AI is a tool. Knowing which questions to ask is the work.
JSA integrates AI-assisted analysis into our domain: patron segmentation, reinvestment modeling, and feasibility work — not to replace analytical judgment, but to process larger datasets faster, surface patterns manual review may miss, and stress-test our models against a broader range of variables.
Computational rigor where it earns its keep.
The gaming industry generates enormous volumes of patron-level behavioral data — trip frequency, theo win, Free Play redemption, time-on-device, offer response. AI-assisted analysis lets JSA work across that full picture rather than relying on aggregated summaries that mask critical segment-level dynamics.
Patron segmentation & reinvestment modeling
Pattern recognition across full patron databases
Identifying micro-segments that aggregate reporting would not surface.
Predictive attrition modeling
Flagging at-risk patrons earlier in the decline curve — while reinvestment can still influence behavior.
Free Play ROI optimization
Machine learning tests reinvestment threshold scenarios across segments to surface the allocation that maximizes net revenue within a defined budget.
Anomaly detection in monthly data
Flagging unusual shifts in segment behavior that may indicate competitive incursion or offer-response patterns worth investigating.
Feasibility & market analysis
Gravity model enhancement
AI-assisted calibration of trade area draw estimates using behavioral and mobility data that traditional demographic models do not incorporate.
Competitive scenario modeling
Running a broader range of competitive entry and market-shift scenarios than is practical with manual sensitivity analysis alone.
Revenue projection stress-testing
AI-assisted variance analysis to identify the assumptions in a feasibility model that carry the most risk — and quantify the impact of those assumptions moving against base case.
Our analysts and principal own every conclusion.
AI accelerates JSA's scrutiny and expands the range of research we can attain and scenarios we can test. It does not replace the nuanced judgment required to interpret results in the context of a specific property, market, and competitive set. Every AI-assisted output at JSA is reviewed, validated, and interpreted by our analyst and principal before it reaches a client.
The ±10% feasibility track record was built on exactly that combination — computational rigor and domain expertise applied together.
Data security & client confidentiality
All patron-level data provided to JSA for AI-assisted analysis is handled under strict confidentiality protocols. Client data is used exclusively for the engagement for which it was provided, is not retained beyond the engagement term, and is never used to train external models or shared with third parties. All patron-level records are de-identified prior to any AI-assisted processing; personally identifiable information is removed before data enters analytical workflows.