Target CPA vs Maximize Conversions

If you run Google Ads, you've hit this fork: should you use Target CPA or Maximize Conversions as your bidding strategy? They sound similar — both chase conversions — but they behave very differently, and picking the wrong one can quietly burn budget or choke your volume.

This guide explains what each one actually does, when to use which, and the concept that ties them together: marginal CPA.

What Maximize Conversions does

Maximize Conversions has one job: spend your entire daily budget to get as many conversions as possible. It doesn't care what each conversion costs — only that it gets the most it can for the money available.

This means:

• It will always try to spend the full budget
• Your cost per acquisition (CPA) floats — it can spike if the auction gets expensive
• It's volume-first, efficiency-second

Use it when you're budget-constrained and want the most conversions that budget can buy, and you don't have a hard CPA ceiling you must respect.

What Target CPA does

Target CPA (now folded into Maximize Conversions with a target in newer Google Ads UI) flips the priority: get as many conversions as possible while keeping the average cost per conversion near a number you set.

This means:

• It will hold back spend if hitting your target requires it
• Your CPA stays near the target (on average — individual conversions vary)
• It's efficiency-first, volume-second
• It may not spend the full budget if it can't find conversions at your target

Use it when you know your economics — you know the maximum you can pay per conversion and still be profitable — and you want the system to respect that ceiling.

The key difference, in one line

Maximize Conversions spends the budget. Target CPA respects the CPA. One is a spend constraint, the other is an efficiency constraint.

Marginal CPA — the concept that decides which to use

Here's the idea most advertisers miss. Your average CPA and your marginal CPA are different numbers, and the marginal one is what matters for scaling decisions.

Average CPA = total spend ÷ total conversions. It's the headline number in your dashboard.

Marginal CPA = the cost of the next conversion if you increase spend. As you push for more volume, you bid on less-qualified auctions, so each additional conversion tends to cost more than the last. Your marginal CPA rises as you scale.

Why this matters: you can have a healthy average CPA of $30 while your marginal CPA — the cost of squeezing out the next few conversions — is already $80. If your product only supports a $50 CPA, those marginal conversions are losing you money even though the average looks fine.

This is the core reason Maximize Conversions can quietly become unprofitable at scale: it keeps buying conversions even as the marginal CPA climbs past your break-even point, because its only instruction is "spend the budget."

Which should you use?

The usual playbook

Most experienced advertisers run a sequence: start on Maximize Conversions to gather conversion data fast (the algorithm needs roughly 30 conversions in 30 days to bid well), then switch to Target CPA once you know your numbers, setting the target near your actual average CPA, then nudge the target down for efficiency or up for volume — slowly, so you don't shock the algorithm.

Work out your numbers first

Before you set any target, you need to know what CPA you can actually afford. That comes from your customer value: if a customer is worth $150 and you want a 3:1 return, your max CPA is $50. Run the numbers:

• CPA calculator — work out your current cost per acquisition and compare against customer lifetime value
• ROAS calculator — the return-on-ad-spend view of the same question, plus the breakeven and target ROAS you need
• LTV:CAC ratio calculator — whether your unit economics support scaling at all
• Conversion rate calculator — because a higher conversion rate lowers your effective CPA without touching bids

One trap to avoid

Don't set a Target CPA far below your current average and expect volume to hold. If your average CPA is $40 and you set a $20 target, Google will simply stop spending — it can't find conversions that cheap, so it pulls back rather than overspend. Lower the target in 10-15% steps, let it stabilize for a week, then step again.

Portfolio bidding — when neither strategy is set per-campaign

Both Target CPA and Maximize Conversions can be applied as portfolio bid strategies, spanning multiple campaigns rather than one at a time. This matters because Google's algorithm optimizes across the whole portfolio, potentially shifting spend between campaigns to hit an overall target rather than respecting a per-campaign ceiling — a campaign that's individually inefficient can still receive budget if it's pulling the portfolio average toward the target, while an individually efficient campaign can get starved if the portfolio doesn't need its volume. Portfolio bidding suits advertisers running several campaigns toward the same underlying business goal (e.g. multiple product-category campaigns all feeding the same margin target); it's the wrong choice when campaigns genuinely have different economics and need independent ceilings.

Seasonality and target adjustments

Google Ads' Seasonality Adjustments feature lets advertisers tell Smart Bidding in advance about a short-term conversion-rate change (a flash sale, a known traffic spike) that the algorithm's historical data wouldn't otherwise predict. Without this signal, both Target CPA and Maximize Conversions bid based on what they've learned from recent performance — a sudden real demand spike can initially look like noise to the algorithm and get under-bid during the exact window it matters most. For planned events with a known start and end date, setting a seasonality adjustment ahead of time avoids the multi-day lag it otherwise takes the algorithm to recognize and adapt to a genuine shift in conversion behavior.