ROAS looks simple: revenue divided by ad spend. In real ecommerce operations, it is not simple at all. A campaign showing 500% ROAS may be profitable, or it may be harvesting brand demand, remarketing to existing customers, selling clearance inventory, ignoring refunds, or reading the wrong conversion value. A ROAS calculator should not announce whether an account is good. It should put revenue, margin, refunds, shipping subsidy, platform fees, payment fees, discounts, inventory, and cash flow into the same decision frame.
Start with one rule: calculate the business break-even ROAS before judging the ad-platform ROAS. Platform ROAS is conversion value divided by ad spend inside Google Ads, Meta, or another ad system. Operating ROAS comes from orders, contribution margin, fulfillment cost, refunds, taxes, payment cost, and inventory risk. Both are useful, but they answer different questions. Platform ROAS helps evaluate traffic and learning. Profit ROAS helps decide whether the company should keep spending.
1. Define which ROAS you are reading
Use at least three levels. Platform ROAS is reported conversion value divided by ad spend. It depends on pixels, APIs, attribution windows, and value settings. Store ROAS is Shopify or backend order revenue divided by ad spend; it is closer to real orders but may still ignore refunds, shipping subsidies, and payment costs. Here the contribution multiple means contribution before advertising divided by ad spend, not after-ad net profit. State that definition before asking whether advertising and fixed costs are covered.
If you only use platform ROAS, attribution can mislead you. Retargeting, brand search, and existing-customer campaigns usually look better. Cold acquisition, creative testing, and new product exploration usually look worse. Low ROAS does not automatically mean bad spend, and high ROAS does not automatically mean scale. First identify the job of the campaign: create demand, capture demand, recover customers, or clear inventory. Different jobs need different ROAS thresholds.
2. Calculate break-even ROAS from contribution margin
The basic formula is break-even ROAS = 1 / contribution margin rate. Contribution margin is not product gross margin. It should deduct product cost, payment fees, platform transaction fees, average discount, expected refund loss, packaging, fulfillment, shipping subsidy, and support cost. Keep the original hypothetical inputs: a pre-discount product reference of $100, product cost $35, processing $4, fulfillment $8, shipping subsidy $7, discount $6 and expected refund loss $4. Contribution before ads is $36; the break-even ratio depends on whether the numerator is before or after the discount. The comparison below separates those bases.
Break-even should not always equal target ROAS. If repeat purchase, subscriptions, accessories, or email lifecycle revenue are proven, first-order ROAS can be slightly lower. If the product is one-time, refund-heavy, slow-moving, or cash-intensive, the target should sit above break-even. Build three lines into the calculator: break-even, healthy, and scale-ready. On the gross-reference basis in this example, 278% is break-even; 350% and 450% are additional hypothetical healthy/scaling lines, not universal standards. Recalculate when the revenue basis changes. This prevents the team from overreacting to one 310% day or killing every cold test at 230%.
Align the revenue basis before comparing ratios
This is one hypothetical order in a single currency, excluding customer taxes and shipping charges. The 100 is a pre-discount reference, not post-discount product revenue. The discount of 6 leaves revenue of 94. Other modeled costs are 35 + 4 + 8 + 7 + 4 = 58, leaving contribution before ads of 36. The last 4 is an expected refund allowance, not an actual refund.
| Line | Amount |
|---|---|
| Pre-discount product reference | 100.00 |
| Discount, deducted once | 6.00 |
| Post-discount product revenue before actual refunds | 94.00 |
| Other modeled costs including the refund allowance | 58.00 |
| Contribution before advertising | 36.00 |
| Hypothetical ad spend | 30.00 |
| Contribution after ads, before fixed overhead and tax | 6.00 |
On the post-discount revenue basis, break-even ROAS is 94 / 36 = 2.61×. On the gross-reference basis it is 100 / 36 = 2.78×, about 278%. Those thresholds use different numerators; do not compare a post-discount platform ratio directly with the 278% reference.
