Should we feed profit or margin into Google instead of revenue?
Our answer
Yes, if margins differ across products. Send each order's profit as a secondary conversion for 30 days, then switch one campaign's bidding to it, starting at the profit return it already gets. Keep revenue if margins are flat.
- Confidence
- Medium
- Strongest proof
- 10 Google rules
- 4 practitioners, 9 opinions or claims
- Sources
- 12 sources
- 23 statements
- G
YD
New here? Take the 1-minute tour
Why this is the answer
3 reasons and our judgmentThe basket, not the product clicked, decides order profit. Savvy Revenue found 55% of 91 buyers who clicked one product bought something else. Young Digital Marketing argues shipping, quantity, returns and basket mix change order profit, so both send each order's actual profit.
Google's own value tools don't follow order profit. Its conversion value rules adjust value only by audience, location and device. Its product value adjustments, a beta announced in May 2026 for Performance Max and Shopping, weight products by brand or category, with no published results.
Profit bidding works in about 80% of Savvy Revenue's accounts, but not under heavy promotions or flat margins. One of its clients raised its target from 1.5 times to 2 times and saw revenue growth fall from 30% to 11% a year.
Our judgment
- Send each order's actual profit rather than a margin per product where you can. Savvy Revenue's check of one product and Young Digital Marketing's list of costs both point to the basket, and Google's cart data reports show whether buyers bought the advertised product.
- Switch only where margins differ and promotions are not heavy, and switch one campaign or portfolio first. Keep revenue as the secondary action so you can switch back. Savvy Revenue sets targets per campaign or portfolio sharing a bid strategy, JudeLuxe says ROAS may still serve with consistent margins and Google says a goal change shows Learning for up to 7 days. The same holds at every monthly spend, since the sources set conditions by margins and promotions. The only spend level named is JudeLuxe's view that ROAS increasingly fails to track what most brands care about above £10K a month.
- Set the profit target above 100% by the contribution you need. With profit before ad cost as the conversion value, a Target ROAS of 100% means the ads use up all of the order's profit. This applies the break-even logic of Setting a target ROAS (Relevant Audience's 1 divided by gross margin after variable costs, with Savvy Revenue's definition of order profit).
- Judge the switch on profit dollars rather than the profit ratio, and expect slower revenue growth. smec argues a high ratio target can cap absolute profit, and Google's 15% uplift claim measures campaign profit rather than a ratio.
- Use product value adjustments only when you cannot send order profit. Keep conversion value rules for audience, location or device differences rather than product margin. Louder says adjustments need clean margin data, and Young Digital Marketing calls value rules the lowest value of its three ways to feed profit.
Our judgment, built from Savvy Revenue, Young Digital Marketing, Google, JudeLuxe, Relevant Audience, smec (Smarter Ecommerce) and Louder: our reading of these sources, not a finding from any one of them.
What the sources say
23 statements from 12 sourcesEchelonn, a Google Ads agency, says 50 to 75% of converting queries in many 'prospecting' campaigns it reviews are the brand's own name, at $6+ a click versus 30 to 40 cents in a brand campaign.
“A new customer isn't always an incremental one. Plenty of them were warmed up by Meta first, then searched your name and bought.”
Menachem Ani, an agency owner, says Performance Max sees revenue, not margin: add cost of goods to the feed and a margin label so you can split the catalog by gross profit before Black Friday.
“PMAX sees revenue. It doesn't see what you keep.”
Google says passing cart data with conversions unlocks reports such as number of orders, average cart size and revenue, and with cost of goods from Merchant Center, gross profit metrics.
“By combining this data with the cost of goods sold (COGS) from your Merchant Center feed, Google Ads can generate additional metrics based on gross profit.”
Google says gross profit reporting needs cart data on purchase conversions plus cost of goods from the product feed. Without both, the reported gross profit will be inaccurate.
“Gross profit calculations require cart data from your purchase conversions and COGS data from your product feed.”
Google says the cost of goods attribute adds gross profit reporting for conversions, and a rough estimate or average is enough. It requires item-level information for each conversion.
“provide a rough estimate of your actual COGS”
Google says conversion value rules adjust value by location, device or audience at auction time for Smart Bidding, but Smart Bidding already uses those signals when the difference shows in reporting.
“If a user in one of those segments converts better, and this is reflected in your reporting, Smart Bidding already accounts for that.”
