Blog › Bidding Strategy
What Is Target CPA Bidding in Google Ads? A Practical Guide
By Jason Stern · · Founder, Enlightened Media (Google Premier Partner)
Target CPA (tCPA) bidding is a Google Ads Smart Bidding strategy that sets your bid for every auction automatically, trying to get as many conversions as possible at or below the cost per acquisition you choose. In current Google Ads it lives inside Maximize Conversions: you turn on the optional target CPA field and enter the average cost per lead you want.
How target CPA bidding actually works
With manual bidding you set a max cost per click and Google uses it everywhere. With target CPA, Google sets a different bid for each individual auction, at the moment someone searches. That is what Google calls auction-time bidding.
To set that bid, Smart Bidding predicts how likely this specific search is to turn into a conversion, using signals you could never weigh by hand at scale:
- The exact search query and the keyword it matched
- Device, operating system and browser
- Location and the searcher's distance from your service area
- Time of day and day of week
- Audience signals, like whether the person has visited your site before
- Ad and landing page combination
If a search looks likely to convert, Google bids higher. If it looks weak, it bids lower or barely competes. Your target is an average, not a cap. Some leads will cost twice your target and some will cost half, and the system aims for the blend to land near your number over time.
One setup note: if you are reading older guides that list "Target CPA" as its own strategy, that is the same thing. Google folded it into Maximize Conversions with an optional target, and Target ROAS into Maximize Conversion Value with an optional target ROAS.
What tCPA needs before you turn it on
Smart Bidding is only as good as the conversion data you feed it. In the accounts I manage, these three things decide whether tCPA works or burns money.
1. Accurate conversion tracking
Every form fill and phone call you care about has to be tracked, and nothing else should be counted as a lead. If tracking is broken or double counting, tCPA will optimize toward the wrong thing with total confidence. If you are not sure yours is clean, start with a conversion tracking audit before touching bidding.
2. Enough conversion volume
The algorithm learns from conversions, so a campaign getting three leads a month gives it very little to work with. Google does not publish a hard minimum anymore. In my experience, campaigns with roughly 30 or more conversions in the last 30 days behave predictably on tCPA, and campaigns well under that tend to swing around. Low volume accounts often do better on Maximize Conversions without a target, or Manual CPC, until volume builds.
3. Only primary conversions that matter
Google Ads lets you mark each conversion action as primary (used for bidding) or secondary (reported only). Your primary actions should be real leads: qualified calls, form submissions, booked appointments. Page views, scroll depth, button clicks and chat opens belong in secondary, if you track them at all.
How to choose a starting target CPA
Start from what you are actually paying now, not what you wish you were paying. Pull your actual cost per conversion for the last 30 to 60 days and use that, or something very close to it, as your first target.
Then sanity check it against what a lead is worth to you. A simple formula for the most you can afford per lead:
Max CPA = average job value x close rate x acceptable marketing margin
Example: a plumber with an average job of $800 who closes 40% of leads and is willing to spend 25% of revenue on acquisition can afford about $80 per lead ($800 x 0.40 x 0.25). If your current CPA is $110, do not set a target of $80 on day one. Start near $110 and work down.
If your current CPA is already well above what a lead is worth, the fix is usually not the bid strategy. It is usually keywords, landing pages or tracking. This breakdown of high cost per lead covers where to look.
The learning period and changing targets gradually
When you switch to tCPA or make a significant change, the campaign enters a learning period while the system recalibrates. Google shows this as a "Learning" status on the bid strategy. Performance is often uneven during this window, so judge results over a few weeks, not a few days.
A common practitioner guideline is to move your target in steps of roughly 15 to 20% at a time, then wait at least a week or two before the next change. That is the approach I use. A big jump, like cutting a $120 target to $60 overnight, usually tells Google to stop competing in most auctions, and lead volume drops hard before it recovers, if it does.
Common target CPA mistakes
- Counting junk conversions. If page views or "clicked phone number" events are primary, tCPA will happily find you cheap page views. Your reported CPA looks great and the phone stays quiet.
