Last updated: 11 July 2026
Why this list uses an RFP lens
A list of Shopping agencies is useful only if it helps a retailer run a better selection process. Public websites can show whether a provider offers feed work, Merchant Center support or retail media expertise; they cannot show which operator will be assigned, what the retainer contains or how that team will handle your catalogue. This guide therefore combines a short evidence-based list with a practical RFP exercise.
Disclosure: Upscale Digital publishes this page and is included. Its inclusion is conflicted. No provider paid to appear, no private fee has been estimated and the sequence below is not a performance ranking.Evidence threshold
We checked current first-party agency pages on 11 July 2026. Inclusion required explicit Shopping, ecommerce, product-feed or retail paid-search capability. We ignored generic awards and did not turn self-reported case-study outcomes into scores because baselines, attribution and promotional periods differ.
The six agencies represent different models: founder-led specialist, integrated performance agency, retail enterprise partner, focused ecommerce PPC agency, feed-platform specialist and integrated search agency.
Shortlist at a glance
| Provider | Published strength to verify | Likely procurement fit |
|---|---|---|
| Upscale | Shopping, PMax, tracking and margin-led bidding | Senior specialist access |
| Impression | Feed, stock, margin, creative and retail reporting | Integrated retail performance |
| Summit | Retail paid search, CSS, feed tooling and omnichannel | Complex retail operation |
| DPOM | Dedicated UK Shopping and ecommerce management | Growing online retailer |
| Onefeed | Feed technology plus Shopping management | Catalogue-data bottleneck |
| Found | PMax, feed optimisation, paid search and CRO | Integrated search and conversion |
1. Upscale Digital
Luke Bright’s profile states a focus on Google Shopping, PMax, ecommerce PPC, conversion tracking and margin-led bidding. The founder-led delivery model is worth considering when the retailer already has internal merchandising and development resources but wants senior Google Ads ownership.Ask Upscale to show where feed changes are executed, how Merchant Center alerts are covered and which creative or engineering tasks require client support. Direct access is valuable only when the surrounding responsibilities are clear.
2. Impression
Impression’s Shopping page describes feed optimisation, use of margin and stock data, creative support and product-level retail reporting. This is unusually relevant public detail for a multi-disciplinary Shopping brief.The proposal should name the feed, analytics and creative specialists assigned and itemise any separate projects. Do not assume every agency capability is included in a paid-media retainer.
3. Summit
Summit’s retail paid-search proposition lists Shopping, PMax, first-party data, profit optimisation, local inventory, attribution, FeedManager and its Productcaster CSS.This depth suits larger retailers, but proprietary tooling requires due diligence. Ask how CSS charges work, whether feed rules can be exported and what data or campaign changes are needed to leave.
4. DPOM
DPOM publishes a dedicated ecommerce and Google Shopping service for UK online retailers. Its narrower model may fit stores that need managed Search and Shopping without a large integrated team.Use a live SKU sample to test whether the team can move from feed diagnosis to implementation. Confirm responsibility for policy issues, source-feed changes and product-level reporting.
5. Onefeed
Onefeed combines feed technology and Shopping management. That model can be useful where catalogue transformation is the central constraint or data must feed several marketing channels.Ask how media insights become feed rules, who owns those rules and whether reporting remains available outside the platform. Technology should reduce operational friction without creating avoidable lock-in.
6. Found
Found lists feed optimisation, PMax, paid search, YouTube and CRO in its PPC proposition. It may suit a retailer that wants product-page and broader search work connected to Shopping.Ask the delivery team to trace one weak product from query and feed evidence to landing-page and campaign action. That exercise distinguishes genuine integration from several services sharing one logo.
The paid RFP exercise
Select 20 products: top sellers, high-margin products with low visibility, high-click products with weak sales and disapproved items. Give each finalist the same product attributes, twelve weeks of performance, stock cover, margin band and return rate. Remove anything commercially sensitive that is not needed.
Pay finalists for substantial diagnostic work. Ask them to identify uncertainties, not produce a free account rebuild. Compare the quality of questions, sequencing and tradeoff reasoning. A useful team will separate eligibility, demand, feed relevance, bidding and page problems before suggesting changes.
What the proposal must allocate
| Workstream | Named owner | Evidence required |
|---|---|---|
| Source and supplemental feeds | Retailer, agency or feed vendor | Rule location, approval and rollback |
| Merchant Center | Operational owner | Alerts, response time and appeal process |
| PMax and Standard Shopping | Media strategist | Product overlap, controls and experiment plan |
| Commercial data | Finance or trading owner | Margin, returns, stock and refresh cadence |
| Creative and pages | Brand, agency or developer | Deliverables, source files and test backlog |
A weighted scorecard for the final two
Use a weighted scorecard only after the working session. Weight Shopping delivery and proposed-team quality more heavily than agency brand. A useful starting allocation is 25% feed and Merchant Center competence, 20% commercial measurement, 20% named-team quality, 15% campaign and PMax reasoning, 10% governance and ownership, and 10% working chemistry. Change the weights to match the brief.
Add an evidence grade beside every score: verified from a current first-party source or reference; demonstrated in the working session; promised in the proposal; or unknown. A high score supported only by promises should not beat a slightly lower score backed by demonstrated work.
| Score area | Evidence to request | Failure mode |
|---|---|---|
| Feed and Merchant Center | Live diagnosis and named implementation owner | Advice with no route to publish fixes |
| Commercial measurement | Margin, returns and stock workflow | Revenue-only optimisation |
| Delivery team | Names, roles and allocation | Senior pitch, junior delivery |
| Governance | Account ownership and rule export | Tool or account lock-in |
Three retailer scenarios
£15k monthly media, 300 stable products
A founder-led specialist or focused ecommerce PPC team may be enough if the retailer maintains clean source data and can implement page changes. Paying for enterprise feed tooling would need a specific business case.
£80k monthly media, 12,000 changing SKUs
Feed rules, automated diagnostics, stock and margin data, category reporting and resilience become material. A retail specialist or integrated performance agency may justify a larger scope because small allocation errors affect substantial spend.
Three countries plus physical stores
The brief now includes local inventory, language, shipping, tax, store value and regional trading. Cross-market governance and analytics may matter more than a boutique’s direct access. These figures are hypothetical complexity examples, not provider minimums or fee estimates.
First-90-day evidence
By day 15, expect an access map, Merchant Center issue register, feed ownership diagram, conversion reconciliation and category baseline. By day 45, expect high-confidence eligibility fixes, documented feed tests and an agreed PMax/Shopping architecture. By day 90, expect test readouts, product actions, unresolved client dependencies and a next-quarter investment plan.
Do not grade the agency on immediate ROAS alone. Promotions, seasonality, brand demand and stock changes can dominate a short comparison. Grade whether measurement is more trustworthy, product eligibility is understood and decisions can be traced to evidence.
Reference calls that reveal delivery quality
Do not ask references whether the agency is “good.” Ask who actually worked on the account, how often senior people attended, what happened during a feed or Merchant Center incident, which recommendation the client rejected and how handover was handled. Choose references with similar catalogue and organisational complexity, not merely a famous logo. Record whether the proposed individuals were involved in the referenced work. Agency case studies usually describe a successful period; a reference call can expose the ordinary operating experience between major wins.
Sources
- Google: issues in Merchant Center
- Google: product reporting
- Google: Standard Shopping versus PMax experiments
Agency links above are first-party service descriptions, not independent endorsements.