Should I use Target ROAS or Target CPA?
Use Target CPA when every conversion is worth about the same and your goal is volume at a predictable cost. Use Target ROAS when order values or margins vary and your goal is profitable revenue.
Target CPA suits customer acquisition, product launches and lead generation where you know what a lead is worth. Target ROAS suits mature ecommerce campaigns with consistent conversion volume and fairly stable order values. It also needs accurate revenue data, ideally adjusted for margin, or it will chase the wrong orders.
Google's advice is to bid toward the right conversion goal on Target CPA first, then report values for at least four weeks or one to two conversion cycles before moving to value-based bidding. When you set the ROAS target, base it on recent conversion value per cost, leaving out the latest days still waiting on conversion lag. Your margin tells you the target you need. History tells you the target the campaign can hit. Start at history and walk toward margin, using the feed or CRM to send better values along the way.
Sources: About Target ROAS bidding - Google Ads Help, Pick the right bid strategy - Google Ads Help
Is Maximize Clicks a good bid strategy?
Maximize Clicks is fine for a short list of jobs, like a launch, a new message test or gathering early data, and wrong for almost everything else. It spends your budget on the cheapest clicks it can find and doesn't care whether they convert.
The mistake I see most in audits is Maximize Clicks with a CPA goal sitting next to it. Google lets you save that setup and the status still says eligible, but the two instructions contradict each other. The auction buys cheap traffic and the CPA number is decorative. Google's own help page says Maximize Clicks isn't a good choice when your goal is a specific cost per conversion.
To check your account, add the bid strategy type column to the campaigns view and read the strategy name, not the status. Any campaign on Maximize Clicks that you judge by CPA should move to Maximize Conversions or Target CPA once it has the volume. That won't replace search term cleanup, so keep adding negatives too. Our piece on CPC strategies goes deeper.
Sources: Determine a bid strategy based on your goals - Google Ads Help
How do I test a new bid strategy without hurting performance?
Run it as a Google Ads experiment on part of the traffic instead of switching the whole campaign. The old strategy keeps running as a control, so you get a real comparison instead of a before and after muddied by seasonality.
Google's recommended setup: pick a campaign that has been running a while and has at least 30 conversions in the last 30 days, or 50 for value-based tests. Split it 50/50, change only the bid strategy, and turn on experiment sync. Let it ramp for two weeks or three conversion cycles, whichever is longer, then run at least 30 more days and ignore the ramp period when you read the results.
If a campaign is too small for that, an experiment won't tell you much. Either pool campaigns with the same economics, or switch the strategy and leave it alone for 14 days before judging it. Whichever route you take, write down the success metric before you start. Pick it afterward and you'll always find one that agrees with whatever happened.
Sources: Test your bid strategy - Google Ads Help, About Value-based bidding using campaign experiments for Search and Shopping - Google Ads Help
How often should I change my Target CPA or ROAS?
Change a target when the business goal changes or the campaign has clearly settled above or below it over a couple of weeks, not every time one day looks bad.
There's a distinction that trips people up. Google says changing a target doesn't trigger a new learning status, and Smart Bidding adjusts bids right away. Switching the strategy itself, or making big budget and targeting edits, is what sends a campaign back into learning. So small, deliberate target moves are fine. Flipping between strategies is not.
Our rhythm: after a strategy switch, hands off for 14 days. The first week often looks worse while the model explores, and pulling back on day six is how accounts stay stuck. After that, review cost per conversion and volume over the last two weeks, leaving out days still waiting on conversion lag, then move the target and repeat. If you want a preview first, the bid strategy simulator shows roughly what a different target would do to cost and volume.
Sources: How our bidding algorithms learn - Google Ads Help, Duration of the learning period for campaigns and what affects it - Google Ads Help
What bid strategy should I use on a small budget?
On a small budget, start with Maximize Conversions on as few campaigns as possible, so every conversion you get lands in one place the model can learn from.
The trap with small accounts is splitting a thin budget across many campaigns, each with three or four conversions a month. At that volume Smart Bidding is guessing, and adding a Target CPA on top usually makes it stop spending. Consolidate first. If a campaign still sits well under about 30 conversions in 30 days, consider optimizing toward an earlier step that still shows real intent, such as a qualified phone call, while you keep reporting on the final outcome.
Keep the money on your highest-intent terms and add negatives every week, because a small budget can't afford to learn on junk. Our minimum for management is $3,000 a month in ad spend. Below that, our small-budget guide is a better starting point than hiring anyone.
Does broad match work with Smart Bidding?
Broad match works well with Smart Bidding when the account has healthy conversion data, and poorly when it doesn't. They're built to run together: broad finds the queries and the bid strategy decides what each one is worth.
The old rule of never touching broad is gone, but the catch is volume. If a campaign isn't getting 30 to 50 conversions in 30 days, the algorithm can't separate good queries from bad ones and broad will wander. Broad on manual bids is worse still, and it's one of the problems we flag most often in audits.
Our sequence: get conversion tracking right, reach roughly a conversion a day, then open up match types. Once that's stable, add conversion values that reflect profit or lead quality, add negatives for the obvious misses, and give it 30 to 90 days before deciding. The advertisers who stick with it through that window tend to do fine. See our notes on the broad match update and the conversion tracking audit.
