Change target ROAS at the wrong time and you can turn a normal dip into a self-inflicted slump that drags on for a week or two. We see it all the time: you look at yesterday’s numbers, they are ugly, and the temptation is to grab the biggest knob in the room, your target. The problem is that Google Smart Bidding treats target shifts like new instructions, so the more often you change course, the longer it takes to settle.

Below is how we think about tROAS and tCPA in real accounts. Not in theory. In the messy, day-to-day reality where tracking lags, promos end midweek, and leadership wants a clean answer by Friday.

Change target ROAS? Get clear on what “learning” really means

Smart Bidding is not fragile, but it is sensitive to big signals. When you change a target, switch bid strategies, or shove budget up fast, Google often needs time to re-balance bids across auctions. That is why things can look jumpy right after an edit.

Google’s own guidance is pretty straightforward: make smaller target moves and give the system room to respond. They specifically call out gradual adjustments, often around 20%, and waiting about a week between changes in their Smart Bidding gradual adjustment recommendations.

One detail that gets missed: the learning period is not only about days on the calendar. It is about how many conversions the system can observe. If you are only getting a handful of purchases or qualified leads per week, the “settling” part takes longer. So yes, a target update can feel like it broke the campaign, but what is usually happening is the algorithm is trying to re-earn confidence with limited data.

Change target ROAS with a simple guardrail: the 20% rule

If you want the safest way to adjust without whiplash, treat your target like a dial, not a light switch.

When you raise tROAS or lower tCPA, you are telling Google to be pickier about auctions. Delivery can tighten, sometimes fast. When you lower tROAS or raise tCPA, you are giving it more freedom to bid, which can restore volume, but it can also loosen efficiency if you overdo it.

A lot of practitioners stick to the 20% approach because it keeps the algorithm from overreacting and keeps your read on performance clean.

• Step 1: Move the target by 10% to 20% max.

• Step 2: Wait 5 to 7 days, or one full conversion cycle for your business.

• Step 3: Review results using the same lens every time: volume, CPA or ROAS, and downstream quality.

If you are trying to get from A to C, you still go A to B to C. It feels slower in the moment. In practice, it is usually faster than thrashing the account and never getting a stable read.

When to change target ROAS or adjust tCPA (and you will not regret it)

You do not need a crisis to justify a target change. You just need a real reason tied to the business and enough data to support the move.

1. Your current target does not match reality. If the campaign has been beating your target consistently, that can be a sign you are being too conservative and leaving volume on the table. On the flip side, if it is missing target for weeks, the target may be unrealistic for the inventory you are trying to buy. Google has also adjusted how target-based strategies behave when budgets are tight, which is why keeping targets aligned with what you truly want matters.

2. Your unit economics changed. If margins tightened, shipping got more expensive, AOV dropped, or your team updated what counts as a qualified lead, your old target might be pointing you at the wrong kind of growth. Your target is not a scorecard. It is a constraint. It should match what you can afford.

3. Volume dropped hard and stays down. If impressions, clicks, or conversions are down 30%+ for multiple weeks, a too-strict target is a common culprit. In that case, loosening the target can help you re-enter more auctions and rebuild conversion signal.

When to wait (even if you really want to change target ROAS)

Most accounts do not get in trouble because nobody ever touched tROAS or tCPA. They get in trouble because targets were edited too often, too early, and without enough conversion signal to justify the confidence.

• You are in learning. If the campaign status is “Learning,” do not stack a bunch of edits on top. Every extra change adds noise.

• You just switched strategies or launched a big restructure. The first 2 to 3 weeks are often messy. You are watching the system explore, not operate in a steady state.

• You have volatility that is normal for your business. If your conversion lag is three days, yesterday’s CPA spike is not a verdict. Check day-of-week patterns, promos, and tracking before you assume the target is wrong.

• You intentionally loosened the target to find new pockets of demand. Give it time to test. Tightening too fast often shuts down the exact exploration you asked it to do.

A quick pre-flight checklist before you adjust tCPA

Before you touch the target, make sure you are not “fixing” the bid strategy when the actual problem is upstream. We do this kind of audit-first pass in almost every account because it saves money and headaches.

If you want us to sanity-check this with you, start with our Google Ads management service page so you can see how we work and what we typically look at before recommending a target shift.

How we change targets without breaking delivery

At PPC Boost, we treat targets as one part of a control system. Measurement, feed quality, creative, landing pages, and budget allocation all push on performance too. If you change five things at once, you cannot tell what helped, what hurt, and what simply needed time.

• One variable at a time: If we change the target, we usually do not change creative or structure in the same breath.

• Small moves: 10% to 20% shifts, then a proper wait window.

• Clean evaluation windows: We account for conversion lag so we are not grading the system too early.

• Profit-first targets: We align tROAS or tCPA to what your business can afford, not what looks “impressive” inside the UI.

If you want to see how this plays out in a lead gen account where strategy and cleanup mattered as much as bidding, take a look at the Pandora CloudCover paid search case study.

And if you want more of our practical Google Ads thinking, browse the rest of the PPC Boost blog for additional guides and account frameworks.

FAQ: Changing tROAS and tCPA without guesswork

How long should you wait after you change target ROAS?
In most cases, wait at least 5 to 7 days and ideally a full conversion cycle for your business. Google also notes you should allow 1 to 2 conversion cycles after a Target ROAS change before judging results in their Target ROAS best practices.

Does changing tCPA always restart learning?
Not always with a loud “Learning” label, but large or frequent changes often trigger learning-like volatility. Treat target edits like meaningful inputs, especially when conversion volume is limited.

Should you change targets when CPA spikes for a few days?
Usually no. First check conversion lag, day-of-week behavior, tracking issues, and whether a promo just ended. If things normalize over the next week, a target change would have been an expensive overreaction.

What if your campaign is not spending its full budget?
A too-strict tROAS or too-low tCPA is a common reason delivery throttles. In that case, loosening the target in small steps is often more effective than simply raising budget.

Conclusion

Changing tROAS and tCPA is a powerful lever, but it works best when you treat it like a measured adjustment, not a panic response. Keep moves small, give the smart bidding learning period space, and tie targets back to unit economics and real outcomes. If you want a hands-on second opinion on whether you should change target ROAS or wait, reach out through our main PPC Boost homepage and we will walk through the data with you.