What "Feed-First" Shopping Ads Management Actually Means

If you've been running Google Shopping campaigns for any meaningful stretch of time, you've probably noticed something: a lot of the manual controls we used to rely on have quietly disappeared, and the ones that haven't disappeared have been gradually made less meaningful by the way Performance Max operates.

The query sculpting techniques that practitioners (myself included) spent years getting good at have become harder to execute cleanly. Match types mean considerably less than they used to. And the net effect of all of this is that Google has been systematically reducing the levers we once pulled to influence Shopping performance, and replacing them with a single, somewhat uncomfortable question: how good is your data?

That's what "feed-first" shopping management means, at least in the way I think about it, and I think it's worth spending some time with because it represents a genuine philosophical shift in how this work gets done rather than just a tactical adjustment.

The Shift That Actually Happened

I remember when Shopping campaign architecture was a real source of competitive advantage. You could separate products by priority tier, sculpt query flow with carefully placed negative keywords, build out custom label hierarchies that gave you bidding control at a granular level, and if you understood those mechanics well you'd outperform the practitioners who didn't by a clear margin. It felt like craft, honestly, and I think a lot of people in this industry (again, myself included) developed a real attachment to that complexity because it rewarded expertise in a way that felt earned.

Performance Max changed that calculation substantially, and I think the industry is still processing what that actually means (and how to regain the same level of... immediate performance from changes). The reality is, when you're running a PMax campaign, you're handing Google a set of assets and signals and asking it to optimize across its entire inventory of placements, and the algorithm's primary input for Shopping-related placements is your feed. The campaign structure gives Google direction, but the feed gives Google the raw material it works with, and the quality of that raw material determines how well the algorithm can do its job. From what I've observed across the accounts we work with at ZATO, you can build a sophisticated campaign architecture on top of a mediocre feed and the results will be mediocre, but a relatively straightforward campaign structure built on top of an excellent feed is more likely to outperform more complex setups with worse data. The feed is the foundation, and foundations determine ceilings, even if they don't guarantee what gets built on top of them.

Put a different way, optimizing your feed isn't the ONLY thing you should be doing when managing Google Shopping Ads and Performance Max campaigns, but it is definitely the place to start.

What This Looks Like in Practice

So what does feed-first management actually look like when you're doing the work?

Title optimization is probably the most direct expression of feed-first thinking, and it's also the place where I see the clearest gap between what most brands are doing and what they could be doing. Your product titles are one of the primary signals Google uses to match your products to search queries, and the difference between a title that's written for your internal catalog system and a title that's written for the way your customers actually search can be significant, both in impression volume and in the relevance of the queries you're matching. Using search term data to inform how you write and structure titles is one of the more reliable ways to improve Shopping performance without touching a single campaign setting, which I find almost philosophically satisfying in the current environment. You can read more about how we approach this kind of Google Shopping optimization over on the blog.

Product type and Google product category are two attributes that get forgotten more consistently than almost anything else I see in a feed audit, and both of them give Google meaningful signals about where to surface your products and for what kinds of queries. Product type in particular is an attribute where I've seen consistent positive impact when it's optimized thoughtfully rather than just filled in quickly, and it's often one of the first attributed we look to optimize when an account is underperforming relative to what its catalog should be capable of generating.

Custom labels are the feed attribute that connects most directly to campaign strategy, and they become important in a PMax world where you have fewer campaign-level levers available to you. They allow you to segment products by a variety of different segments. That could be by margin, by promotional status, by performance tier, by whatever dimension is meaningful to the specific business, and then treat those segments differently in your campaign structure. I think of custom labels as one of the primary remaining ways to maintain intentionality about where your budget goes and which products get the most aggressive targeting, and the fact that they live in the feed rather than the campaign is itself a pretty good illustration of why feed-first thinking matters.

Supplemental feeds/data sources are a tool that helps make all of this executable at scale, because they allow you to make targeted changes to specific attributes without touching the primary data source every time something needs to change (though the most ideal way to manage will always be to edit the source directly if you can!). For the accounts we manage, supplemental feeds for custom labels and promotional data are often parts of the setup, and they give us the ability to respond quickly to competitive moments or client needs at scale (across multiple, random IDs) without the complexity of going back to the primary feed. We've written specifically about setting up supplemental data sources in Merchant Center Next if you want to go deeper on the mechanics of how this actually works.

The Part I Want to Be Careful About

I want to say something here that I think is easy to miss when people talk about feed-first management as if it's a cure for everything, because I've seen it become a kind of magical thinking in some conversations and I don't want to contribute to that.

The feed is the most important variable within Google's ecosystem. It's not the most important variable in whether a business succeeds with Shopping ads. Your offer, your pricing, your product-market fit, your conversion rate once someone arrives on your site... these things matter more than any feed optimization, and a beautifully structured feed sending high-quality traffic to a product page that doesn't convert is still a bad outcome. I think it's actually part of our job as practitioners to be clear about this with clients, because the expectation that feed work alone will produce transformational results is one that leads to difficult conversations later.

What feed-first management does, is give you the highest possible chance of being matched to the right queries, shown to the right people, and competing effectively in the auctions that matter to your business. What happens after the click is still on the business to get right, and I've found that the brands who understand that distinction tend to be better partners to work with over time, because they're thinking about the whole system rather than just the part we control.

For businesses that have the product, the pricing, and the on-site experience working in their favor, a feed-first approach to Google Shopping and PMax is where the most meaningful and durable performance gains tend to live. Not because it's glamorous work (it really isn't, and anyone who tells you otherwise has either never done a feed audit or has a much higher tolerance for filling in attributes than I do), but because it's where Google is actually making decisions, and working with that reality rather than against it is, it seems to me, the most honest and sustainable path forward in a landscape that keeps changing the rules on us.

‍