Choosing between Standard and Enhanced
Last updated: August 20, 2026
The Conversation Editor’s AI Model section lets you pick between Standard and Enhanced. Both drive the same chat flow — same photo card logic, same support routing, same draft generation — but they run on different underlying models with real quality trade-offs.
The short version
| Standard | Enhanced | |
|---|---|---|
| Review quality | Good | Noticeably better |
| Context awareness | Basic | Stronger |
| Sentiment detection | Good | Better edge-case handling |
| Default | No | ✅ |
If you’re not sure, stay on Enhanced. It’s the default because the review-quality difference is meaningful for most stores.
What “better” actually means
Follow-up questions
When a customer says “I love it,” Enhanced tends to ask a specific, thoughtful follow-up (“What’s been your favourite detail?”) where Standard might ask something more generic (“Can you tell me more?”). Specific questions produce specific answers, which produce better reviews.
Sentiment on ambiguous messages
If a customer writes “yeah, it’s fine I guess,” Enhanced is more likely to read that as lukewarm and follow up to surface what’s holding them back. Standard may take “fine” at face value and wrap the review early.
Draft writing
The final review is drafted the same way in both models — from the customer’s own words, preserving their exact phrasing without changing meaning (see How the review draft is written). The difference between Standard and Enhanced is in the conversation — the follow-up questions, sentiment reading, and support detection above — not in how the draft is written.
Support detection
Both models catch obvious support signals (“haven’t received it”, “it’s broken”). Enhanced is better at catching subtler ones — customers who are frustrated but haven’t used a keyword, or who buried a real issue inside a mostly-positive message.
When Standard makes sense
- You’ve tested both and genuinely prefer Standard’s voice for your brand.
- You’re on a legacy plan and watching your allowance. Standard conversations never count toward a legacy plan’s Enhanced-review allowance or trigger auto-recharge. See Billing on legacy plans.
For most stores, Enhanced produces reviews that are worth the difference.
Switching between models
Switch at any time from the Conversation Editor. Changes apply to conversations that start after you save — conversations already in progress keep the model they started with.
You can experiment freely. Try Standard for a week, look at the review quality, switch back to Enhanced if the quality drop isn’t worth it.
What doesn’t change between models
- Photo card behaviour — same gate logic, same sentiment check, same “ask once” rule
- Support routing — same triggers, same cooldown
- Draft structure — same voice preservation rules
- UI and colours — whatever you configured in Customizing the chat
The model only affects the quality of the AI responses themselves, not the flow around them.
Billing impact
On current plans: Enhanced conversations use a credit from your plan’s included AI-collected reviews, and only when the customer submits. Standard-model conversations never draw credits, on any plan. See Billing & limits.
On legacy plans, the model you pick changes what you’re charged. Enhanced reviews count against your allowance; Standard reviews don’t. On the original reviews / pro / enterprise ladder that allowance is 200 / 1,000 / 2,000 conversations depending on tier; on the later Starter / Growth / Scale ladder it’s a monthly credit pool. See Billing on legacy plans for both.