referee the split

From there, reallocate by marginal return: each month, shift budget toward wherever the last dollar performed best. Judge those moves on blended numbers, because platform dashboards each claim credit for the same orders; marketing efficiency ratio (MER), total revenue divided by total ad spend, is the referee that can’t be gamed. The right split is discovered, not declared, and it drifts with auctions, seasons, and creative cycles.

The budget line everyone forgets: creative

Most founders set an ecommerce marketing budget that is 100% media and 0% production, then wonder why performance decays. On paid social, creative is the primary performance lever. Ads fatigue, and accounts that scale keep a steady pipeline of new hooks, angles, and formats to test.

That pipeline costs money: UGC creators, editing, statics, landing page variants. Treat creative production as a standing share of your paid media budget rather than an occasional favor you do the account. A common planning heuristic reserves roughly 10-15% of media spend for production: a starting point to adjust against your testing velocity, not a benchmark. Either way the principle is fixed: a media budget with no creative budget attached is a plan to run the same ads until they stop working. Compare ecommerce advertising costs across brands and the ones that look expensive on production usually look cheap on CAC.

Pacing: when to raise the budget

Raise budget on evidence, not the calendar. A new quarter is no reason to spend more; these are:

• Marginal CAC is comfortably under allowable CAC. You have room before the math breaks.
• Recent creative tests produced new winners. Fresh proven ads absorb budget; tired ones just spend it.
• Performance held through your last increase. Scaling in steps beats doubling overnight, which restarts learning.
• Operations can handle the volume. Inventory, shipping, and support must scale with spend, or ads acquire customers your fulfillment disappoints.

The same logic runs in reverse. When marginal CAC crosses allowable CAC and new creative isn’t fixing it, cutting budget isn’t failure, it’s the system working. The budget breathes with the account.

Deciding how much to spend is also deciding who manages it

Everything above assumes someone watches contribution margins, marginal CAC, learning phases, and creative pipelines every week. That’s a real operating job. Running it yourself at smaller spend levels is often right while validating. Once a percentage-point improvement in CAC outweighs a management fee, specialists become a math decision, not a comfort one. At that point, read our guide on how to choose the right paid media agency, including the questions that expose weak ones on the first call.

However you staff it, hold the budget to the standard we’ve laid out: built from your unit economics, sized to your stage, concentrated enough for signal, allocated by marginal return, and fed by a real creative pipeline. How much to spend on ecommerce ads was never a percentage. It’s the output of a system, and now you have the system.