With ad spend of 30, revenue ROAS is 94 / 30 = 3.13×, while contribution before ads divided by ad spend is 36 / 30 = 1.20×. Only 6 remains after advertising. A contribution multiple of 1× covers ads, not fixed overhead or net profit; the two ratios do not share one threshold.
Revenue of 94 already deducts the discount; do not subtract 6 again. When actual refunds are reflected in revenue, reconcile or replace the allowance of 4 rather than mechanically deducting the same refund twice. Platform conversion value may use revenue or a profit-related value; inspect the actual setting before carrying consistent inputs into the tool.
References: Shopify sales definitions · Google Ads conversion values
3. Google Target ROAS is a bidding constraint, not your profit model
Google Ads documentation explains that Target ROAS uses Smart Bidding to predict conversion value for an auction and adjust bids toward the return goal you set. It requires conversion values, and some campaign types need recent conversion volume. Google also notes that setting a target too high may limit traffic. Operationally, Target ROAS is a bidding constraint, not a profit statement. If you set 500%, the system tries to reach that ratio inside the conversion value it can see. It does not know your real refund rate, support cost, inventory pressure, or cash-flow limit unless those values are passed correctly.
Do not treat Target ROAS as “higher is always better.” Too high a target can restrict volume, reduce learning, and leave the algorithm with only the easiest conversions. Too low a target may generate more orders while weakening profit. Use the calculator to define the business target, then compare it with recent actual ROAS, conversion volume, budget utilization, new-customer share, and product mix before changing bids. A drastic target change is often more dangerous than the target itself.
4. Review ROAS with attribution, refunds, and inventory
A weekly ROAS meeting should not report only account-level ROAS. Split by cold acquisition versus remarketing, brand versus non-brand, new customer versus returning customer, market, and product group. Add AOV, contribution margin, refund rate, and inventory cover. A campaign with high ROAS but low inventory can create stockouts and support pressure. A campaign with lower ROAS but high-quality new customers may deserve more testing if email and repeat purchase are proven.
Reconcile platform revenue with backend revenue. Ad platforms may overstate or understate value because of attribution windows, duplicate events, currency conversion, modeled conversions, or bad purchase triggers. GA4 and ad accounts can also disagree because event-scoped and item-scoped metrics are not interchangeable. The meeting should not become a fight about which tool is perfectly right. Define the decision source: which number is used for budget, which cost model is used for profit, and which early signals are used for creative decisions.
5. Inputs a practical ROAS calculator needs
- Ad spend by channel, campaign, market, and date.
- Order revenue, with clear handling of tax, shipping, discounts, and refunds.
- Product cost by SKU or realistic blended cost.
- Payment, Shopify, third-party gateway, and platform transaction fees.
- Fulfillment cost: packaging, pick-pack, warehousing, freight, and exceptions.
- Refund and support assumptions: return rate, return shipping, chargebacks, and service cost.
- Repeat assumptions: 30/60/90-day repeat, email revenue, subscription, or accessory attach rate.
The calculator should output current ROAS, break-even ROAS, acceptable CPA, and ad cost per order. A stronger version also asks: how much discount can we afford, what happens if refunds rise by three points, and whether a 10% AOV lift changes the bidding room. That is a better operating question than “what ROAS is good?”
6. When to scale budget
Scale only when four conditions hold. First, tracking is trustworthy and purchase is not duplicated or missing. Second, ROAS is above the healthy line, not barely at break-even. Third, order quality is stable: refunds, support, fulfillment, and inventory can absorb the volume. Fourth, the result is not only brand demand or remarketing. Do not jump budget after one good day. Review seven-day trend, conversion volume, learning state, and inventory, then adjust in small steps.
Low ROAS also does not always mean pause. Diagnose cold-start learning, creative testing, attribution delay, page friction, pricing, payment, inventory, and market mismatch first. Ads bring users to the page; they cannot compensate for weak product economics, trust, or fulfillment. A useful ROAS review should produce more than “increase or cut budget.” It should create actions for landing page, pricing, market split, SKU pause, creative, event QA, email, and inventory.