How Insights Work
Cybex AI Insights follows a streamlined workflow that takes you from raw data to production-ready analytics. Every insight starts in the Data Hub and can be refined, saved, and operationalized across Cybex BI Module.
Create a Dataset in Data Hub
Select or build a dataset for a specific insight subject from the Cybex Retail AI Data Hub. Datasets are conformed to the platform's unified data model, ensuring consistency across all downstream analytics.
Create Instructions (Prompt) for the Insight
Author a detailed prompt based on the dataset that specifies the desired output formatting, analysis depth, and advanced insight capabilities - including ML-based anomaly detection, statistical analysis, time-series forecasting, and retail-specific KPIs.
Submit Dataset + Instructions to AI LLM
Send the dataset and prompt to the AI engine. The LLM processes the data using advanced retail domain knowledge, applies ML models, runs statistical computations, and generates structured insight results with narrative analysis.
Review Results and Refine the Prompt
Evaluate the AI-generated insights. If additional depth, different groupings, or alternative visualizations are needed, augment and resubmit the prompt. The iterative loop lets you dial in exactly the analysis required.
Create Permanent Queries for Result Sets
For insight result sets you want to operationalize, create additional permanent queries that connect directly to the source dataset. Render these in Cybex BI Module for pivot-style analysis and ad-hoc exploration.
Save Insights as AI Templates
Save refined prompts and their configurations as reusable AI Templates. Templates can be scheduled, shared across teams, and applied to new datasets - ensuring consistent analysis methodology across the organization.
Access Insights in Cybex BI Module
Access all saved insights directly within Cybex BI Module. Live dashboards, scheduled refreshes, and drill-down capabilities make insights available to every stakeholder - from store managers to the C-suite.
Insights by Application Module
Every Cybex retail application generates rich AI insights. Select a module to explore the types of advanced analytics, ML models, and forecasting capabilities available.
Allocation Insights
- Size-curve optimization by store cluster
- Demand-based store tiering and ranking
- Weeks-of-supply balancing and deficit alerts
- Transfer recommendation engine
- New store allocation modeling
Merchandising Insights
- Price-volume-mix (PVM) decomposition
- LY/TY comp analysis with variance drivers
- Margin and markdown optimization
- Category performance scorecards
- Vendor profitability analytics
Sales Audit Insights
- Transaction anomaly detection (z-score, isolation forest)
- Discount compliance and override analysis
- Void/return pattern recognition
- Revenue leakage quantification
- Payment reconciliation automation
CRM & Loyalty Insights
- RFM segmentation and migration tracking
- Churn prediction with intervention triggers
- Customer lifetime value forecasting
- Personalized promotion optimization
- Cohort analysis and retention curves
Assortment Planning Insights
- Cluster-based assortment optimization
- Seasonal buy depth and breadth modeling
- Style performance scoring and ranking
- Attribute-level demand analysis
- Open-to-buy forecasting
MPOS Insights
- Real-time clienteling recommendations
- Cross-sell and upsell analytics
- Associate performance scoring
- Basket analysis and affinity mapping
- Omnichannel fulfillment optimization
Insights Across Retail AI Topics
Our blog series covers 24 deep-dive topics - each a source of specialized AI insights. From forecasting and price optimization to loss prevention and hybrid cloud deployment, every topic maps to actionable analytics in the Cybex platform.
Advanced Insight Capabilities
Every insight generated through the Cybex workflow benefits from built-in advanced analytics capabilities that go far beyond basic reporting.
ML-Driven Anomaly Detection
Isolation forests, z-score analysis, and clustering algorithms automatically surface outliers in sales, inventory, and customer behavior data.
Statistical Analysis
Regression, correlation, variance decomposition, and hypothesis testing built into every insight prompt - no separate tooling required.
Time-Series Forecasting
ARIMA, exponential smoothing, and neural network models for demand, sales, and traffic forecasting with confidence intervals.
Retail-Specific KPIs
Pre-built calculations for sell-through, weeks-of-supply, GMROI, comp sales, AUR, UPT, conversion, and 50+ specialty retail metrics.
Natural Language Narratives
AI-generated executive summaries that explain what happened, why it happened, and what to do next - in plain language ready for leadership.
Reusable AI Templates
Save and share insight configurations as templates. Schedule automated runs, apply to new datasets, and maintain consistent methodology across teams.
Ready to unlock AI-powered insights across your retail operations?
From dataset to dashboard - see how Cybex Retail AI Insights transforms decision-making.