Overview
Switching decision engines is easier than you think. This guide maps concepts from your current platform to KaireonAI equivalents and outlines a practical migration path for each.Pega Customer Decision Hub (CDH)
Concept mapping
| Pega CDH | KaireonAI | Notes |
|---|---|---|
| Propositions | Categories | Top-level grouping for offers (e.g., Credit Cards, Loans) |
| Actions | Offers | Individual items recommended to customers |
| Strategies | Decision Flows | Visual flow editor with enrichment, scoring, and arbitration |
| Properties | Schema Fields | Customer/account attributes stored in schema tables |
| Adaptive Models | Algorithm Models | Per-offer propensity models with automatic retraining |
| Channels | Channels | web, mobile, email, sms, call-center, etc. |
| Contact Policies | Qualification Rules | Frequency caps, suppression rules, eligibility filters |
| NBA Designer | Studio | The main configuration workspace |
| Interaction History | Interactions | Outcome records (impressions, clicks, conversions) |
| Customer Profile | Customer Schema | Enrichment data loaded at decision time |
Key differences
- No Pega PRPC dependency — KaireonAI runs as a standalone API service; no application server required.
- Formula engine — computed values use a safe expression language instead of Pega expressions.
- REST-first — all operations are available via REST API, not just the real-time channel.
Migration steps
- Export propositions and actions — Map your Pega proposition hierarchy to KaireonAI Categories and Offers. Use the bulk import API:
POST /api/v1/categories/importandPOST /api/v1/offers/import. - Recreate strategies as Decision Flows — Rebuild your Pega strategies in the visual Decision Flow editor. Map each strategy component (enrichment, scoring, arbitration) to the corresponding Decision Flow stage.
- Migrate customer data — Create schemas matching your Pega property definitions, then load data via connectors (S3, Snowflake, Databricks, etc.) or the Data Pipeline module.
- Replicate contact policies — Translate Pega contact policies into KaireonAI Qualification Rules with equivalent frequency caps and suppression windows.
Adobe CDP / Adobe Journey Optimizer (AJO)
Concept mapping
| Adobe CDP / AJO | KaireonAI | Notes |
|---|---|---|
| Segments | Segments | Customer audience definitions |
| Journeys | Journeys | Multi-step customer journey flows |
| Offers (Offer Decisioning) | Offers | Items surfaced to customers |
| Collections | Categories | Groupings of related offers |
| Decisioning | Decision Flows | Rules + ML scoring to rank offers |
| Placements | Channels + Creatives | Channel-specific creative variants |
| Profiles | Customer Schema | Unified customer data model |
| Events | Triggers | Real-time event-driven actions |
| Eligibility Rules | Qualification Rules | Who can receive which offers |
| Sandboxes | Tenants | Isolated environments per team or use case |
Key differences
- No CDP dependency — KaireonAI handles decisioning without requiring a separate CDP. Customer data is loaded via schemas and connectors.
- Simpler pricing — no per-profile or per-event charges; flat platform fee.
- Open APIs — every feature is accessible via REST; no proprietary SDK required.
Migration steps
- Export segments — Recreate Adobe segments as KaireonAI Segments using the segment builder or
POST /api/v1/segments. - Map offers and placements — Import Adobe offers into KaireonAI Offers, and map placements to Channel + Creative combinations.
- Rebuild journeys — Use the KaireonAI Journey builder to recreate Adobe Journey Optimizer flows. Map journey triggers to KaireonAI Triggers.
- Connect data sources — Replace Adobe data feeds with KaireonAI connectors (S3, Snowflake, Kafka, BigQuery, etc.) to keep customer data flowing.
Braze
Concept mapping
| Braze | KaireonAI | Notes |
|---|---|---|
| Campaigns | Runs | One-time or scheduled batch executions |
| Canvas | Journeys | Multi-step orchestration with branching logic |
| Segments | Segments | Audience targeting criteria |
| Content Cards | Creatives | Channel-specific content variants |
| Connected Content | Connectors | External data integrations |
| Currents | Data Pipelines | Event streaming and ETL |
| Custom Events | Triggers | Events that initiate actions |
| Custom Attributes | Schema Fields | Customer data attributes |
| Engagement Reports | Dashboards | Performance analytics and metrics |
| Message Templates | Content Templates | Reusable content with variable substitution |
Key differences
- Decisioning built-in — KaireonAI combines messaging orchestration with ML-powered offer ranking; Braze requires external decisioning.
- Next-best-action focus — real-time personalized recommendations per customer, not just campaign blasts.
- Formula engine — dynamic per-customer computed values without external APIs.
Migration steps
- Export segments and campaigns — Map Braze segments to KaireonAI Segments. Convert campaign logic into Decision Flows (for real-time) or Runs (for batch).
- Migrate Canvas flows to Journeys — Rebuild Braze Canvas multi-step flows in the KaireonAI Journey builder with equivalent branching and delay steps.
- Move content — Import Braze message templates and Content Cards as KaireonAI Creatives and Content Templates.
- Replace Currents with Pipelines — Set up KaireonAI Data Pipelines to replicate Braze Currents event streams. Connect the same downstream destinations (S3, Snowflake, etc.).
General tips
- Start with a pilot — migrate one category or campaign first to validate the mapping before doing a full cutover.
- Use starter kits — apply a pre-built industry template to get a working baseline:
POST /api/v1/starter-kits/{kit}/apply. - Run in parallel — keep your existing platform live while validating KaireonAI recommendations side-by-side using A/B experiments.
- Leverage the API tutorial — see the API Tutorial for detailed request/response examples.