YourProductFeedIsTheMostImportantDocumentInYourEcommerceBusiness

More important than your brand guidelines. More important than your trading report. More important than your agency contract. There is one document that determines whether Google shows your products or your competitor's - and most ecommerce founders have never read it properly. Most agencies have never fixed it.

Let me tell you about a document you've been ignoring

There is a file sitting behind your Google Shopping campaigns right now. It is called your product feed. It is the data file that tells Google what you sell - your product names, descriptions, categories, prices, availability, and dozens of other attributes that Google uses to decide when to show your products and to whom.

If this file is accurate, well-structured, and written in the language your customers use to search - Google can match your products to the right searches at the right moment. Your ads appear. Your customers find you. Your campaigns perform.

If this file is written in the language of your warehouse - internal SKU codes, manufacturer references, season codes, generic product names that mean nothing to a customer with intent - Google cannot make the match. Your ads do not appear. Your customers find your competitors. Your campaigns spend your budget and return nothing.

This is not a technical edge case. It is the most common and most expensive failure mode in Google Shopping. And it is almost entirely invisible in standard reporting, because impression share lost due to poor feed quality does not appear as a line item in your monthly report. It shows up as revenue you never made, customers you never reached, and a market position that quietly erodes while the ROAS holds steady on the traffic you do capture.

Why the feed matters more than the campaign

In Google Search campaigns, you write the keywords. You define which searches trigger your ads. The campaign structure is, to a significant degree, under your control.

Google Shopping does not work this way. In Shopping - and in Performance Max - Google reads your product feed and decides which searches are relevant to your products. You do not choose the keywords. Google chooses them, based on what your feed tells it about what you sell.

This means your product titles are your keywords. Your product descriptions are your targeting signals. Your categories, attributes, and custom labels are the data Google uses to decide where your budget goes and who sees your ads.

A poorly constructed feed is not just a technical problem. It is a targeting problem. It means Google is making decisions about where to spend your budget based on incomplete, inaccurate, or poorly written product information. It means your ads appear for the wrong searches - or do not appear at all.

No bid strategy compensates for this. No budget increase fixes it. No campaign restructure addresses it. Until the feed is right, everything else is optimising the wrong thing.

The warehouse language problem

The most prevalent form of feed failure in practice is what we call warehouse language: product titles written for internal stock management systems rather than for customer search intent.

Here is what it looks like. A fashion brand sells a navy midi wrap dress. In their stock system, this product is identified as "Occasion Dress - Style D218 - Midnight - Size 12." The product title in their feed reflects this internal naming convention.

Meanwhile, their customer is on Google. They type "navy midi wrap dress for wedding guest." Google reads the product title. It reads "Occasion Dress - Style D218 - Midnight." It cannot confidently match these to the search. It finds a competitor whose title reads "Navy Midi Wrap Dress Wedding Guest Occasion." It shows the competitor.

The brand's ad did not appear. Their budget was not spent on this search. But their competitor's was. And the customer bought from the competitor.

Multiply this across a catalogue of several hundred products. Multiply it across several thousand daily searches. Multiply it across 18 months of running Shopping campaigns with an unoptimised feed. The cumulative invisible revenue loss is substantial - and it does not appear anywhere in the reporting.

What Google actually reads

Understanding why feed quality matters requires understanding which feed attributes Google prioritises when matching products to searches.

Product title is the most important attribute in the feed. Google weights title content heavily when determining query relevance. The most impactful keywords for your product should appear early in the title - colour, material, gender, product type, size descriptor, and any occasion or use-case signals that are relevant to how your customers search.

Product description is secondary but valuable, particularly for longer-tail queries and for providing context that reinforces the title. A well-written description increases the probability that Google classifies your product correctly and serves it for relevant searches.

Google product category tells Google's taxonomy which category your product belongs to. Incorrect or overly broad categorisation means Google is uncertain about your product's context, which reduces its confidence in serving it for specific searches.

Custom labels are your opportunity to segment products by any attribute you choose - margin tier, season, performance level, new vs established. They do not affect Google's matching, but they determine how your campaigns are structured and how you control which products receive which level of bidding and budget attention.

GTIN and MPN - where these exist - allow Google to identify your product in its catalogue and serve it for product-specific searches. Missing GTINs on products where they exist reduces your eligibility for certain placements.

