Decisioning and Next Best Action
Receive a versioned customer context rather than rebuilding customer intelligence, and return one permitted customer action — including no action — with the policy checks and the reason recorded.
Runnable demonstrationintegrated workflow — The stages are joined by persistence, identity and authorisation inside the platform. This does not mean integrated with an operator system.
A journey you can run end to end on a synthetic dataset for your market, today.
Internal, simulated and disconnected. Nothing has been sent to or executed in an operator system.
The Next-best-action and Retention Decisioning offer runs on its own workbench, on a sample for the organisation's market or on a workbook the operator returns; the pack-dropout decision also runs inside Customer Intelligence's journey on the same sample data.
Buyer and operator
- Buyer
- CVM / Product Director
- Operated by
- Decision owners and the approvers named in the policy
The result this product produces
- One ranked, permitted action per customer moment, or an explicit no action
- The policy each decision was made under, identified and versioned
Built today
- Policy checks, option ranking and an explicit no-action outcome run deterministically.
app/_lib/demo/configured-demonstration.ts - The deciding policy is a versioned, typed profile rather than a literal.
app/_lib/contracts/scenario-profile.ts - The Next-best-action and Retention Decisioning offer runs end to end on /operator/workbench/next-best-action: its workbook is registered with schema heuritel.nba-ranking v1 and engine v1.
app/_lib/offers/engines/next-best-action.ts
Still to build
- Client-editable policy, which needs an approval path that does not yet exist for configuration
- Execution: the ranked action is proposed and decided; no channel or charging system receives it
- Observed outcomes and attribution: the run records what was recommended, never what happened
Propositions that include this product
- Managed CVM and ExperimentationWe run your value-management engine and prove what it earned, with your team in the approval loop.
- Next-best-action and Retention DecisioningOne ranked, permitted action per customer moment — including no action — with the reason recorded.
- Customer-care AI Resolution AssistantResolve more contacts at first touch, with the agent in control and every action evidenced.
Streams beneath this product
The workspace for this product
2 workspace screens belong to this product. They open for a member of an organisation whose subscription includes it.
Sources of this description
PRODUCT_02_DECISIONING_NEXT_BEST_ACTION.md§1 Paid product propositionapp/_lib/demo/configured-demonstration.ts— the deterministic evaluation both the screen and the persisted run call