Ecommerce SEO
Product feed optimisation
The job here is specific: fix the feed first, because in Shopping it determines more than the website does.
What does this work cover?
Fix the feed first, because in Shopping it determines more than the website does. In practice that means diagnosing disapprovals and the attribute errors behind them, and then the harder half: title and description structure that matches how people actually search.
Why this work exists
This is where the gap between what a business believes about itself and what is on record turns into lost revenue.
It belongs to Ecommerce SEO and is described here because it is the piece clients recognise and ask for.
The nearest piece of work to it is testing whether the product filter is usable with one thumb. Where that is already handled, this moves faster than the estimate; where it is not, the two get done together or neither of them holds.
What good looks like
It is healthy when the people who disagreed with it still use it.
The gap between that and where most accounts start is why this is scoped as product feed optimisation rather than as a line inside something larger. Deciding what happens to a URL when a product is renamed is the usual second step once it is in place.
When this is the right work
Three observations mean this work is due:
- A significant share of the feed is disapproved and nobody is watching
- Titles are internal stock codes
- Availability and price in the feed disagree with the website
If none of those is true, something else is probably the constraint, and the Ecommerce SEO overview maps the alternatives. The usual neighbours are finding the bestsellers that no page is built to capture demand for and making the stock feed and the site agree before the customer finds out.
What the work covers
What is in scope:
- Diagnosing disapprovals and the attribute errors behind them, usually the first thing to show a result, so it goes early.
- Title and description structure that matches how people actually search, which changes often enough that a single pass has a shelf life.
- Identifier, brand, category and variant attributes completed properly. Usually scoped at half the size it turns out to be.
- Supplemental feeds for margin labels and promotional data. This is where the theory meets the account and usually loses.
How it runs
Three stages. Most of the risk sits in the first one.
Bringing in whoever owns the system
Developers, agency partners or whoever actually controls the platform get involved before recommendations are written, not after they are rejected.
Proving it in one place first
Changes get made to a small, representative slice and measured there before being applied everywhere. It is slower for three weeks and faster for the rest of the year.
Clearing the contradictions
Where two systems disagree about the same fact, that gets settled before anything is built on top of either of them.
What each part runs into
In practice each of those four runs into something specific:
- Diagnosing disapprovals and the attribute errors behind them, harder in Birmingham because of a buying pattern that resolves faster than the reporting cycle
- Title and description structure that matches how people actually search – in Houston, complicated by the fact that a sector mix that pulls the brief in several directions
- Identifier, brand, category and variant attributes completed properly, which in Sydney runs into a seasonal peak that arrives before the content is indexed
- Supplemental feeds for margin labels and promotional data: the Kota version of this has to deal with a reporting line that averages away the market in question
The mistake most often made here
Managing bids on a feed that is a third disapproved. No bid strategy compensates for products that cannot show.
The reason it matters is that the failure is invisible from the report: the activity happened, the number did not move, and the conclusion drawn is usually the wrong one.
How it is measured
Feed approval rate, impression share in Shopping, and return on ad spend against margin. Reported monthly against a baseline taken before anything changed.
Two people should be able to calculate it independently and get the same answer. If they cannot, it is the wrong measure.
What you end up with
The thing you are left holding is a clean, fully-attributed feed with disapprovals resolved and a monitoring routine so they do not return.
Comes with a date and a next review, so it ages visibly instead of quietly.
How it differs by sector
By sector, the same scope produces different priorities:
- B2B services: A buying committee, a long cycle, and attribution that has to survive both. so this work has to account for that before anything else.
- Ecommerce and retail, where margin per order decides everything, and the product feed is usually the weakest asset. which moves this work earlier in the plan.
- In Home and trade services the binding constraint is that emergency intent, service-area geography, and a phone that has to be answered.
The industry pages carry the rest of that detail. B2B services and Ecommerce and retail are the two where this piece of work most often decides the outcome.
Common questions
What makes a good product title in a feed?
The words a buyer would use, in the order they would use them: brand, product type, the distinguishing attribute, then size or colour. Internal stock codes and marketing names match nothing. The title is the strongest matching signal in the feed, which is why it repays rewriting.
Why do products get disapproved?
Most commonly a missing or invalid product identifier, a price or availability that disagrees with the landing page, a restricted-category classification, or an image that breaks policy. The disapproval reason is reported per product, so this is diagnosable rather than mysterious; the problem is usually that nobody is looking.
How long does this take?
The diagnostic part is usually two to three weeks. What follows depends on what the diagnosis finds, and we scope the second phase after the first rather than quoting both upfront.
How much of our time does this take?
A few hours in the first fortnight, mostly answering questions about history that cannot be reconstructed from the data. After that, a monthly review. If the brief is still open, Log file analysis and Long task are the two that change it most often.
What do we actually get at the end?
A clean, fully-attributed feed with disapprovals resolved and a monitoring routine so they do not return. Documented, so it survives whoever was in the room, and specific enough to act on without a follow-up conversation. Teams that get this right have usually read AI Platform Optimization for Ecommerce and retail, and argued about Analytics and Measurement for Ecommerce and retail.
Ask about product feed optimisation
Give us the site and one question. If the honest answer is that you do not need this, we will say so.
Let us look at where you actually stand
We will show you what we would measure and why, so you can judge the approach before committing to it.