Target ROAS (Return on Ad Spend) is one of Google Ads’ most powerful smart bidding strategies. You tell Google what ROAS you want, and the algorithm adjusts bids on every auction to try to hit that target.
But Target ROAS depends entirely on the quality of the conversion data you send to Google. If your tracking is broken, incomplete, or sending wrong revenue values, the algorithm will make poor bidding decisions — and your campaigns will underperform no matter how strong your ads or products are.
How Target ROAS Bidding Actually Works
When you set a Target ROAS of 400%, you are telling Google: for every dollar I spend, bring me four dollars in revenue. Google’s algorithm then looks at each individual auction, estimates the probability that this specific user will convert and at what value, and bids accordingly.
The key word is estimates. Google builds these estimates from your historical conversion data. If that data is wrong, the estimates are wrong, and every bid the algorithm makes is wrong.
Why Tracking Accuracy Matters More for ROAS Than for CPA
With Target CPA, Google tries to hit a fixed cost per conversion. It does not need to know the value of each conversion — only whether a conversion happened.
With Target ROAS, Google needs to know three things:
- That a conversion happened
- What revenue value it generated
- How that revenue compares to what was spent to acquire it
A conversion value error of 20% does not create a 20% ROAS error. It compounds across thousands of auctions and learning cycles, often producing actual ROAS that is 40–60% off your target.
Three Ways Bad Tracking Destroys Target ROAS Performance
1. Missing Conversions
If your purchase event only fires on 80% of orders, Google sees 20% less revenue than you actually generate. It thinks your campaigns are producing 20% less ROAS than they are. Smart bidding responds by either reducing bids (if ROAS looks too low) or misallocating budget across campaigns that appear to be performing differently than they actually are.
2. Wrong Revenue Values
If your purchase event sends the product list price instead of the actual amount paid (after discounts, or inclusive of shipping depending on your setup), Google’s ROAS calculations are wrong from the start. A 15% average discount means Google thinks it is generating 15% more revenue per order than it actually is. Campaigns will overbid, actual ROAS will consistently fall short of target, and you will spend more than you should chasing a target the data is artificially inflating.
3. Duplicate Conversions
If your Shopify thank you page fires the purchase event twice — a common issue when customers refresh the order confirmation page — Google sees double the revenue. The algorithm thinks campaigns are performing twice as well as they are, raises bids, and your actual ROAS ends up at roughly half what Google Ads reports. This is one of the most damaging and hardest-to-diagnose tracking errors for ROAS campaigns.
How to Check If Your Conversion Data Is Accurate
Compare these three numbers over the same 30-day period:
- Shopify total orders
- GA4 purchase events (found in Reports → Monetisation → Ecommerce purchases)
- Google Ads conversions (found in Campaigns → Columns → Conversions)
All three should be within 5–10% of each other. If Google Ads shows significantly more conversions than Shopify orders, you have duplicate tracking. If significantly fewer, you are losing purchase data.
Also compare total revenue: pull the total conversion value from Google Ads and compare it to your Shopify revenue for the same period. More than a 10% gap means your value parameters are wrong.
How Much Tracking Error Is Acceptable for Target ROAS?
For Target CPA bidding, up to 15% error in conversion count is workable. For Target ROAS, you want under 5% error in both conversion count and conversion value. The higher your ROAS target, the less tolerance you have for errors. A small tracking gap has a proportionally larger effect on a 600% ROAS target than on a 200% one — because the algorithm is making finer bid adjustments to hit a tighter margin.
Setting Up Target ROAS Without Killing Your Campaigns
Once your tracking is clean, follow these steps when switching to Target ROAS:
- Have at least 50 conversions in the last 30 days before enabling Target ROAS — smart bidding needs historical data to build estimates from
- Set a realistic initial target — start at your actual average ROAS from the last 30 days, not your ideal target. Increase it gradually by no more than 15% every 4 weeks.
- Do not change anything during the learning phase — any change to bids, budgets, or conversion actions resets learning. Allow 4–6 weeks for the algorithm to stabilise.
- Use conversion value rules — if certain products or customer segments are worth more to your business, use Google’s Conversion Value Rules to weight them so the algorithm optimises for actual business value, not just reported revenue.
Fix Your Tracking Before You Switch Bidding Strategies
Switching to Target ROAS with broken tracking is one of the most common reasons Shopify store owners see their ad performance collapse after making what should be an improvement. The algorithm is powerful — but it optimises for exactly what you measure. If what you measure is wrong, the optimisation goes in the wrong direction at scale.
We audit your full Google Ads conversion tracking, GA4 purchase events, and Shopify setup to verify that every data point you send to Google is accurate. Book your free Shopify tracking audit here before you change your bidding strategy.