What it Does

• Identifies teams that users have previously bet on (supporting Winners, Losers, or Any interaction).
• Monitors upcoming popular events featuring these same teams.
• Notifies users when their teams are playing again.
• Prioritizes recommendations based on the user's past affinity and current event popularity.

How it Works

1. Track Behavior: Tracks teams users have bet on, recording the outcome (Win/Loss) and recency.
2. Monitor Inventory: Scans for upcoming sporting events that meet a minimum popularity threshold (bet count).
3. Match: Matches users to upcoming events where their team is playing.
4. Filter: Applies filters for specific Markets, Classes, Types, and Odds ranges.
5. Rank:
   • First, ranks the user's affinity teams by score (recency/amount).
   • Second, selects the best specific bet selection for that event (lowest odds within range).

Configurable Parameters

• Audience Type (audience_type): Defines the target segment. Options:
  • 'Winner': Users who won their last bet on this team.
  • 'Loser': Users who lost their last bet on this team.
  • 'Any Prior Winner': Users who have won at least once on this team.
  • 'Any Prior Bet': Users who have placed any bet on this team.
• Timing:
  • min_hours_before_event: Minimum hours before start time to trigger.
  • max_hours_before_event: Maximum hours before start time to trigger.
• Quality & Filters:
  • min_popularity: Minimum unique users betting on an event for it to be eligible.
  • min_proposed_odds / max_proposed_odds: Min/Max decimal odds for the recommended selection.
  • categories: List of sport categories to include (e.g., 'Football', 'Basketball').
  • market_names / class_names / type_names: Lists to filter specific inventory types (or 'Any').

Output Data

• userid: Identifier of the recipient
• team: Name of the team matched (the "affinity" team)
• category_name: Sport category (e.g., Football)

Upcoming Event Details:

• event_name: Full name of the upcoming event
• event_key: Unique identifier for the event (useful for deep links)
• event_date_time: Scheduled start time
• selection_name: Name of the specific bet selection (e.g., "Arsenal to Win")
• selection_key: Unique identifier for the selection
• market_name: Name of the betting market (e.g., "Match Winner")
• market_key: Unique identifier for the market
• type_key: Identifier for the bet type
• odds: Current decimal odds for the selection

Context Fields (Historical):

• last_event_name: Name of the specific event the user previously interacted with
• last_event_date_time: Date and time of that previous event
• last_selection_name: The specific selection they bet on last time
• last_market_name: The market of their previous bet
• last_odds: The odds they got on their previous bet
• last_bet_potential_payout: Potential payout of the previous bet (used for scoring/ranking)

Required Events and Tables from Client Integration

Event Stream:

• Key fields: userid, selection_name, bet_status, event_name, bet_action_time, brandkey, event_date_time.
• Requires BET_SETTLED and BET_PLACE events.
• Requires betslip_status (with 'won' and 'loss' values).

Inventory Snapshot:

• Key fields: eventdatetime, categoryname, selectionname, selectionkey, eventname, denprice, numprice, decprice, eventkey, marketname, marketkey, typename, typekey, classname, classkey.