AI adoption across customer-facing functions is ever-growing, and the customer engagement landscape is undergoing a significant transition. Marketing teams are moving away from static, campaign-led engagement models towards systems capable of making contextual, real-time decisions across channels, products, and customer journeys.
The shift is not simply about automating marketing operations. Increasingly, enterprises are looking at AI as a decisioning layer that can help identify the right customer, the right context, the right message, and the right moment, all dynamically and at scale.
In conversation with Express Computer, Jacob Joseph, VP-Data Science, CleverTap discusses how enterprises are rethinking customer engagement strategies, why traditional campaign-based marketing is gradually losing relevance, and how AI systems are evolving from assistive tools into contextual decision-making engines.
Enterprises are moving beyond campaign-centric marketing
According to Joseph, one of the biggest changes underway is the gradual decline of traditional campaign-led engagement strategies.
Historically, marketers spent a large portion of their effort on operational execution, creating campaigns, managing channels, defining segments, scheduling communication, and monitoring delivery cycles. As customer journeys became more fragmented and digital touchpoints multiplied, these manual workflows became increasingly difficult to manage.
“Most of the time, while creating those campaigns and executing those campaigns, a lot of thinking of a marketer was spent on the operational guidance,” Joseph points out. AI, however, is beginning to automate many of these repetitive operational layers. Joseph explained that enterprises are now using AI not only to create campaign variations quickly, but also to orchestrate decision-making across targeting, content, timing, and customer relevance.
“The marketer will come with the problem statement. Then the AI will select the right segment, the right creative, the right product, the right offer, the right channel, and the right time and cadence,” he explains.
This marks a broader shift in how enterprise platforms themselves are evolving, from feature-centric tools towards systems focused on delivering business outcomes.
AI adoption is becoming easier as enterprise mindsets evolve
Joseph believes the enterprise conversation around AI adoption has changed dramatically over the past year.
Earlier, organisations often needed significant convincing before they were comfortable operationalising AI within customer engagement workflows. According to him, the hesitation was not only about technology maturity but also about changing established ways of working.
“We had to really convince the marketer to adopt AI because there was a shift needed in the way they worked,” he says. However, with generative AI becoming mainstream across industries, that resistance is fading rapidly.
“Now, the problem of convincing them to use AI is not there. Now it is how to use it,” Joseph observes.
This change, he believes, is accelerating enterprise experimentation with AI-led engagement systems, particularly in areas such as workflow automation, segmentation, and customer lifecycle management.
Human oversight will continue to remain critical
Even as AI systems become more capable, Joseph says enterprises will continue maintaining human oversight over important customer-facing decisions.
According to him, most organisations are currently progressing through a phased AI adoption journey. Initially, AI acts as an assistive layer, generating recommendations while humans retain final control.
“There is a human in the loop. It’s a co-pilot kind of thing,” he explains.
Over time, as confidence in AI systems increases, enterprises may automate larger portions of engagement workflows. However, Joseph does not expect organisations to hand over complete control to autonomous AI systems, particularly for sensitive or high-impact decisions.
“There will be certain critical business-level implications which they wouldn’t like to delegate to autonomous agents,” he says.
He also acknowledges that fully autonomous systems can introduce risks if guardrails and oversight mechanisms are not implemented properly.
Contextual relevance is redefining hyper-personalisation
Joseph avers the industry often misunderstands hyper-personalisation by focusing too heavily on static customer preferences and historical behaviour.
Most enterprises already understand broad customer preferences, preferred brands, categories, demographics, or buying patterns. However, the real challenge lies in understanding relevance in the present moment.
He points out that true hyper-personalisation is less about historical profiling and more about contextual understanding. A customer who preferred one product category previously may currently be exploring entirely different interests or life-stage requirements.
This is where AI systems are becoming increasingly valuable. Instead of relying on static segmentation logic, AI can dynamically generate context-aware content, recommendations, and engagement journeys tailored to individual customer situations. Previously, such large-scale contextual customisation was difficult because marketing teams lacked the operational bandwidth to create thousands of content variations manually.
Explainability and governance will define trustworthy AI systems
As enterprises deploy AI more deeply into customer engagement environments, Joseph believes explainability and governance will become increasingly important.
AI systems must not only make decisions but also explain why those decisions were taken. “Whatever decisions you take, there should be some sort of reasoning to that,” he says.
Joseph explains that enterprise AI systems increasingly involve multiple interconnected agents working simultaneously across segmentation, campaign creation, content generation, targeting, and orchestration workflows.
“Each of these moving parts needs to have some sort of a judge or QA agent,” he notes.
This layered oversight becomes important not only for governance but also for building enterprise trust in AI-generated outputs over time. Organisations, he believes, are far more likely to scale AI adoption when they can understand the logic behind system recommendations and decisions.
Specialised AI models will emerge across industries
Joseph also expects enterprises to increasingly adopt specialised AI models tailored to specific industry contexts and use cases.
“There are use cases for both,” he says, referring to general-purpose and specialised AI models.
Industries such as healthcare and pharmaceuticals, for example, may require domain-specific models capable of handling specialised terminology, regulatory contexts, and highly contextual datasets.
“If you are in the pharmaceutical space, maybe you need a very highly specialised model rather than a general large language model,” Joseph explains. According to him, the choice of model architecture will depend heavily on the complexity of the business problem, the nature of customer interactions, and the level of contextual understanding required.
Customer engagement platforms are evolving into decision engines
Looking ahead, Joseph believes enterprise engagement platforms themselves are undergoing a structural evolution.
Instead of functioning purely as campaign management tools, these systems are increasingly becoming intelligent decision engines capable of orchestrating end-to-end customer engagement journeys autonomously.
The broader industry direction, according to him, is towards systems that can interpret business objectives directly, generate engagement strategies automatically, and continuously optimise customer journeys in real time.