Google says Smart Bidding uses active conversion value rules at auction time, and campaign-level rules do not affect other campaigns except within portfolio bid strategies.
“Conversion value rules created at the campaign level will not affect bidding performance or reported conversions for other campaigns”
Google says Smart Bidding's Learning status lasts up to 7 days while ads keep serving, budget or target changes never trigger it, and results should be judged over 2 conversion cycles or 2 weeks.
“If you have past conversion data in your account, Smart Bidding never starts from scratch.”
Google says to use Maximize conversion value for the most value within budget and Target ROAS for a set efficiency, which needs at least 15 conversions in 30 days. Report two or more distinct values.
“Target ROAS should have at least 15 conversions in the last 30 days at the conversion tracking level.”
Relevant Audience, an agency selling Google Ads management, says break-even ROAS is 1 divided by margin (a 40% margin breaks even at 2.5) and that break-even is a floor, not a goal.
“a single account-wide ROAS target tends to underfund high-margin items and overfund low-margin ones. Splitting campaigns or product groups by margin band is the usual fix.”
Savvy Revenue's Andrew Lolk says profit bidding works in about 80% of the accounts where the agency runs it. One client moved to a 2x POAS target and revenue growth fell from 30% to 11%. No sample size.
“Revenue growth dropped from 30% YoY to 11%”
Savvy Revenue, a Google Ads agency, says to import profit as a secondary conversion for 30 days, set the first target at the profit return you already get, then switch in weekly steps. Skip it under heavy promotions.
“you cannot simply set your target to the new POAS number on day one”
A trade press piece describes a Google beta that lets advertisers multiply conversion values for specific products or attributes at campaign level, such as the highest-margin products. No Google or advertiser data.
Search Engine Roundtable reports a Google Ads beta, Product Value Optimization, that lets advertisers adjust values toward profit, sell-through or best-sellers. Google gives no performance data.
“Create adjustments to optimize towards what matters, like profit, seasonal sell-through, or best-sellers”
Jeremy Young says grouping campaigns by product margin is not the same as bidding for profit, since shipping, returns and basket mix change order profit. He recommends feeding actual profit as the conversion value.
“not the same as actually bidding for profit”
Louder says clean margin data in the product feed is essential, and that Shopping or Performance Max campaigns that exist only to split products by margin may now be consolidated.
“Clean product margin or lifetime value data is essential”
Jeremy Young, whose agency sells profit tracking, says sending profit per order to Google Ads as conversion value beats value rules. One unnamed client went from break-even to its best week, no figures given.
“the first one is sort of the lowest value one”
Google lets Performance Max and Shopping advertisers raise conversion values for chosen products, by brand or category, to favor profit or best-sellers without restructuring campaigns. No performance figures.
“without altering campaign structure”
A Google Ads agency case study of one home and garden retailer: low-margin products took 38% of spend for 12% of profit. After margin-tier campaigns and a feed clean-up, ROAS rose from 4x to 12x in 8 months, no control.
“Low-margin products consumed 38% of ad spend for just 12% of profit”
smec, which sells Google Shopping software, argues a strict Target POAS ceiling in Smart Bidding favors products with strong history and costs volume. Its example is an illustration with no data.
“strictly enforcing a high Target POAS”
JudeLuxe, a UK agency selling profit-focused Google Ads management, says POAS matters most when margins vary across products, and ROAS may still serve consistent margins. POAS needs SKU-level margin data.
“If your margins are consistent across products and your costs are stable, ROAS may still serve you well.”
Google describes a beta profit goal for Performance Max and Standard Shopping that uses cart-level conversions and cost of goods sold from Merchant Center. It gives no steps for getting access.
“Optimize for profit with your Performance Max and Standard Shopping campaigns using data from your cart-level conversions and cost of goods sold in Merchant Center.”
Google says gross profit optimization in Smart Bidding gives on average a 15% uplift in campaign profit over revenue-only bidding. Footnote: Shopping campaigns, Sept to Oct 2023, profit is revenue minus COGS.
“Customers who use gross profit optimization in Smart Bidding see on average a 15% uplift in campaign profit, compared to revenue-only bidding optimizations.”
In the full answer
For readers who sign in- What to do this week3 steps
- The numbers behind it3 figures from cited sources
- Where each source standsA chart placing each source on the question
- The strongest disagreementThe best case against the answer, from a cited source
- How sure we areWhat its confidence rests on, and what would change the answer