- Setting the target too low. An aggressive target throttles volume. Google bids so cautiously that you lose most auctions and spend drops below budget. If your campaign is underspending on tCPA, the target is often the reason.
- Changing it too often. Adjusting the target every few days keeps the campaign in a near constant state of relearning.
- Treating every conversion as equal. A 15 second call and a detailed quote request are not worth the same. If both count as one conversion, tCPA will chase whichever is cheaper to get. Set a meaningful minimum call duration, and if you assign conversion values, base them on real lead value rather than mixing arbitrary numbers for calls and forms.
- Expecting it to fix a weak campaign. Smart Bidding optimizes the traffic you give it. Bad keywords and a slow landing page stay bad.
tCPA vs Maximize Conversions vs Target ROAS vs Manual CPC
- Maximize Conversions (no target): spends your full daily budget trying to get the most conversions possible, with no cost control per lead. Useful for gathering data in a new campaign. Risky if budget is high relative to demand.
- Maximize Conversions with target CPA: same goal, but with a cost per lead guardrail. Best fit for lead gen businesses with steady conversion volume and leads of similar value.
- Target ROAS: optimizes for conversion value instead of count. Right for ecommerce, or lead gen when you import real revenue per lead. Wrong if every conversion carries the same placeholder value.
- Manual CPC: you set bids yourself. Still useful for very low volume campaigns or tight testing, but it ignores the auction-time signals Smart Bidding uses.
For most local service businesses I work with, the progression is Maximize Conversions to build data, then add a target CPA once volume is consistent.
Use offline conversion imports so tCPA finds real customers
The biggest upgrade most lead gen accounts can make is telling Google which leads actually turned into customers. By default, tCPA only knows that someone filled out a form. It has no idea whether that person was a $9,000 job or a spam submission.
Offline conversion imports close that gap. Your CRM records the Google click ID (GCLID), or you use enhanced conversions for leads, and you send back events like "qualified lead" or "closed sale" to Google Ads. Once those events have enough volume, you can make them primary and let tCPA bid toward people who look like your real buyers instead of anyone who submits a form.
This takes some setup between your CRM and Google Ads, but in the accounts I manage it is one of the most reliable ways to improve lead quality without raising spend.
Frequently Asked Questions
Is target CPA the same as Maximize Conversions?
Almost. In current Google Ads, target CPA is an optional setting inside the Maximize Conversions strategy. Without a target, Google tries to get the most conversions for your budget. With a target CPA set, it tries to get the most conversions while keeping the average cost per conversion at or below the number you enter.
How many conversions do you need for target CPA bidding?
Google no longer publishes a strict minimum, and tCPA can technically run with little data. In practice it works best when a campaign has steady conversion volume. In the accounts I manage, around 30 conversions in the last 30 days is a comfortable starting point. Below that, results tend to be more erratic.
Why did my conversions drop after setting a target CPA?
The most common cause is a target set well below your actual historical cost per conversion. Google bids conservatively to hit it, wins fewer auctions, and spends less. Other causes include the learning period after a change and recent tracking changes. Try raising the target closer to your real CPA and adjusting in gradual steps.
How often should I change my target CPA?
Not often. Each meaningful change can restart learning, so frequent edits keep performance unstable. A common practitioner approach is adjusting in steps of about 15 to 20% and waiting at least one to two weeks between changes, judging results over several weeks rather than day to day.
The Bottom Line
Target CPA bidding is a strong tool when the tracking is clean, the volume is there and the target is grounded in real numbers. When any of those is off, it quietly works against you. I personally manage every account at Enlightened Media, including bid strategy and conversion setup. Want a second set of eyes on yours? Request a free audit and I will tell you what I see.
Ready to Talk to a Google Ads Specialist?
Jason reviews every submission personally. Free audit, no sales pitch, responds within 1 business hour.
Get My Free Audit →Or call directly: 504-229-0603