Profit-Based Bidding in Google Ads: Switching From ROAS to POAS Without Losing Volume

A Target ROAS campaign hits its 5x goal and everyone relaxes. The campaign is "profitable," the dashboard is green, and budget gets pushed in. Then the quarterly P&L lands and the contribution margin is thinner than the dashboard implied. Nothing is broken. The campaign did exactly what you asked. The problem is what you asked: you told Google to maximize revenue per dollar, and it did, including by leaning into the products that produce revenue but very little profit. Target ROAS is doing its job perfectly. Its job is just not the one you actually care about.

The shift from ROAS to POAS, profit on ad spend, closes that gap. It is not a new bid strategy and not a new product to buy. It is the same value-based Smart Bidding you already run, fed a more honest number. Google optimizes toward whatever conversion value you send it, and the entire move comes down to changing that value from revenue to profit. The mechanics are straightforward; the discipline is in migrating without throwing away the learning your current campaigns have accumulated.

The flaw hiding inside Target ROAS

Smart Bidding strategies use Google's AI to optimize for conversions or conversion value in every auction. Target ROAS is the value-based one: you state the average conversion value you want per dollar spent, and Google predicts each potential conversion's value and bids to hit that target. The catch is entirely in the word "value." For most accounts, the conversion value being sent to Google is revenue, the order total. So Target ROAS is, precisely, Target Revenue On Ad Spend.

Consider two products at the same 100 dollar price. One is a high-margin item that nets you 60 dollars of profit; the other is a near-loss-leader that nets you 5 dollars. To a revenue-fed tROAS campaign, these two sales are identical: both are 100 dollars of conversion value. Google has no reason to prefer the profitable one, because you never told it which one is profitable. Worse, the low-margin product often converts more easily (it is cheaper, more discounted, more competitive), so the revenue-optimizing system will frequently lean toward it, scaling exactly the sales that build revenue and erode profit. The green 5x dashboard can sit on top of a product mix that is quietly drifting toward your worst margins. The metric is not wrong; it is answering a question about revenue while you are asking a question about profit.

What POAS actually is

POAS reframes the target around profit. Instead of "how much revenue did I get per ad dollar," it asks "how much gross profit did I get per ad dollar." A 5x ROAS on a 20%-margin order is a 1x POAS, which is break-even before overhead. A 3x ROAS on a 70%-margin order is a far healthier 2.1x POAS. POAS exposes the truth that revenue hides: two campaigns with identical ROAS can have wildly different profitability, and the one with the lower ROAS is sometimes the better business.

Crucially, POAS is not a separate Google product. Google's own documentation on value-based bidding states plainly that the value you optimize toward can be revenue, profit margins, or customer lifetime value, depending on your strategic focus. That single sentence is the permission slip. Optimizing on profit is not a workaround or an unsupported trick; it is one of the intended uses of the value-based bidding you are already running. You are not switching strategies. You are changing the definition of value from revenue to profit, and letting the same Smart Bidding machinery optimize toward the better target.

How to feed profit to Smart Bidding

You change what Google optimizes for by changing the conversion value you send it. There are two practical ways to do that, and most mature accounts end up using a combination.

• Margin-adjusted conversion values. Instead of sending the order total as the conversion value, send the order's gross profit. This requires knowing your margin at the product level and computing the profit value at the moment of conversion, typically in the data layer or in the conversion tracking setup, so the value passed to Google is profit rather than revenue. This is the most precise method, because the value is true per-order profit, but it depends on having clean, current margin data, which is why a solid product-level data foundation matters as much here as anywhere. The signal-quality groundwork in first-party data in PPC is the prerequisite for trustworthy profit values.
• Conversion value rules. Conversion value rules let you adjust reported conversion values, and Google uses those rules in real time within Smart Bidding to optimize Target ROAS and Maximize conversion value. They are coarser than true per-order margins, because they adjust by conditions such as location, device, or audience rather than by exact product margin, but they are simpler to deploy and useful for expressing differences like "new customers are worth more than returning ones" on top of a profit base.

For Shopping specifically, the cleanest path is margin data flowing through the product feed so each product carries a profit-adjusted value, which is why feed quality and structure become a profitability lever, not just a discovery one. We cover the feed and account plumbing this rides on in the Google Ads Merchant Center setup guide. The point common to both methods is the same: Smart Bidding will optimize toward whatever number you define as value, so the entire job of moving to POAS is getting profit, rather than revenue, into that field.

Migrating without losing volume

This is where most POAS attempts go wrong. An advertiser flips every conversion value from revenue to profit on a Monday, the absolute value of every conversion drops sharply (profit is a fraction of revenue), and the campaign's tROAS target, still set for revenue-scale numbers, is suddenly wildly out of calibration. Smart Bidding sees the value signal collapse, delivery becomes erratic, and volume craters. The strategy was sound; the transition was a cliff.

