Is Performance Max worth it for small ecommerce stores?

The campaign type is not size gated. What limits a small account is that automated bidding learns from conversions, and a thin flow of conversions produces slow, noisy learning. The campaign will still spend the budget while it works that out.

Three boundaries make the difference. Exclude brand queries so the campaign cannot meet its target by buying people who were already coming. Give it a product feed, because for a store the feed is the strongest input the campaign has. And point it at a single conversion action that means an order, with a value attached, rather than a mix of actions that includes newsletter signups and add to cart.

With those in place, a small store gets a campaign that behaves close to Shopping with extra placements attached. Without them, the most common outcome is a campaign that reports a good return built largely on demand the store already had.

Set the budget so that it can buy a meaningful number of clicks a day rather than a handful. A budget that only affords two or three clicks daily gives the system almost nothing to learn from, and the account spends months in the same state.

Putting the entire catalogue into one asset group is the first. A store with a few hundred products and mixed margins ends up with budget flowing to whichever products convert most easily, which is usually the cheapest ones, and the campaign optimises towards low value orders that clear the target on paper.

The second is leaving conversion tracking loose. If the campaign counts more than one action as a conversion, or counts a purchase twice because a theme tag and a channel tag both fire, the bidding is optimising towards a number that does not exist. Fix measurement before you judge the campaign, not after.

The third is judging it in the first fortnight. Every change to budget, target or assets restarts part of the learning, and a small account that adjusts weekly never gets a clean read. Set the campaign, leave it alone through a full purchase cycle, then change one thing.

The fourth is uploading no assets at all and then wondering why the campaign behaves like Shopping. That is a legitimate way to run it, and it is a choice worth making deliberately rather than by omission.

Choose Standard Shopping when you need to see and steer where the money goes. It exposes search terms and lets you add negatives directly, which matters most in the first months of an account when you are still learning which queries your products actually serve.

It also suits catalogues with a strong internal split: a few products carrying most of the margin, or a long tail that should be bid differently from the core range. Campaign priorities and product groups give you that split explicitly, and the reporting shows you the effect immediately.

And it suits a store that has not yet proved the unit economics. Running a manual or lightly automated Shopping campaign for a period gives you real cost per click and conversion rate data for your own products, which is the input you need to decide what any automated target should be.

The two are not exclusive. A common shape for a growing store is Standard Shopping for the core range where control matters, a brand search campaign held separately, and Performance Max carrying the wider catalogue once there are enough conversions to feed it.

Start with the feed. Titles that lead with the terms a shopper types, correct product types and categories, real identifiers, clean images and accurate availability do more for a small Performance Max campaign than any setting inside it.

Then build one campaign with one or two asset groups aligned to how the catalogue actually splits, not to how the website navigation splits. Attach the brand exclusion list and the account level negatives. Set a value based bidding target that reflects your margin, and give the campaign a budget it can actually spend at that target.

Supply real assets rather than leaving Google to assemble the campaign from the feed alone: several headlines, a description set, product photography and lifestyle imagery in the required shapes, and video if you have it. If you do not supply video, expect the campaign to use what it can generate.

Then decide the review date in advance. Look at search themes and the placement reporting for obvious mismatches, check that brand really is excluded by reading the search terms, and hold the structure steady until the review date arrives.