Reviews & UGC
Social proof closes sales, so Zubby AI pulls your product reviews into its corpus and cites them back to shoppers in chat — “this one has 4.8 stars across 200+ reviews, and buyers say it runs warm.” This page covers where reviews come from and how the agent uses them.
Zubby has three distinct “review” surfaces in the dashboard, and they answer different questions. It’s worth knowing which is which:
| Page | What it’s about |
|---|---|
| Product reviews | Star-rating reviews of your products, ingested from Yotpo / Judge.me / Loox / Okendo / native sources. The AI cites these. |
| Reviews & Feedback | Shopper thumbs-up / thumbs-down on the AI’s answers — quality signal for the agent, not product reviews. |
| Review Queue | Conversations the AI may have mishandled, queued for you to inspect and feed retraining. |
Product reviews (UGC the agent cites)
This is the page most merchants mean by “reviews.” Open Catalog → Product reviews. It’s a read-only viewer over the reviews Zubby has ingested, so you can see exactly what’s in the corpus the AI quotes from.
Where reviews come from
Connect a review provider under Integrations and Zubby starts pulling reviews automatically. Supported sources include:
- Yotpo
- Judge.me
- Loox
- Okendo
- Native — reviews already attached to your store.
Each ingested review stores its rating, title, body, author, verified- buyer flag, source, language, the linked product, and the original external timestamp.
What the page shows
Three summary cards sit up top:
- Total reviews — overall count, with how many are from verified buyers.
- Average rating — mean stars across every source.
- Sources — how many providers are feeding reviews, listed by name.
Below that you can filter by rating (1–5 stars) and by source, then browse the latest reviews — newest external timestamp first. Each row shows the star rating, a source badge, a verified-buyer badge where applicable, the language, the review title and body, the author, and a link to the product’s catalog detail page.
Read-only by design
How reviews feed the agent
Ingested reviews join the same retrieval layer as your catalog and knowledge base. When a shopper asks “is this good for wide feet?” or “how’s the quality?”, the agent can surface relevant review snippets as evidence instead of making unsupported claims. Verified-buyer reviews carry more weight as trustworthy proof.
Reviews reduce hallucination risk
Reviews & Feedback (ratings on AI answers)
Separate from product reviews, Reviews & Feedback (under the AI feedback area) aggregates the thumbs-up / thumbs-down ratings shoppers leave on the AI’s own messages in the widget. It’s a coaching surface for the agent, and it’s read-only — you act on issues from the conversation, not from here.
It shows:
- Total feedback, with the change versus the prior 7 days.
- Positive and Negative percentages and counts.
- 7-day volume of ratings.
- A recent feedback list — each entry quotes the rated message, shows who left it, and deep-links to the full conversation.
- Negative themes — common words pulled from negative- feedback notes, so repeated complaints stand out.
The intended workflow: spot a recurring negative theme, open the conversation for context, then close the gap by adding the missing fact to your knowledge base so it doesn’t happen again.
Review Queue (conversation moderation)
Review Queuesurfaces whole conversations the AI might have mishandled so you can flag them for retraining. It’s about chat quality, not product reviews. Filter chips narrow the list to:
- Low CSAT — conversations rated 2/5 or lower.
- Negative sentiment — flagged negative by the agent.
- Low confidence — confidence score at or below 0.4.
- Escalated — handed off to a human.
Each card shows the conversation summary, CSAT, sentiment, and confidence badges, plus a link to open the full thread. Use it as a weekly QA pass over the AI’s worst interactions.
Gotchas
- No product reviews showing? Connect a source under
Integrationsfirst — nothing ingests until a provider is linked, and ingestion runs on the background worker. - Don’t confuse the three pages.Product reviews = UGC the AI cites; Reviews & Feedback = thumbs on AI answers; Review Queue = flagged conversations.
- Edit reviews at the source. Zubby mirrors the provider, so corrections and replies belong in Yotpo, Judge.me, Loox, or Okendo.
Related
- Catalog & product sync — reviews link back to product detail pages.
- Product recommendations — social proof shown alongside suggestions.
- Knowledge base — turn recurring negative feedback into permanent fixes.
- Integrations — connect your store, then add Yotpo, Judge.me, Loox, or Okendo.