Not every product deserves the same ad budget. Smart bidding reads per-SKU ROAS, margin, stock cover and return rate, then sorts products into actionable groups and derives a target return for each from that group's own measured performance. From there you can act three ways: tag products with custom labels so your existing campaign rules pick them up, hold the money-losers out of your feeds, or set target ROAS and daily budgets directly on your live Google Ads campaigns. The last one shows you the current value against the new one, limits any single change to 50%, needs your approval, and records the previous value so it can be rolled back.
Performance Max gives you one target for a whole campaign, but a campaign is not one product. Inside it sit high-margin items that could take far more spend, thin-margin items being bid up on volume, and products that are almost out of stock and should not be bid on at all.
Most teams know this and cannot act on it, because the per-SKU data lives in one system and the bidding controls live in another.
How it works
Four signals per product, not just ROAS
Scoring combines return on ad spend, profit margin, stock cover in days, and return rate. ROAS alone rewards a product that sells well and comes straight back, or one about to go out of stock.
Products sort into groups
Every product lands in one of five groups — scale hard, scale with control, bid conservatively, test small, or stop. The group is a decision, not a score you have to interpret.
Each group gets a target from its own performance
The suggested target return is derived from what that group actually achieves, bounded by a floor and ceiling appropriate to its risk. When a group has too little measured data, FeedGraph says so and falls back to a policy default rather than presenting a guess as a measurement.
Three ways to act on it
Write the groups into your catalogue as custom labels so your existing campaign rules pick them up. Hold the loss-makers out of your feeds. Or set target ROAS and daily budgets directly on your live Google Ads campaigns.
Direct changes are shown, capped and reversible
Before anything is sent, FeedGraph reads the campaign’s current value and shows you current against proposed. Any single change is capped at 50% of the current value, shared budgets and portfolio strategies are called out, and the previous value is recorded so the change can be rolled back.
What you get
- ✓Per-SKU scoring from ROAS, margin, stock cover and return rate
- ✓Five actionable groups rather than a leaderboard
- ✓A target return per group, derived from that group’s measured performance and labelled when it is not
- ✓Custom labels 0–4 written to your catalogue for use in your own campaign rules
- ✓Target ROAS and daily budget writes to live Google Ads campaigns, each behind an approval and each reversible
- ✓A supplemental feed you can hand to Merchant Center
- —It does not write bids to Meta. Meta’s catalogue is managed by FeedGraph, but bid and budget changes there are not supported — Google Ads only.
- —It cannot act at all until you turn Write Actions on, and even then every individual change waits for a person to approve it. There is no autopilot.
- —It needs data to be useful. Margin and return-rate figures have to be supplied; without them, scoring falls back on fewer signals and says so.
- —One change can move a value by at most 50%. Reaching a much higher target takes several approved steps, deliberately.
Questions
Does it actually change my Google Ads account?
Yes, if you let it. It sets target ROAS and daily budgets on campaigns you select, through the Google Ads API. It cannot do anything until Write Actions is switched on for the workspace, and every change is staged with a before-and-after and waits for your approval. The previous value is recorded so any change can be rolled back.
What stops it spending far more than I meant?
Three things. A single change can move a value by no more than 50% of what it is now. Budgets shared between campaigns are refused unless you confirm that explicitly, because otherwise the change would reach campaigns you did not name. And nothing goes out at all without a person approving that specific change.
What if I do not want it touching my account?
Leave Write Actions off — that is the default. You still get the scoring, the groups and the recommended targets, and you can act on them by writing custom labels to your catalogue or generating a supplemental feed, both of which stay entirely inside FeedGraph.
See Smart bidding in use
More in Performance intelligence
All features →Advertising overview
One screen showing which products make money and which waste budget across Google and Meta — with what to do about each.
View feature →Ad performance
Product-level spend, ROAS and waste across Google and Meta, day by day — what to scale, fix or pause.
View feature →Google Ads Studio
Scores your Performance Max readiness across ten areas, lists what is holding you back, and executes campaign changes.
View feature →Ready to try it?
FeedGraph is in early access. Apply to join and we'll onboard you personally — no credit card required.
Join early access