Stop Chasing a Higher ROAS. Do These 8 Things Instead

Here’s a pattern we see often: a store reports a healthy ROAS in the ad dashboard, yet the bank balance barely moves. Or the opposite: ROAS slips, someone tightens the target, spend collapses, and revenue goes with it.

Both come from treating ROAS as a dial to turn instead of a result to understand. The eight tactics below fix the inputs that produce the number.

What Is a Good ROAS? Start With Break-Even

ROAS (return on ad spend) is the revenue an ad platform attributes to your ads divided by what you spent on them. Earning $5 of revenue for every $1 of ad spend is a ROAS of 5.0×. It says nothing about profit, which is why a “good” ROAS doesn’t exist in the abstract.

A 4:1 return looks great until you remember that product cost, shipping, payment fees, and returns can eat three-quarters of your revenue, leaving a 4:1 ROAS at roughly zero profit. BrightBid’s 2026 ROAS profitability guide makes the same point.

Your break-even ROAS is simply 1 divided by your contribution margin: the share of revenue left after product cost, shipping, fees, and returns, but before ad spend.

Once you know your break-even, every decision below gets easier. You’re no longer chasing a high ratio; you’re hunting for the most profit at a ratio you can afford.

What Changed for ROAS in 2026?

Google’s Performance Max and AI Max and Meta’s Advantage+ campaigns now decide much of who sees your ads and what you pay. That shifts where you can win:

• Bidding is automated, so inputs matter more. The product feed, creative, and conversion signals you provide are the main way you steer the system.
• AI shopping surfaces use the same data. Google’s AI Mode and Gemini draw on your Merchant Center product data, so feed work now pays off in more than one place. (We cover this in our Google Shopping guide.)
• Costs have generally trended up. With auction costs rising on both platforms, wasted spend hurts more than it used to.
• Platform-reported ROAS overlaps. Each platform takes credit for conversions it touched, so dashboards can flatter you. Tactic 7 deals with this.

Tactic 1: Fix the Feed Before the Bids

For Shopping and Performance Max, your product feed is the targeting. Google reads it to decide which searches each product fits, and industry guides such as Search Engine Journal’s Performance Max for ecommerce describe the feed as the foundation of the campaign. Weak data limits both reach and conversion quality.

• Titles: lead with what the product is, then add the details shoppers search by: brand, material, size, color, model.
• Images: clean, high-resolution, product-forward, with a separate image for each color variant.
• Identifiers: real GTINs and brand names help Google match your product to the right queries.
• Price and availability: keep them in sync with your site. Mismatches cause disapprovals and wasted traffic.
• Custom labels: tag products by margin, bestseller status, or season. You’ll use these in Tactic 2.

Tactic 2: Split Campaigns by Margin and Role

Treating a 60%-margin product and a 15%-margin product with the same ROAS target guarantees you’ll overspend on one and starve the other.

• Separate by margin. Use custom labels to group high-margin and low-margin products. Give high-margin items a lower ROAS target to win more volume, and hold low-margin items to a stricter one.
• Separate hero products from the long tail. A handful of bestsellers usually deserve their own budget and learning, instead of competing with a thousand slow movers.
• Separate brand from non-brand. Branded searches convert easily and make blended results look better than they are.
• Move targets in steps. Jumping a target ROAS far above your recent history can make a campaign stop spending. Raise it gradually and watch total revenue, not just the ratio.

Tactic 3: Treat Creative as Your Targeting

With broad automated targeting, your ad itself decides who stops scrolling and who converts. On Meta especially, creative is the biggest lever you still control. Google’s Performance Max also rewards strong assets: images, video, and headlines.

• Test angles, not just variations. A new color on the same message teaches little. Test different promises: price, speed, social proof, problem-solution, founder story.
• Lead with proof. Real reviews, user-generated content, and before-and-after demos usually outperform polished brand imagery in cold audiences.
• Refresh on a schedule. Fatigue quietly lowers ROAS. Plan a steady flow of new creative instead of waiting until performance drops.
• Give Performance Max real assets. Include short videos and several image sizes, grouped into themed asset groups for promotions and seasons.

