Gift shoppers who cannot phrase a keyword search get a curated recommendation out of a conversation instead.
Ocula Product Discovery Optimization Agents
Product and content teams delegate catalog-wide enrichment and copy to a governed agent team instead of working SKU by SKU.
An always-on product optimization flywheel where a team of agents enriches each SKU, generates product content, activates it across every channel, and refreshes it based on performance and market signals.
What the agents do
Takes a launch brief, scopes the work, applies the retailer's source, taxonomy and brand rules, runs its specialist agents, and escalates only the exceptions.
- Built by
- Ocula, which builds and runs the agent team on its own platform
- Contributed by
- Ocula (Builds and runs the agent team)
- Industry
- Retail
- Audience
- B2B, B2C
01
The problem
Why this exists
Retailers now compete for discovery in AI search, shopping assistants and their own onsite search, and each channel reads a product record differently. That makes the work per SKU: attributes that match the retailer's taxonomy, descriptions written in the shopper's language, and product rules applied the same way across thousands of items.
Read more
Large catalogs change faster than lean product-data, content and merchandising teams can fill them. The retailers quoted in Ocula's submission describe the size of the gap: one building-supplies retailer adds around 10,000 products a month and completes around 1,000; a health and beauty retailer wants product set-up to fall from 14 weeks to 7; a specialist retailer spends 30 to 40 hours a week on manual attribute entry.
General-purpose assistants speed up single tasks, but retailers report uneven classification, invented detail, and product rules that are not applied consistently at scale. Keeping sources traceable and rules enforced across a whole catalog is a different job from writing one description well.
The agents are built for multi-brand retailers in home improvement, hardware, electronics, specialty retail, building supplies, multi-brand fashion, farm and fleet, and multi-category retail, where catalogs are large, complex and change constantly.
02
Outcomes
What changes when this is in place
- AO World increased organic traffic by 15% after optimizing its products for AI search, measured in its own analytics.
- Blain's Farm & Fleet reports a 95% gain in team efficiency, alongside a 15% increase in conversions.
- The Sole Supplier lifted traffic to product pages by 24%, with average rankings improving by about 1.5 to 2 positions.
Read more
- Product-description coverage at The Sole Supplier rose from 25% to 100%, including around 80% of products that had no description.
- A 30-day controlled test at The Conran Shop, 95 enriched SKUs against 95 held back, attributed 13.3% more organic traffic to descriptions and 9.4% to metadata.
- The same test attributed £4,000 of ChatGPT-referred sales in 30 days to the enriched products.
03
How it works
Architecture
The agents sit alongside a retailer's existing systems as a product-data and content layer. Catalog data goes in, enriched attributes and channel-ready copy come back approved, and the retailer's product information system stays the system of record.
Read more
Work enters as a file upload or over a direct connection and is mapped into a shared product model. An orchestration agent reads the requested outcome and applies the retailer's rules: which sources are trusted or blocked, which taxonomy and attribute values apply, which units and formats are permitted, and which specialist agents handle each stage. Governance, enrichment, copy, and QA then run in order, with a human review step wherever the rules ask for one.
The enrichment agent keeps the evidence behind each attribute, so QA checks claims against sources instead of plausibility, repairs and rechecks what a rule can fix, and escalates only the exceptions. Rules and working preferences persist across sessions, so when product information changes only the affected products are regenerated. A performance agent closes the loop by naming the products losing visibility in discovery and what should change.
Tech stack
Google CloudDurable workflow orchestrationModel gateway across multiple model providersCSV, JSON and XLSX catalog files
Resources
Next step
Request an intro
Ocula is behind this solution. Ask about delegating catalog enrichment to an agent team, and the MACH Alliance will put you in touch with the person who can answer.
Not ready to talk to anyone? MACH Insider is the Alliance mailing list. New solutions, program openings, and the research behind them, once a month.
Everyone credited on this solution
- OculaBuilds and runs the agent team
Every organization named here has a page listing the rest of its work on the Hub.
Related solutions
A stated goal becomes a finished look, so a shopper who cannot phrase a query still gets a usable answer.
Solution Studio Shopping Greeter
Solution Studio
A shopping assistant that lifted conversion 25%, taken from a split test to all traffic.
Ask Frasers Shopping Assistant
Verification
Contributor submission
The detail on this page comes from the contributor's own account of their own work. The MACH Alliance reviews a submission before publishing it and does not independently verify the figures in it.
What this page does not establish
- No independent audit of the retailer results is published; the figures come from each retailer's own analytics.
- The submission names AO World, Argos, The White Company, Kingfisher Group, and Blain's Farm & Fleet as live, and gives results for three of them.
- Pricing, implementation time, and the share of each catalog under agent management are not stated.
Where this came from
Written from Ocula's submission to the Agent Solutions Hub, an eight-page PDF titled 'Ocula Technologies x MACH Agents'.
Status: Live in productionAdded Oct 30, 2026A listing is an editorial decision, not a MACH Alliance certification.
Built something worth showing?
Put your agentic solution in front of the enterprise leaders who use the hub to decide what to build next. Submissions are open to MACH Alliance Members, Collaborators, or Ambassadors.