The title rewrite: what good looks like

Rewriting product titles for search intent is the single highest-leverage activity in Google Shopping management. It costs nothing except time, it requires no technical infrastructure, and its impact compounds over time as the feed accumulates conversion signal on high-intent queries.

The formula for a well-optimised product title varies by category, but the underlying principle is consistent: titles should contain the exact words and phrases your target customer uses when searching for a product like yours, in a sequence that front-loads the most distinctive and high-intent terms.

The principle is always the same: think about how your customer searches, not how your warehouse identifies the product. The stock code is irrelevant to Google. The customer's search language is everything.

Margin segmentation: the advanced lever

Once the feed is optimised for search intent, the next layer is segmentation by margin. This is where the feed becomes a tool for profitability management rather than simply a visibility tool.

Custom labels allow you to tag each product with a margin tier: high margin, mid margin, low margin, or excluded. These tags flow through to your campaign structure, where each tier can be managed with a different ROAS or POAS target that reflects its break-even threshold.

Without margin segmentation, Google's algorithm allocates budget based on conversion signal. Products that have converted frequently in the past receive more budget. Products with less history - often newer or more premium lines - receive less. This creates a self-reinforcing cycle where your highest-volume products dominate spend regardless of their margin contribution.

With margin segmentation, you can intervene in this allocation. High-margin products receive aggressive bids even if their historical volume is lower. Low-margin products receive conservative bids or hard budget limits even if their historical conversion rate is strong. The account optimises toward profit rather than toward volume.

This structural change - feed segmentation combined with tier-based campaign architecture - is what separates Google Shopping management from Google Shopping optimisation. Most accounts are managed. Few are optimised at this level.

What "feed management" actually means in most agencies

Feed management is a line item in most agency proposals. It sounds comprehensive. In practice, in the majority of agencies, it means one of two things: the feed has been submitted to Google Merchant Center and is technically active, or a feed management tool like DataFeedWatch or Channable has been connected to the account.

Neither of these is feed optimisation. A submitted feed that says "Boot - AW24 - SKU7741B" is a submitted, active, technically functional feed that is costing you impressions every single day. A feed tool that automates the submission of bad data automates the problem rather than solving it.

True feed optimisation requires human judgment applied to title language, category selection, and attribute completeness - informed by search term data from the account, customer search behaviour, and an understanding of how Google's matching works in practice. It is not a tool. It is a skill. And it is underrepresented in most agencies because it is time-consuming and the results, while significant, are not instantly visible in the metrics most agencies report on.

The compound effect of feed quality

The impact of a well-optimised feed compounds over time in a way that campaign optimisation alone does not. When your titles match high-intent searches accurately, Google serves your ads to the right audience. Those audiences convert at a higher rate. Higher conversion rates improve your Quality Score equivalent in Shopping, reducing your effective CPC over time. Lower CPCs improve your ROAS without increasing your budget. Better ROAS signals allow more aggressive bidding, increasing impression share on the searches that matter.

The virtuous cycle runs in the opposite direction with a poor feed. Broad, inaccurate matches generate low conversion rates. Low conversion rates increase effective CPCs. Higher CPCs compress ROAS. The account looks like it needs more budget, when what it actually needs is better data.

This is why the feed is the foundation of everything in Google Shopping. Not the bidding strategy. Not the campaign structure. Not the audience signals in Performance Max. The feed. Getting it right is the first job. Everything else is built on top of it.

Where to start

If you have never conducted a proper feed audit, the starting point is your top 20 products by spend. For each one, conduct the search test: open Google in incognito, search for the product using customer language, and observe whether your ad appears and whether your title reflects the search intent.

Then pull your search term report for Shopping campaigns and filter for your highest-spend queries. Look at the relationship between the search terms Google is matching and the product titles that triggered them. Where there are gaps - where generic, low-intent terms are generating spend on your highest-margin products, or where your best products are missing from high-intent searches - you have found your optimisation priorities.

Rewrite titles starting with your highest-spend, highest-margin products. Apply search intent language. Remove internal codes, season references, and manufacturer jargon. Front-load the attributes that your customers search for first.

Then measure the impact on impression share, click-through rate, and - most importantly - conversion rate by product. Feed quality improvements take two to four weeks to show in performance data as Google recrawls the feed and updates its matching. The results, when they come, are durable in a way that bid changes rarely are.

Your product feed is not a technical document. It is your presence on Google Shopping. Treat it accordingly.