Tactic 4: Feed the Algorithms Better Data

Automated bidding is only as smart as the conversion data behind it. Three upgrades make a real difference:

• Send purchase value, not just purchase count. Value-based bidding needs accurate order values, ideally excluding tax and shipping, so the system chases bigger baskets.
• Strengthen tracking. Use Meta’s Conversions API alongside the pixel and Google’s enhanced conversions so fewer purchases are lost to browser and consent limits.
• Separate new customers from returning ones. Platforms naturally favor shoppers who are cheap to convert, often people who would have bought anyway. Use new-customer goals (Performance Max supports bidding higher for new customers) to push growth.

If you can, go one step further and feed margin-adjusted value, so algorithms optimize for profit rather than revenue.

Tactic 5: Right-Size Retargeting and Brand Search

Retargeting and brand search produce the prettiest ROAS numbers because they reach people who were already about to buy. That makes them easy to over-fund.

• Cap retargeting. Keep it large enough to recover abandoned carts, but small enough that it isn’t soaking up budget that could bring new customers.
• Check what Performance Max is really doing. Use brand exclusions or review search-term and channel reports so you can see non-brand performance on its own.
• Protect brand terms separately. A small, cheap brand campaign is fine. Just don’t let it hide a weak prospecting engine.

Not sure how to balance all this across platforms? Our guide to splitting budget between Meta and Google walks through it.

Tactic 6: Raise Conversion Rate and Order Value

ROAS is revenue divided by spend. You can lift it without touching the ads by getting more from the traffic you already pay for.

• Fix the product page. Clear pricing, delivery times, return terms, reviews, and fast load speed all lift conversion. See our breakdown of product page elements that lift conversion.
• Reduce checkout friction. Fewer form fields, guest checkout, and the payment methods your customers expect.
• Raise average order value. Bundles, free-shipping thresholds, and “complete the set” suggestions turn the same click into a bigger basket.
• Match the landing page to the ad. If the ad promises a specific product or offer, send people straight to it, not to the homepage.

Tactic 7: Measure With MER and Incrementality

Platform ROAS is useful for tuning campaigns and comparing creative. It’s unreliable as a verdict on the whole business, because every platform claims the sales it touched.

Here’s a simple illustration of the double-counting problem described in this MER vs. ROAS explainer. Say you spend $10,000 across two platforms. One reports $20,000 in revenue and the other $17,500, so the dashboards claim $37,500. But your store only earned $30,000 from those ads. The difference is double-counting, and summing the platform ROAS figures overstates results by 25%.

• Track MER. Marketing efficiency ratio is total revenue divided by total ad spend across every channel. It can’t be double-counted, so if spend rises and MER falls, you know efficiency really dropped.
• Watch new-customer revenue separately. MER blends new and returning buyers, so also track new-customer share and cost per new customer.
• Test incrementality. Pause or reduce a channel in one region or time window and see what happens to total sales. It’s the closest thing to proof of what a channel adds. Retargeting is a good first test.
• Judge by profit. A lower ROAS that delivers more total profit beats a high ROAS on tiny volume.

Tactic 8: Diagnose Performance Across the Funnel

When ROAS drops, you need to know at which stage the problem appears. Looking at campaign results alone isn’t enough to find the source.

• CTR: Are the ad message and creative grabbing attention?
• CPC: What is your traffic costing you?
• Landing page views / clicks: Are people actually reaching the page?
• Add-to-cart rate: Do the product and price meet shoppers’ expectations?
• Checkout rate: Are you losing people during checkout?
• Purchase conversion rate: Is your traffic turning into sales?

Reading these metrics together tells you whether to optimize on the ad side or on the website and purchase journey.

A 30-Day Plan to Improve ROAS

• Week 1: Baseline. Calculate break-even ROAS, audit conversion tracking, and set up a weekly MER and new-customer view.
• Week 2: Feed and structure. Fix titles, images, and identifiers. Add margin and bestseller labels and split campaigns accordingly.
• Week 3: Creative and audience. Launch a new creative round with distinct angles. Cap retargeting and separate brand from non-brand.
• Week 4: Verify and adjust. Run one incrementality test, review MER, and adjust targets in small steps.

Change one major thing at a time and give each change a couple of weeks to settle before you judge it.

ROAS isn’t an optimization target. It’s a performance indicator that results from your optimization work.

Frequently Asked Questions

Sources: BrightBid, Good ROAS for Google Ads in 2026 · Search Engine Journal, Performance Max for Ecommerce · DEV Community, Summing platform ROAS overstates your results