The days after a release are when you most need a clear answer, and when the signals are hardest to read. Star ratings move slowly, review volume jumps around, and one loud complaint can look like a trend. Your AI assistant can compare the new version with the one before it on equal terms and tell you how it landed. In this article, you'll learn what this check includes, how to ask for it, and how to read the verdict.
Overview
When you ask how a release went, your assistant compares the new version with the previous one and returns:
- A verdict up front: landed, mixed, or backlash, with the counts behind it
- Review volume, average rating, and sentiment for each version, measured over the same number of days so the comparison is fair
- New vs. carried-over complaints: themes that appeared with the new version, kept separate from issues users already had before
- Early warnings: a complaint theme that's growing fast gets flagged even if the star rating hasn't moved yet
This is a read-only request. Nothing in your account changes.
Why sentiment comes first
Your average rating is built from every review over time, so a bad release can take days to show up in it. What people write changes much faster. That's why the assistant looks at sentiment and complaint themes, not just stars, and flags a rising issue before it reaches your rating.
How to ask
Once AppFollow MCP is connected to your AI assistant, you can ask in plain language. For example:
We shipped 6.2 nine days ago. Did it go well?More things to ask
- "Average rating and review volume for 6.2 vs 6.1, by store."
- "What are people complaining about in 6.2 that nobody mentioned in 6.1?"
- "Any spike in 1-star reviews in the last 48 hours? Summarize the likely cause."
- "Sentiment split for the last 14 days vs the 14 before."
💡 Haven't connected yet? See AppFollow MCP and AI Toolkit for setup steps for each supported client.
Reading the verdict
| Verdict | What it means |
|---|---|
| Landed | Rating and sentiment held or improved, and no significant new complaints appeared |
| Mixed | Some things improved, but at least one new complaint theme is worth a closer look |
| Backlash | Rating or sentiment dropped, driven by issues that came with the new version |
F.A.Q.
Why compare the same number of days for each version?
A version that's been live for months will always have more reviews than one released last week. Comparing equal windows, for example the first nine days of each version, keeps the numbers fair.
What if I don't remember exactly when the version shipped?
Just name the version. Each review includes the app version it was written on, so the assistant can see when reviews for that version started coming in.
Why are sentiment or themes missing for some reviews?
Sentiment labels and themes come from AppFollow's review analysis, which is available on some plans. Without it, the assistant reads the review text directly, so you'll still get a summary of what changed.