Most Google Ads accounts treat all products the same. One campaign, one budget, one strategy — and then they wonder why performance is all over the place.
The problem is simple: not all products are equal. Some drive profit. Some drive volume. Some just consume budget and return nothing. If you don’t know which is which, you’re optimizing blind.
That’s why the first thing I do in any new account is label the product catalogue.
What labeling actually means
Product labels in Google Ads are custom attributes you attach to your Shopping feed. They don’t affect what Google shows — they affect how you structure campaigns and bidding.
A basic label system looks like this:
hero— top 20% of products by revenue, always onlong-tail— consistent sellers, moderate bidsnew— recently added, needs data before judgingdrain— spend without return, either fix or exclude
Once you have these labels, you can put each group in its own campaign with its own budget and bid strategy. Heroes get aggressive bids. Drains get excluded or capped. Long-tail runs on a moderate target ROAS.
Why this matters more than bid strategy
Everyone debates Smart Bidding vs. manual, Target ROAS vs. Target CPA. But bid strategy is downstream of campaign structure. If your structure is broken — if your hero products and your drain products are in the same campaign — no bid strategy will save you.
Google’s algorithm will allocate budget based on what it thinks will convert, but it doesn’t know your margins. It doesn’t know that a product with a high conversion rate is actually your worst margin product. You have to tell it.
Labels are how you tell it.
The automation angle
Doing this manually doesn’t scale. If you have 500 products, you’re not relabeling weekly by hand.
That’s what I built the Performance Labelizer for. It pulls your product performance data, compares revenue, ROAS, and click share against thresholds you define, and writes the labels back to the feed automatically. It runs on a schedule. You set the rules once.
The result is a feed that self-organizes around performance. Products that start draining get caught early. New products get a fair evaluation window before being judged.
Where most people get this wrong
The most common mistake is setting thresholds too aggressively. If you label anything under a 400% ROAS as a drain and exclude it immediately, you’ll gut your long-tail. Long-tail products often have lower individual ROAS but contribute meaningfully to total revenue at scale.
The second mistake is labeling once and forgetting. Seasonality matters. A product that drains in January might be a hero in November. The system needs to update.
Build the labels. Automate the refresh. Let the structure do the work.