How Much Should You Spend on Ads? ROAS, Margins & the Next Dollar
A practical advertising budget guide explaining ROAS versus profit, contribution margins, campaign measurement, and responsible scaling—with an illustrative scenario.
Adapted and substantially expanded from Northlane Creative’s Advice Column #2, originally emailed September 7, 2026. This article is an educational guide, not a client-results case study.
“How much should we spend on ads?” sounds like a budget question, but it is really a measurement and economics question. A campaign that creates sales can still lose money. A low-volume campaign can also reveal a profitable path worth expanding. The right starting number is the amount you can responsibly test while learning what happens after someone clicks.
The useful distinction: return on ad spend describes attributed revenue relative to ad cost. It is not the same as net profit or a complete return-on-investment calculation.
ROAS is a ratio—not a profit statement
Return on ad spend (ROAS) is attributed revenue divided by advertising spend. Spend $1,000 and record $3,000 in attributed sales, and the campaign’s reported ROAS is 3.0×, also written as 300%. That says nothing by itself about the cost of goods, labor, fulfillment, refunds, processing fees, or overhead.

To estimate whether that campaign is economically sustainable, examine the contribution margin left after the variable costs that rise with each sale. The simple break-even ROAS, when ad cost is the only remaining acquisition expense, is approximately 1 ÷ contribution margin. At a 25% contribution margin, that is 4×; at 40%, it is 2.5×. Real acquisition economics may also include agencies, creative work, or other costs.
| Illustrative scenario | Attributed revenue | Contribution at stated margin | Less $1,000 ad spend | Contribution after ad spend* |
|---|---|---|---|---|
| 30% margin / 3× ROAS | $3,000 | $900 | −$1,000 | −$100 |
| 40% margin / 3× ROAS | $3,000 | $1,200 | −$1,000 | +$200 |
| 50% margin / 3× ROAS | $3,000 | $1,500 | −$1,000 | +$500 |
*Illustrative math, not client data. Figures are before fixed overhead and other acquisition expenses not included in the margin assumption. Platform attribution may differ from actual incremental sales.
Before choosing a monthly budget, decide what you need to learn
A budget is useful when it can answer a question. Are people searching for this service? Does the ad attract the right visitors? Are they contacting you? Does the resulting sale justify the cost? A low-cost market may produce enough data with a modest daily test. A high-competition market may require more budget or a narrower objective to collect meaningful information.
Start with one offer, one geographic market where relevant, a clear conversion action, and a defined review date. Do not evaluate success solely on clicks. Quality matters: a dozen relevant inquiries may be more useful than hundreds of visits with no fit.
Follow the full customer path, not just the ad dashboard
If customers drop out at checkout, more traffic will send more customers through a confusing checkout. If a contact form sends leads into an unattended inbox, higher ad spend simply exposes that process failure. Examine your pages on mobile and test the handoff all the way to the person responsible for responding.

| Stage | What to measure | Warning sign |
|---|---|---|
| Landing page | Engaged visits, page clarity, relevant CTA | People leave before they understand the offer |
| Consideration | Product view, form start, key interactions | Visitors need information that the page hides |
| Action | Completed lead, checkout, purchase | The journey breaks or asks too much |
| Follow-up | Qualified responses, booked meetings, sales | Leads exist but nobody owns the next step |
Scale where there is evidence—not just room to spend
Our original email discussed a hypothetical $20,000 increase in monthly spend at 3× ROAS, equivalent to $60,000 in additional attributed monthly revenue if that return held. Across twelve months, the arithmetic is $720,000 in attributed revenue. But this is a scenario, not a projection: higher spending can face diminishing returns, limited demand, different audience quality, and more expensive bids.

An incremental test provides stronger evidence. Increase one well-performing campaign in controlled steps while watching conversion quality, acquisition cost, and contribution after expenses. In parallel, test a different audience, category, or offer when the existing one approaches saturation.
What you should receive from an agency or internal team
Ask for a report that translates spend into decisions, not a long presentation of vanity metrics. A useful monthly review includes: what was spent, what actions occurred, how conversions were measured, what changed, what the team learned, where the next dollar should go, and which website or follow-up improvements could outperform additional advertising.
- Budget versus actual spend, by campaign and objective.
- Conversion tracking definition and testing status.
- Qualified inquiry or order quality, where data is available.
- Revenue and contribution assumptions, not ROAS in isolation.
- Actions taken, experiments completed, next actions and owners.

Editor’s note: Automated bidding systems such as Google Ads Target ROAS use recorded conversion values; their targets should be chosen from valid economics and reliable tracking, not copied from another company’s dashboard.
Resources worth exploring
These references provide implementation details and additional reading; they are offered for research rather than as paid endorsements.
- Google Ads: About Target ROAS — the platform’s explanation of value-based bidding.
- GA4 recommended events — recommended ecommerce and lead-generation measurement events.
- Google Search Central — improve the pages visitors reach before buying more traffic.
Your practical next steps
- Write down revenue, variable cost, contribution margin, and desired customer acquisition cost.
- Pick a campaign goal and confirm conversion tracking with a test event.
- Review the entire landing-page-to-sale process on a phone.
- Run one bounded experiment with a decision date.
- Expand only when the incremental results support it.
Brett Luke · Director, Northlane Creative
Brett writes practical guides about small-business marketing, websites, branding, sales systems, and the decisions behind client projects.
Keep reading. Keep learning.
Explore another guide, compare an approach, or follow along as we document the work behind real projects.
Looking for the next topic? Visit our Articles & Guides library for more practical reading.