Kayo Sports
1:1 AI decisioning (message, creative, channel, timing, frequency, promo per individual)
Kayo Sports grew subscriptions +14% in FY24 and cross-sell +105% with a 1:1 AI decisioning engine on Braze, taking the number of possible personalized actions per customer from 300 to 1.2 million.
Key points
- 1:1 AI decisioning chooses message, creative, channel, timing, and promo per subscriber.
- On Braze and BrazeAI Decisioning Studio, ten reinforcement learning models.
- +14% subscriptions in FY24, +105% cross-sell.
- Actions per customer from 300 to 1.2 million, evidence B confirmed.
Objective
Increase subscriptions and reduce churn for a sports streaming service by orchestrating a 1:1 relationship at scale rather than fixed-rule campaigns.
The deployment
Kayo Sports, Australia's largest sports streaming service, built its Customer Cortex on Braze and BrazeAI Decisioning Studio. Ten reinforcement learning models trained on its first- and third-party data analyze each subscriber, and the decisioning chooses for them the message, the creative, the channel, the timing, the frequency, and the promotion. Orchestration runs through Canvas on email, push, in-app, and SMS. The number of possible actions per customer went from 300 to 1.2 million.
Results Proof B
Quantified Braze customer story cited by name, corroborated by the Australian martech press Mi3 (specialized, T4). Consistent but without major general press, hence B.
How it works
Documented architectureThe stack in detail
- plateforme Braze customer engagement platform that carries the profiles, the channels, and the orchestration
- outil BrazeAI Decisioning Studio next-best-action decisioning engine: choice of message, creative, channel, timing, frequency, and promo per subscriber
- outil Braze Canvas multichannel journey orchestration on email, push, in-app, and SMS
- llm Dix modeles de reinforcement learning (Customer Cortex) in-house models trained on Kayo's first- and third-party data, which turn behavior into signals for the decisioning
How it runs, concretely
For ops teams-
1Data unification Data team
First- and third-party data (viewing, subscriptions) feed the Customer Cortex.
-
2Modeling Data / AI team
Ten reinforcement learning models turn this data into usable signals.
-
3Next-action decision AI (BrazeAI Decisioning Studio)
Decisioning Studio chooses message, creative, channel, timing, frequency, and promo for each subscriber.
-
4Orchestration and loop AI / CRM team
Canvas delivers on email, push, in-app, and SMS; the customer's reaction feeds the next choice.
Each subscriber's engagement and viewing behavior. The reinforcement learning learns from the reaction to each message; without that feedback, the system falls back on generic rules.
How your customers perceive this type of use
Sourced studiesLe pricing algorithmique est le terrain le plus inflammable : 68% des consommateurs disent se sentir leses quand les marques utilisent le pricing dynamique et 80% jugent plus dignes de confiance les marques aux prix constants (Gartner, 2024). L'equite percue varie selon le secteur : le pricing dynamique n'est juge juste que par 33% a 40% des repondants selon qu'il s'agit de concerts ou de cinemas (YouGov, 17 marches). Le prix personnalise par les donnees individuelles est le plus rejete : 47% des Americains s'y opposent fermement (Consumer Reports, 2024).
Acceptance conditions
- La constance des prix comme signal de confiance : 80% jugent plus fiables les marques aux prix stables (Gartner 2024)
- Le secteur conditionne l'equite percue : le pricing dynamique est mieux tolere pour les cinemas (40% le jugent juste) que pour les concerts (33%) (YouGov 2024)
Red lines
- Le pricing dynamique percu comme abus : 68% se sentent leses (Gartner 2024)
- Le prix individualise a partir des donnees personnelles : 47% d'opposition ferme (Consumer Reports 2024)
- Les frais caches et hausses imprevues, vecus par 79% des consommateurs sur un an et associes a la perte de confiance (Gartner 2024)
Sources: Gartner 2024 · YouGov 2024 · Consumer Reports 2024
How to replicate
Inference, not sourcedData prerequisites
- per-subscriber engagement and consumption data, unified
- enough interaction volume to train reinforcement learning
- multichannel consent
Org prerequisites
- data team able to build and maintain RL models
- CRM team that sets goals and guardrails
- subscription model with churn to manage
Possible stack
- Braze + BrazeAI Decisioning Studio
- any engagement platform with a next-best-action decisioning engine
The plan, step by step
- Step 1Unify per-subscriber engagement and consumption data in the platformDeliverable: Unified customer profile fed continuously
- Step 2Define the metric the decisioning optimizes (retention, cross-sell) and the guardrails (max contact pressure)Deliverable: Documented decisioning goals and constraints
- Step 3Activate the decisioning on a pilot lifecycle (e.g. onboarding or reactivation) with a control groupDeliverable: First next-best-action program in production, measured vs control
- Step 4Broaden the action space: channels, promotions, frequencies, creativesDeliverable: Extended action catalog + incremental measurements per wave
- Step 5Add in-house predictive models (churn, propensity) upstream of the decisioningDeliverable: Proprietary scores plugged into the decision engine
First step: Unify per-subscriber engagement data and define the metric the decisioning should optimize (retention, cross-sell) before opening up the action space.
The same pattern, proven in other industries
Sources
- S1 Kayo Sports leverages AI to create 1:1 messaging at scale (Braze) Interested party archive pending
- S2 Kayo customer and revenue chief says AI-decisioning switch has crushed churn (Mi3) Secondary archive pending
An error, newer info, a source?
This page lives on its accuracy. If a figure has moved, if the deployment has changed, or if you have a higher-quality source, tell us. Every sourced correction is verified before publication.