Google Ads Bidding Change: Why Your ROAS Dropped

Short answer: On August 17, 2026, Google changed how Target CPA and Target ROAS bidding works for campaigns that are “Limited by budget”. These campaigns used to beat their targets, sometimes by a lot. Now Google bids so results land closer to the target you set. The rollout finished on August 27, 2026. If your ROAS fell or your CPA rose since mid-August, check whether your target is far below what the campaign was actually delivering.

This is one of the biggest Google Ads bidding changes of 2026, and it affects a lot of eCommerce and lead generation accounts. Below I explain what changed, how to find the campaigns that are affected, and the options you have to fix performance.

What Google changed on August 17, 2026

Google first announced the update in early July 2026, as Search Engine Roundtable reported. According to Google’s own FAQ, the change started rolling out globally on August 17, 2026 and was complete on August 27, 2026.

The old behaviour worked like this. When a campaign had a target but not enough budget to spend more, Smart Bidding often found cheaper conversions than the target asked for. A campaign with a 300% Target ROAS could quietly deliver 600%. That looked great, but Google says it made results hard to predict: when you raised the budget, you could not know what performance to expect.

The new behaviour is simple. A budget-limited campaign now bids toward the target you actually set. Google’s own example: if your Target CPA is $10 but you have recently been getting conversions for $5, the campaign will deliver closer to a $10 CPA unless you change the target.

In other words, the target is now the main control for efficiency, and the budget is the control for volume. That is how most of us always assumed it worked.

Which campaigns are affected (and which are not)

Per Google’s FAQ, the change applies only when both of these are true: the campaign uses a target-based strategy, and it is limited by budget.

Maximize conversions and Maximize conversion value without a target are not part of the change. Google presents them as alternatives you can switch to. If you need a refresher on how each strategy works, read my guide to Google Ads bidding strategies.

Why your ROAS dropped after the bidding change

If performance fell after mid-August, the most likely reason is a gap between your target and your real results. Many advertisers set a low Target ROAS (or a high Target CPA) months ago to give the campaign room, then forgot about it. While the campaign was budget-limited, Google ignored that loose target and delivered better results anyway.

After August 17, Google started taking the loose target seriously. It bid more aggressively, won more expensive auctions and spent the same budget at a lower return. Nothing was “broken”. The campaign simply did what the target told it to do.

This hits eCommerce Shopping and Performance Max campaigns hardest, because many stores keep budgets tight for cash-flow reasons and run with targets well below their real ROAS. If you are not sure how ROAS relates to your actual profit, my post on ROAS vs MER vs CAC explains which number to trust.

How to find the affected campaigns in your account

1. Filter by bid strategy. List every campaign using Target CPA or Target ROAS, including portfolio strategies.
2. Check the status column. Look for “Limited by budget”, now or over recent weeks. A campaign that was only occasionally limited can still be affected.
3. Compare target vs actual. Put the target next to the actual CPA or ROAS for the 30 to 60 days before August 17. A big gap means a big risk.
4. Compare before and after. Look at conversions, conversion value, CPA and ROAS for a few weeks before August 17 and the weeks after August 27.
5. Open Recommendations. Google says it shows target recommendations for budget-limited campaigns using target-based bidding. Campaigns with fewer than 7 conversions do not get a recommended target.
6. Prioritise by spend. As Optmyzr points out, fix the campaigns where spend and the target gap are both large first, not just the ones with the biggest gap.

If your conversion tracking is not reliable, fix that before you touch targets. A target based on wrong conversion values will be wrong after any change. My Google Ads audit checklist covers what to check first.

Your options: five ways to respond

There is no single right fix. The best option depends on what the target was meant to do and what the business can afford. These are the main choices, drawing on Google’s guidance and the advice published by Search Engine Journal and Optmyzr.

1. Set the target to match recent actual results. This is Google’s own suggestion. If you were getting 600% ROAS on a 300% target, raise the target toward 600% to keep the same efficiency. Use performance from before August 17.
2. Set the target from your unit economics. Work out the ROAS or CPA you can actually afford, based on margin and customer value, and use that instead of history.
3. Keep the target and add budget. If the low target was deliberate because you want more volume, the campaign now spends in a more predictable way. Adding budget is safer than it used to be.
4. Adjust in steps. Large target changes can push a campaign back into learning. Moving in smaller steps over a couple of weeks is usually calmer.
5. Switch to a strategy without a target. Maximize conversion value (for eCommerce) or Maximize conversions (for leads), with no target set, is outside this change. It can make sense for a budget-limited campaign where you simply want the most value from a fixed spend. Watch efficiency closely, because there is no target holding it.

What I recommend for eCommerce Shopping and Performance Max

For online stores, I would work through it in this order:

• Confirm the tracking. Make sure purchase values are correct and not double-counted. My guide to Google Ads conversion tracking with GA4 and GTM walks through this.
• Decide what the target is for. Is it the minimum ROAS you can live with, or just a number someone set long ago? That answer decides whether you raise it or keep it.
• Do not simply keep raising the target if volume collapses. A very strict target makes Smart Bidding skip auctions. If revenue drops and a higher target does not bring it back, test a strategy without a target on that campaign instead.
• Judge by conversion cycles, not days. Give each change at least one full conversion cycle before deciding. For most stores that means one to two weeks.
• Keep checking. Targets drift as prices, margins and seasons change. Review them monthly, not once.

Shopping and Performance Max also depend on a clean product feed, so if results are still weak after fixing the bidding, look at feed quality and campaign structure. My post on Google Ads for eCommerce and Shopify stores covers that setup.

Mistakes to avoid after the change

• Changing everything at once. Change one campaign or one portfolio at a time so you can see what worked.
• Trusting old forecasts. Google says Performance Planner forecasts were updated for the new behaviour and stabilised after the rollout finished. Use forecasts made after August 27.
• Forgetting portfolios and shared budgets. If a campaign sits in a portfolio strategy, you must change the target at the portfolio level.
• Blaming the algorithm for a tracking problem. If conversion values are wrong, no bid strategy will fix it.

Frequently asked questions

Need help fixing your Google Ads after the bidding change?

If your Shopping, Performance Max or Search campaigns have not been the same since mid-August, I can review your targets, budgets and tracking and tell you what to change. See my Google Ads management service, or request a free ads audit.

Sources

• Google Ads Help: FAQ about changes to target-based bid strategies
• Search Engine Roundtable: Google Ads changes with bidding for campaigns limited by budget
• Search Engine Journal: Google is ending target overperformance
• Optmyzr: Google’s August 17 bidding change, what advertisers need to do