What to check before you touch anything
- How long, and how consistently, has actual ROAS been beating target? A one-off overshoot from a couple of big orders is different from sustained overperformance across several weeks in a row.
- Is the campaign currently budget-constrained ("Limited by budget") on top of this consistent overperformance? If so, first check whether scaling can be solved simply by raising the budget, with no target change at all.
- What share of Impression Share is currently being lost to rank? A persistently high share shows there's untapped volume in the auction the campaign could capture with a looser target.
- Is there enough accumulated data at the current tROAS to draw a confident conclusion? A one-off, not-yet-settled overshoot could be statistical noise rather than a signal to scale.
- Is the business prepared to accept a temporary ROAS dip in exchange for revenue/volume growth? This is a business decision, not a technical one, and it's worth stating explicitly before making any changes, rather than judging a lowered target as "a mistake" after the fact.
- What does the competitive landscape look like specifically in the share of the auction the campaign is currently missing out on. Before lowering tROAS, check Auction Insights for your core keywords: weak competition (low overlap rate, few players in the auction) means the missed volume is likely genuinely open space, and lowering the target will probably capture it safely. If, instead, rank-lost share is rising alongside stronger competitors (a growing overlap/impression share for them), the volume you're missing is competitors' increasingly expensive territory, and lowering tROAS could lead not to cheap scaling but to a sharp rise in acquisition cost with no proportional sales growth. It's also worth manually checking the search results for your key terms — are competitors strong on offer and messaging specifically where you're planning to expand, or is this space where competitors are actually weaker than your offer?
Possible approaches
- The standard approach when ready to scale is to lower tROAS in small steps (similar to the step-based approach used for tCPA, 10–15% at a time), with a 1–2 week pause between changes to let things settle, rather than dropping the target sharply straight to the volume you want.
- If the campaign is budget-constrained while consistently beating target, it's often more logical to try raising the budget first, leaving tROAS unchanged, and only move to lowering the target if that alone isn't enough to reach the volume you need.
- Google will typically flag either a budget increase or a lower target ROAS directly, with an estimated lift in conversion value — worth checking that estimate before adjusting parameters manually.
- You can test lowering tROAS through a Campaign Experiment on a slice of traffic, comparing the volume and efficiency of the new and old target before rolling the change out to the whole campaign.
- If the business isn't ready for a permanent ROAS reduction but wants to test scaling, you can treat it as a temporary, time-boxed experiment (say, during a seasonal demand spike) rather than a permanent target change.
- If your competitive analysis shows the missed volume is mostly open space with weak competition, scaling by lowering tROAS is low-risk and worth doing with confidence. If, instead, the missed volume sits in territory held by strengthening or already-strong competitors, it's smarter to either limit scaling to the segments where competition is weaker, or pair the target reduction with a review of your offer/messaging for the part of the auction where competitors are genuinely stronger.