Custom labels: everyone
says use them - here's how
to actually do it
Using custom labels is the floor. Every PPC account running Shopping or PMax should already be there. What separates campaigns that scale from campaigns that plateau is how those labels are built, maintained and aligned to the business.
Here's how to set them up, the approaches worth considering and why the strategy needs to evolve with the business.
Custom labels are not the competitive edge. How you build and use them is.
Why custom labels matter
By default, Shopping and PMax campaigns are segmented by the attributes already baked into your feed - product ID, brand, product type, category. These are fine as a starting point, but they're Google's structure, not yours.
Custom labels hand that control back. They let you slice and segment your product catalogue in any way that makes sense for how your business actually operates - by margin, by stock age, by strategic priority, by performance. Any attribute Google doesn't natively surface, you can create.
The result: campaign structures that align to your business goals rather than platform defaults.
How to adjust custom label values
There are three main routes, each suited to different setups:
Edit directly in Merchant Center
Google Merchant Center lets you apply overrides to individual products manually. Straightforward for small catalogues or one-off changes, but it doesn't scale - you'd spend more time managing overrides than running campaigns.
Layer values via a secondary feed
A supplemental feed sits alongside your primary feed and writes custom label values against matching product IDs. Typically built in Google Sheets, it's updated manually or on a schedule. Good middle ground - no dev dependency, works for most accounts, and gives you decent control without a big technical lift.
A basic supplemental feed only needs two columns: ID and the custom label attribute. Match on product ID exactly as it appears in your primary feed and GMC will do the rest. Make sure your fetch schedule is set tight enough to catch stock and price changes before they matter.
Rule-based automation at scale
Tools like DataFeedWatch, Channable or Feedonomics let you write conditional rules that assign custom label values dynamically - no spreadsheet maintenance required. If price drops below a threshold, the label updates. If stock drops below a set quantity, it shifts segment. The logic runs automatically every time the feed refreshes.
For larger catalogues or accounts where label values need to change frequently, this is the right approach. The upfront setup takes more time, but the ongoing maintenance drops significantly.
Approaches worth considering
There is no single correct custom label strategy. Every account is different. Every business has different goals, different margin profiles and different pressure points. What works well for a high-AOV furniture brand looks nothing like what a fashion brand needs during a peak sale period.
That said, the following approaches come up repeatedly and are a solid starting point to adapt from:
Treat new products differently from the start
New arrivals often need a different push - especially if you're in a fashion or seasonal vertical where recency drives purchase intent. Labelling new stock separately lets you control budget allocation and bid strategy independently of your evergreen range, rather than letting them compete in the same bucket.
Segment by what the data tells you
Splitting products into tiers based on performance - typically conversion value (whether you use default transaction value or profit proxies) and conversion rate - lets you apply different bid strategies and budgets to each tier. Star performers get the most headroom. Low performers get tighter control or get cut.
Once you're across hundreds or thousands of products, a tool that auto-assigns based on live data becomes significantly more reliable than a quarterly spreadsheet review you keep putting off.
What job is this product actually doing?
Not every product in the catalogue is trying to do the same thing. Some SKUs are the high-margin drivers you want to scale. Some are cash-flow products that need to keep moving at volume. Some are new range launches you're investing in to build data. Labelling by SKU role lets you structure campaigns around these different objectives rather than treating all products as interchangeable.
Old inventory is a cost, not just a product
Stock sitting in a warehouse too long is tying up cash. Labelling by stock age - typically banded into ranges like 0–30 days, 31–90 days, 90+ days - lets you give older inventory a harder push when you need to, without manually hunting through the catalogue every time clearance is on the agenda.
This one tends to be overlooked until it hurts. Building the label early makes the clearance moment much easier to execute.
Bid based on what actually makes money
This is one of the more powerful approaches and, frustratingly, one of the most underused. Knowing which products generate the most contribution margin per sale changes how aggressively you should be bidding on them. A high ROAS SKU with a thin margin is not the same as a lower ROAS SKU with a strong margin - but without margin-tier labels, your campaign structure treats them identically.
Combining margin tier with price point adds another dimension - useful when you're trying to identify where you can push spend hardest or where you need a floor.
Combining label values
You have five custom label slots (0–4). Each approach above could live in its own slot - or you can combine values within a single label to create more specific segments when one attribute alone is not enough.
Examples of combinations that make sense in practice:
The value of combining is fewer labels to manage and more precise campaign targeting in one shot. The trade-off is that combined values can multiply quickly if you are not disciplined - keep it simple enough that someone else could maintain it.
Set and forget doesn't work
Custom labels are not a one-time setup. Business needs shift. Catalogue changes. Margin profiles move. What was relevant last quarter might be wrong today.
Overlooking age of stock means scrambling when clearance pressure hits and the campaign structure is not set up to respond quickly. Ignoring margin tiers means leaving contribution margin decisions to an algorithm that has no idea what anything actually costs you.
The accounts that get the most out of custom labels are the ones that treat them as a living part of their feed strategy, not a checkbox ticked at setup.
Free consultation
Not sure if your GMC is set up to grow?
If you have doubts about whether your Merchant Center and feed structure are actually supporting your business goals, book in a free consultation. No pitch - just a straight read on where things stand.
Book a free consultation →Further reading
The Google Shopping Feed Checklist
A step-by-step checklist covering the feed attributes that drive Shopping and PMax performance - before you touch a single bid.
The 5 Shopping feed mistakes I find in 9 out of 10 accounts
The most common feed errors holding back Shopping performance - and what to fix first.
Google Shopping Price Comparison
How competitive pricing affects your Shopping impressions and what to do about it without racing to the bottom.