Overview
DinMo Intelligence uses your modeled data to generate predictions and recommendations that you can use in segments, campaigns, and analysis.
Choose an intelligence
Estimate the likelihood that a profile or subscription will perform a defined event within a future time window.
Rank products, offers, brands, or categories; find audiences; distribute a campaign audience between targets.
Estimate customer value and identify customers at risk of becoming inactive.
Requirements and output formats vary by intelligence. Start with its data prerequisites rather than assuming that every engine accepts the same models or source.
From data to activation
Prepare the data: connect a supported source and define models, identifiers, relationships, and relevant history.
Configure and train: choose the learning inputs and inspect data quality and evaluation results.
Generate outputs: define the business population, constraints, and refresh schedule.
Activate and measure: use published results, check current eligibility, and measure the business outcome.
For Product Recommendations, one intelligence can support several recommendation scenarios. Training learns the model; each scenario defines what to recommend, to whom, and when to refresh. Outputs are not recomputed after every individual customer action.
Start with an intelligence
Event Propensity Scoring: predict a business event for profiles or subscriptions and use the output safely in an activation policy.
Prepare your data for Product Recommendations: supported warehouses, required models, mappings, and data checks.
Choose a recommendation use case: personalized lists, related items, preferences, and audiences.
Build and allocate audiences: overlapping lists versus one assignment per customer.
Understand performance: distinguish delivery, offline relevance, and measured commercial impact.
Intelligence access depends on your workspace subscription and permissions. Contact your DinMo administrator or customer success contact if it is not available.
Last updated