AI Insights is the review-intelligence side of the platform. Instead of manually reading through large volumes of reviews to understand what customers are really saying, the platform analyzes them for you and turns them into structured signals.
That means the agency can move from reacting to individual reviews toward understanding patterns, weak spots, and opportunities across the whole review set.
Open it from Dashboard -> AI Insights. The Insights tabs (Analytics, QR Analytics, AI Insights, Search AI) sit on the Dashboard page, not in the sidebar.
What AI Insights is
#AI Insights is an AI-powered dashboard that analyzes customer reviews and produces structured intelligence from them. The goal is not just to summarize reviews, but to extract useful operational signals from them.
Steps and details
- Review score beyond simple star averages
- Sentiment analysis with trends
- Category scoring for business-relevant themes
- Keyword extraction across positive and negative language
- Recommendations with action priority
- Source-level review breakdowns
- Alert detection for meaningful change
Who can use it
#From the agency side, AI Insights is available when the platform has a valid default AI provider configured. For customers, access is controlled through the custom plan feature settings like other higher-value features.
Steps and details
- Enabled gives the customer full access
- Upgrade keeps the feature visible but gated
- Hidden removes it from the customer experience entirely
Provider requirement
#AI Insights works with either OpenAI or OpenRouter. The platform simply needs one default AI provider that is ready and working.
Steps and details
That makes it easier to adopt than Search AI, which specifically needs OpenRouter.
- OpenAI works
- OpenRouter works
- One default AI provider is enough
Minimum review requirement
#AI Insights needs at least 5 text reviews to produce a useful analysis. Star-only reviews are not enough on their own because the feature depends on written content to detect themes and sentiment patterns.
How analysis is triggered
#There are three practical ways AI Insights gets created: manual generation, generating for all locations, and the automatic weekly run.
Steps and details
- Manual generation when you click the generate action
- Generate all locations for multi-location organizations
- Weekly automatic generation for organizations with enough text reviews
- Each organization is capped at 3 analyses per day, and failed or in-progress attempts count toward the cap, so repeated retries on a broken provider will exhaust the day's allowance
What happens during analysis
#The system first gathers the relevant text reviews, computes the baseline metrics such as ratings and response rates, then sends the review set through the configured AI model to classify sentiment, identify categories, extract keywords, and generate recommendations.
Steps and details
If the review volume is large, the reviews are handled in batches of up to 200 reviews, with each review truncated to its first 300 characters before analysis, then merged into one final analysis.
- Review collection
- Sampling where necessary
- Baseline statistical calculation
- AI analysis and recommendation generation
- Alert detection after the analysis is complete
How to read the dashboard
#The dashboard is organized to move from top-line summary into deeper explanation. The review score is the quickest directional number, but the real value comes from category scoring, source differences, and recommendation quality.
Steps and details
- Review score for the top-line view
- Sentiment chart for positive, neutral, and negative distribution
- Score trend over time
- Quick stats for review count, average star rating, and response rate
Categories, keywords, and recommendation quality
#AI Insights turns review text into business-relevant themes such as friendliness, cleanliness, wait time, or quality of work. This is where the analysis becomes actionable instead of merely descriptive.
Steps and details
The recommendation layer then converts those patterns into immediate actions, shorter-term improvements, and longer-term strategic ideas.
Source and location analysis
#The feature can break insights down by review source and by location. This matters because one platform may skew more negative than another, and one branch may be performing much worse than the rest of the business.
Steps and details
That makes AI Insights especially useful for agencies managing multi-location clients or clients with review activity spread across several platforms.
Historical insights and reporting
#Previous analyses are stored so you can compare trends over time. AI Insights can also feed scheduled reporting, which makes it easier to show progress instead of only presenting one isolated snapshot.
Alerting and what it watches for
#AI Insights automatically checks for meaningful issues after each analysis, such as declining sentiment, urgent negative themes, low satisfaction, or large review-response gaps.
Steps and details
When important issues are detected, the platform can send notifications with severity-based messaging so agencies know where to focus first.
AI Insights vs Search AI
#Steps and details · includes a table
| Area | AI Insights | Search AI |
|---|---|---|
| Primary purpose | Analyze your actual review data | Monitor visibility across AI search platforms |
| Data source | Your customer reviews | AI-generated search responses |
| Provider requirement | OpenAI or OpenRouter | OpenRouter only |
| Best for | Service quality, reputation patterns, and review strategy | External discoverability and AI recommendation visibility |