Nội dung
AI is moving very quickly into marketing and customer care, but the way businesses are implementing it is far from uniform. For Vietnamese marketers, what matters is not only the rising level of adoption, but also the challenge of choosing the right architecture, governing data, and organizing teams so AI truly creates value instead of making systems more complicated.
- Key points:
- 90% of CX organizations have tested or deployed AI, showing that the technology has shifted from “if” to “when and how.”
- Businesses are still split evenly among end-to-end, hybrid, and best-of-breed models, with no standard AI architecture emerging yet.
- The martech stack risks overlapping functions, causing data discrepancies and raising questions about ROI if operational discipline is lacking.
- AI is also changing the path for training young marketers: automating work does not mean automatically creating a new generation of marketing leaders.
AI in CX is mainstream, but every company is taking a different path
According to Five9’s 2026 Business Leaders CX Report, as cited by MarTech, 90% of customer experience (CX) organizations are now in the testing or deployment stage for AI. This is a very strong signal: AI is no longer a “future project” but infrastructure that is already moving into operation.
However, the consensus stops at the question of whether AI should be used. On the question of how to use it, the market is still fairly evenly split between end-to-end, hybrid, and best-of-breed approaches. In other words, businesses are heading toward the same destination—applying AI in CX—but choosing completely different routes.
MarTech shows that the same thing is happening in infrastructure: 84% of organizations are in the process of moving from on-premises systems to the cloud, but most are still operating in a hybrid model. This clearly reflects a cautious mindset: businesses do not want to “burn the bridge” while AI models, vendors, and customer expectations are still changing quickly.
For marketers, the lesson here is not just “adopt AI.” More importantly, they need to define which customer journey goal AI will serve: reducing response times, improving personalization, or optimizing productivity for customer service teams. Otherwise, AI may add another layer of technology without necessarily improving the actual experience.
Source: MarTech, “AI adoption hits 90% as CX deployment paths diverge”.
What is the martech stack saying when asked about ROI?
In the second analysis, MarTech raises a very thought-provoking angle: if the entire martech stack could “speak,” how would it answer the question of why costs are rising while results are not increasing accordingly?
The familiar story in many marketing departments goes like this: the CMO invests in more tools so the team can work faster, target more accurately, and personalize better. But after 18 months, campaign speed has not improved much, while reports are producing different numbers for the same metric. At that point, the issue is no longer whether the tools are good, but how much overlap the stack has accumulated.
The article points to a fairly typical example: customer data platforms (CDPs) and marketing automation platforms (MAPs) can both build audiences, use part of the same customer data, but define segments differently. When functions overlap, businesses not only spend more budget but also easily create inconsistencies in operations and reporting.
The key takeaway for marketers is that AI only makes this problem more visible; it does not solve it on its own. If data is fragmented, processes are vague, and responsibilities across platforms are unclear, AI can amplify the chaos instead of streamlining it. So before adding another layer of AI, businesses need to review their existing martech architecture: which tools create real value, and which are simply duplicating the functions of others.
Source: MarTech, “If your martech stack could talk, what would it say?”
AI can replace tasks, but who will train the next generation of marketers?
The third piece in this week’s picture is not about technology, but about people. MarTech asks: if AI does most of the execution work that used to be assigned to junior marketers, how will the future generation of marketing leaders learn the trade?
In traditional marketing, experience is built through an apprenticeship model: running campaigns, making mistakes, receiving feedback, and gradually understanding what makes a good decision. This is the process of developing “judgment” — a decision-making ability that is very difficult to replace with theory alone. If AI takes over too much of the foundational work, young marketers may lose the necessary opportunities to learn through hands-on experience.
The article emphasizes that the issue should not be understood simply as whether AI will replace newcomers. Instead, the more important question is: as job tasks change, how will businesses design training paths and allocate work so junior marketers still have the chance to develop strategic thinking?
This is especially important for marketing teams that are accelerating AI adoption. If everything is automated at the execution layer, businesses may save time in the short term but create a long-term gap in the talent pipeline. Effective AI is not just AI that helps work faster, but AI that is introduced into an organizational model capable of nurturing human capability.
Source: MarTech, “Who trains tomorrow’s marketers if AI does the work?”
What this means for the Vietnamese market
For Vietnamese businesses, these three signals point to a fairly clear conclusion: AI is no longer just a standalone technology experiment, but a broader challenge involving system architecture, data, and people. Many local brands tend to buy more tools to solve individual pain points, but without data standardization and clear governance, AI can easily become a new cost layer rather than a growth engine.
From a CX perspective, businesses should start with specific use cases such as classifying customer requests, supporting automated service, or personalizing content along the purchase journey. From a martech perspective, they need to review platforms with overlapping functions to avoid “buying on top of buying.” And on the people side, marketing teams should preserve room for junior marketers to take part in planning, analysis, and campaign optimization rather than handing everything over to tools.
In short, AI in marketing in Vietnam will be more effective if it is deployed as an operational transformation program, not as a technology accessory.
See more marketing news and guides at https://marketing365.vn.
Follow more updates from Marketing365 to stay on top of the latest marketing trends.
Read more articles in the same category at Digital Trends.
Reference sources
- AI adoption hits 90% as CX deployment paths diverge
- If your martech stack could talk, what would it say?
- Who trains tomorrow’s marketers if AI does the work?
