Hightouch

Hightouch is a marketing technology company represented in this wiki by its article on AI Decisioning and the personalisation gap in modern marketing.

Key facts

• Type: Company / marketing technology platform
• Source role: Publisher of the article “Under the hood of AI Decisioning, part one: Overcoming the personalization gap” ^[src-023]
• Key framing: 1:1 personalisation requires learning-based decisions for each customer rather than static segments, journey builders, and rules ^[src-023]
• Concepts introduced here: AI Decisioning, Personalisation Gap, Reinforcement Learning for Marketing, Agentic Marketing, Marketing Bandit Optimisation, Customer Feature Matrix, Treatment Personalisation, Contextual Bandits [src-023, src-024, src-025, src-026]

What it does

The source positions Hightouch’s AI Decisioning around a shift from rule-based marketing automation to learning-based marketing. Instead of asking marketers to predefine every journey branch, AI agents optimise customer-level experience decisions across goals, dimensions, inputs, and guardrails ^[src-023].

The article also connects this marketing problem to the bandit cluster already in the wiki: reinforcement learning provides the broad framework, Multi-Armed Bandits balance exploration and exploitation, and Contextual Bandits use customer context to choose the best option for an individual rather than for an average segment ^[src-023].

In the second article, Hightouch expands the reinforcement-learning layer. It maps marketing into an agent-action-environment-reward loop: the agent chooses content, channel, timing, offer, and frequency; customers and their data are the environment; business outcomes such as purchases, reactivation, clicks, and lifetime value become rewards ^[src-024].

In the third article, Hightouch narrows in on multi-armed bandits as the allocation layer for marketing. The article explains why testing subject lines, send times, offers, and creatives quickly becomes combinatorial, and how bandits dynamically shift send volume toward better-performing actions while keeping some exploration active ^[src-025].

In the fourth article, Hightouch completes the AI Decisioning stack with contextual bandits. It describes how customer data becomes a Customer Feature Matrix, how models predict expected reward for customer-action combinations, and how the system learns both audience-level patterns and individual exceptions ^[src-026].

Related

• See also: AI Decisioning, Personalisation Gap, Reinforcement Learning for Marketing, Agentic Marketing, Marketing Bandit Optimisation, Customer Feature Matrix, Treatment Personalisation, Contextual Bandits, Multi-Armed Bandits

Source references

• ^[src-023] Hightouch — “Under the hood of AI Decisioning, part one: Overcoming the personalization gap”
• ^[src-024] Hightouch — “Under the hood of AI Decisioning, part two: Reinforcement learning”
• ^[src-025] Hightouch — “Under the hood of AI Decisioning, part three: Multi-armed bandits”
• ^[src-026] Hightouch — “Under the hood of AI Decisioning, part four: Contextual bandits”