Before you sign a shopping feed contract, check these tools and providers

What to take away

Start with the requirement list, not the shortlist

The common mistake is to open three vendor websites, compare feature grids and pick the longest channel list. That tells you what each supplier wants to sell, not what your catalogue needs.

Write requirements before you speak to anyone. List current channels and the two you plan to add. Note product count, price change frequency and data ownership. Add marketplaces with high return rates, because feed accuracy matters most there.

Then rank each requirement as essential or desirable. A retailer with 400 SKUs and one warehouse has a different list from a marketplace seller with 40,000 SKUs across three fulfilment centres. The same software rarely suits both.

If you are still mapping options, the shopping feeds: England market guide for 2027 sets out how channel mix and feed requirements differ across the UK, which helps you decide what belongs on the essential list.

Before shortlisting, decide who signs off feed changes. If merchandising, marketing and finance all have a veto, put that in the requirements. Providers that need a ticket for every attribute change will slow you down.

Now add named tools to the longlist. DataFeedWatch suits lean teams wanting self-service feed management. Channable suits retailers running Google Shopping and marketplace feeds together. Feedonomics suits larger catalogues needing managed onboarding. Productsup suits enterprise product content syndication. GoDataFeed suits price-sensitive retailers that want self-service control.

Keep the longlist to types you can compare. Shopping Feed suits marketplace-first sellers. CedCommerce suits Shopify and Magento sellers needing marketplace connectors. ChannelAdvisor, now part of Rithum, suits established sellers wanting a managed service. Shopify's native Google and YouTube app suits small catalogues with simple needs.

Map the tool types you actually need

Feed work splits into four jobs: extraction, connectors, monitoring and optimisation. Few providers do all four well. Naming the jobs stops you buying overlap.

Extraction and transformation tools pull data from your platform, clean it and shape it into channel formats. PIM systems such as Akeneo and Sales Layer can feed this step. Channel connectors push data to Google, Microsoft, Meta and marketplaces. DataFeedWatch, Channable and GoDataFeed sit here.

Monitoring tools watch for disapprovals, price mismatches and stock errors. Feedonomics and Productsup sell this as part of a managed or enterprise service. Optimisation tools handle titles, imagery and bid inputs. Channable and Productsup both offer optimisation features.

A small retailer can often cover extraction and connectors with a native integration plus a spreadsheet. A retailer running seasonal promotions across five channels usually needs a dedicated platform for all four jobs.

Imagery is the job teams forget. Google's Product Studio for Merchant Centre lets merchants generate and edit product imagery inside Merchant Centre, which changes what you need from a feed provider's creative module. The Product Studio for Merchant Centre guidance explains what the tool covers and where it stops.

Write down which of the four jobs you will keep in-house. That single decision removes half the vendor field. Ask each provider which jobs are native and which sit with partners. A connector built by another company can lag behind a channel specification.

Build a scoring rubric before the demos

A rubric forces the same questions on every supplier and gives you a written record if the decision is challenged later. Weight the categories to match your requirement list, then score each provider out of five.

Score each category, multiply by the weight and total the result. A provider that wins on price but scores two on data handling will rarely survive a peak season.

Run the rubric on named providers. Score DataFeedWatch, Channable and Feedonomics on the same sheet. Then score GoDataFeed and Productsup. The comparison becomes concrete rather than abstract.

Run the rubric twice: once on the demo, once after a trial feed. Demo scores are claims. Trial scores are evidence. The gap between the two is the most useful number in the exercise.

Score on your own before the group discussion. Written scores from each stakeholder expose disagreements that a meeting tends to smooth over. Keep the rubric in the contract file.

Check the channel rules each provider claims to support

Every channel publishes its own specification, and the provider's job is to keep up with changes. Ask how it tracks them.

Google's Content API for Shopping is the documented route for programmatic feed management, and the Content API quickstart shows the request structure and authentication model. A provider that cannot describe how it uses the API, or an equivalent, is likely pushing files and hoping.

Microsoft Advertising runs a separate API with its own campaign and product structures. The Microsoft Advertising API documentation is the reference to check when a provider claims Microsoft coverage, because the feed and the ad campaign are managed in different places.

Get the answers in writing. Verbal assurances about channel coverage are the most common source of post-signature disputes. The shopping feeds supplier comparison in England shows how the main provider types differ on exactly these points, which is useful when you are checking claims against the field.