Migrate as a ramp, not a switch.

1. Measure your true product-level margins first. POAS is only as good as the profit data behind it. If you do not know margin by product, that is the work to do before touching bidding, not after.
2. Re-baseline the target before you change the values. A profit-based conversion value is numerically smaller than a revenue-based one, so a 5x revenue tROAS is not a 5x profit tROAS. Calculate the equivalent profit target so you do not accidentally tighten or loosen the campaign when you switch the values.
3. Phase the values in. Rather than swapping 100% of conversion value from revenue to profit overnight, transition deliberately, and where possible on a subset of campaigns first, so you can compare profit-fed campaigns against revenue-fed controls before committing the whole account.
4. Give Smart Bidding room to relearn. Changing the conversion-value signal is a meaningful change, so expect a recalibration window and avoid stacking other big edits on top of it. The flexible-target lever is useful here for managing the transition without choking volume, which we walk through in flexible ROAS targets.
5. Judge the result on profit and on volume together. The win condition is not just higher POAS; it is higher total profit. A campaign that improves POAS while collapsing volume may make less money overall, so watch absolute profit, not only the ratio.

Done this way, the migration keeps the learning your campaigns already built and shifts the optimization target underneath it, rather than resetting everything and hoping volume returns.

When plain ROAS is still the right call

POAS is not universally superior, and pretending it is leads to over-engineering accounts that do not need it.

• Uniform margins. If your catalog is roughly uniform in margin, optimizing on revenue and optimizing on profit produce nearly the same bidding behavior, and the added complexity of profit values buys you little.
• Unreliable margin data. POAS fed bad margins is worse than honest ROAS, because you will confidently optimize toward the wrong number. If you cannot trust your product-level cost data, fix that before switching, and run revenue ROAS in the meantime.
• Lead generation without clear per-conversion economics. When a conversion is a lead whose value is uncertain until later, profit per conversion may be too noisy to feed directly, and a value-rule or lifetime-value approach is a better intermediate step than raw POAS.

The honest framing is that POAS is the right default for catalogs with meaningful margin variance and trustworthy cost data, and an unnecessary complication for everyone else. The deeper question of how profit thinking should reshape the rest of your paid program, including budget allocation, is its own subject, and it connects directly to the cross-channel measurement discipline in running Meta and Google Ads together.

Frequently asked questions

What is the difference between ROAS and POAS?

ROAS, return on ad spend, measures revenue per ad dollar; POAS, profit on ad spend, measures gross profit per ad dollar. Because revenue ignores margin, two campaigns with identical ROAS can have very different profitability. POAS corrects for that by optimizing toward profit, so it reflects the number that actually determines whether the advertising makes money.

Is POAS a separate Google Ads bid strategy?

No. POAS uses the same value-based Smart Bidding, such as Target ROAS or Maximize conversion value, that you already run. Google's documentation states the conversion value can be revenue, profit margin, or lifetime value, so you achieve POAS by feeding profit as the conversion value rather than revenue. The bid strategy does not change; the definition of value does.

How do I send profit instead of revenue to Google Ads?

Two ways, often combined. Send margin-adjusted conversion values, computing each order's gross profit at conversion time so the value passed to Google is profit, which requires product-level margin data. Or use conversion value rules, which Google applies in real time within Smart Bidding to adjust reported values by conditions like location, device, or audience. Feed-level profit values are the most precise route for Shopping.

Will switching to POAS hurt my campaign volume?

It can if you switch abruptly. Profit values are numerically smaller than revenue values, so flipping everything at once leaves your existing tROAS target miscalibrated and can disrupt Smart Bidding. Migrate gradually, re-baseline the target to its profit equivalent, phase the values in, and allow a relearning window. Judge success on total profit and volume together, not on the POAS ratio alone.

When should I not use profit-based bidding?

When your catalog has roughly uniform margins (revenue and profit optimization behave almost identically), when your product-level cost data is unreliable (POAS fed bad margins optimizes toward the wrong number), or in lead generation where per-conversion profit is too uncertain to send directly. In those cases, well-run revenue ROAS, or a value-rule approach, is the better choice.

References

• About Smart Bidding, Google Ads Help
• About Target ROAS bidding, Google Ads Help
• About Smart Bidding using value-based bidding for Search and Shopping, Google Ads Help
• About conversion value rules, Google Ads Help
• Impact of conversion value rules on Smart Bidding and performance, Google Ads Help
• About Maximize conversion value bidding, Google Ads Help