Test data handling on your own catalogue

A trial on a sample of your products beats any demo. Insist on it before commercial terms are agreed.

Pick 200 products that represent your worst data: bundles, variants, products with long titles, and items that change price daily. Load them through the provider's onboarding process and watch what happens.

Record the error rate, the time to first successful sync and the number of manual fixes needed. Ask the provider to explain three errors you cannot resolve yourself. The quality of that explanation tells you more about support than any service level agreement.

Document the trial as you go. Screenshots of the error queue and the timestamps of each sync are stronger evidence than a summary written from memory.

Check what happens at the edges. Can you override a category mapping? Can you exclude a product without deleting it? Can you see a change history for a single attribute?

The shopping feeds best tools: data and sources guide breaks down how to assess feed data quality before you commit. It pairs well with a live trial on your own catalogue.

Put compliance in the procurement pack

Feed content is advertising. Titles, images and promotional text all fall under the advertising codes, and the Advertising Standards Authority's Copy Advice service exists so retailers can pre-clear campaigns before they run. The ASA Copy Advice service is worth using when a feed carries claims that a provider's template might repeat across thousands of listings.

Direct marketing is a separate regime. If your feed work feeds email or SMS campaigns, consent and suppression rules apply. The ICO's direct marketing advice generator walks through the questions that determine whether your planned contact is lawful.

Ask each provider two compliance questions. Where does consent data sit in the workflow? And can you export an audit trail of who changed a feed attribute and when?

Name the person who signs off promotional copy in feed templates. A template repeats one mistake across every listing that uses it. A provider that treats compliance as a marketing afterthought will hand you a problem at the worst moment. Put the answers in the rubric and score them.

Price the whole contract, not the licence

The monthly licence is the smallest number in most feed contracts. Setup fees, data migration, custom connector work and overage charges often exceed it in year one.

Build a three-year cost model. Include licence, onboarding, professional services, channel fees outside the provider, and weekly internal time to run the feed.

For example, a team paying £400 a month might spend £6,000 on onboarding and £9,000 a year in internal hours, so the licence is under a third of the real cost.

Ask about overage. DataFeedWatch and GoDataFeed publish plans that scale with SKU count or feed updates. Channable prices by items and channels. Feedonomics and Productsup usually quote per project or enterprise scope. Get thresholds in the contract, not in a sales email.

Ask whether the quoted monthly price holds for the contract term, and what notice period applies at renewal.

Check the exit terms. Can you export your mappings, rules and history in a usable format? How long after termination does access continue? A provider that will not answer this before signature will not answer it afterwards.

The shopping feeds vendor due diligence in England checklist covers the financial and contractual checks that sit alongside the cost model, including what to ask for when a supplier is small or recently funded.

Plan the implementation before you sign

Implementation is where most feed projects slip. The contract should name the onboarding steps and who owns each one.

Agree a data freeze window for the initial load, so upstream changes do not corrupt the first sync. Name the person on your side who approves category mappings. Book the first channel go-live as a limited launch, with a rollback plan.

Set a review date 30 days after go-live. At that review, compare the trial error rate with the live error rate. If the live number is worse, the onboarding was incomplete, not the software.

Ask for training that covers failure modes, not just the happy path. Staff need to know how to handle a mass disapproval, a broken connector and a price feed that stops updating on a Saturday.

Agree an escalation route before go-live. A shared inbox is not a support process when a feed stops updating during a sale. A provider that resists a written implementation plan is telling you how support will feel later.

The shopping feeds software selection in England guide sets out how to run the selection process itself, from requirement list to signed contract, if you need the fuller sequence.

Common questions

How many providers should we shortlist?

Three is usually enough. Two makes comparison thin and four or more stretches the trial phase beyond what most teams can run properly. Score all three on the same rubric.

Do we need a separate tool for each sales channel?

Not always. A single feed platform with strong connectors can cover Google, Microsoft and the main marketplaces. The exception is a channel with unusual data requirements, where a specialist tool often earns its fee.

What should the trial period cover?

At least one full price change cycle and one promotional event. That exposes how the provider handles bulk updates, disapprovals and support requests when volume is high.

Can we keep feed management in-house?

Yes, if your catalogue is small and stable. Once you run multiple channels with frequent price changes, the internal hours usually exceed the licence cost of a platform, so model both options before deciding.