What do experts say about CDPs in the Vietnamese market?

Updated on October 19, 2026: figures reviewed and updated.

What do experts say about CDPs in the Vietnamese market?

You have probably heard of the customer data platform (CDP), and many companies already use one. In Vietnam, though, the CDP still looks like a new trend. This article collects what local experts have to say about CDP in Vietnam, along with marketing automation, AI, data and related topics.

Background

In recent years, “personalization” has become a key goal for marketing and sales teams at many companies. To personalize anything and improve the customer experience, though, a business needs data it can analyze to find user insights.

For that reason, many Vietnamese companies have turned to customer data platforms (CDPs) to collect and consolidate data from their online and offline sources. Pairing a CDP with marketing technology promises to show a business how to win customers and store what it knows about them, so that it can deliver the best possible customer journey.

What is a customer data platform (CDP)?

A customer data platform (CDP) is software that gathers and organizes customer data from many different touchpoints, so that it can be used by other software, systems and marketing efforts. A CDP collects real-time data and structures it into centralized profiles for individual customers.

The CDP Institute defines a CDP as “packaged software that creates a persistent, unified customer database that is accessible to other systems”.

A customer data platform builds customer profiles by integrating data from many first-party, second-party and third-party sources. That data includes CRM and DMP data, transaction systems, web forms, email, social media activity, websites and e-commerce behavioral data, among others.

What is customer data?

CDPs exist because customer data has become important to both business operations and marketing activities. So what exactly counts as customer data?

Customer data is the information people leave behind when they use the internet and when they interact with companies, whether online or offline. In other words, it comes to a business through its websites, blogs, e-commerce portals and in-store interactions. Customer data is very valuable to businesses, even though recent regulations (such as GDPR) have changed the way organizations collect and manage it.

A CDP collects and organizes 4 main types of customer data

Identity data

Identity data forms the foundation of each customer profile in a CDP. This type of data lets a business identify every customer uniquely and avoid costly duplication of customer records. Identity data includes:

• Name information, such as first and last name
• Demographic information, such as age and gender
• Location information, such as address, city and zip code
• Contact information, such as phone number and email address
• Social information, such as Twitter and LinkedIn
• Professional information, such as job title and company
• Account information, such as specific user IDs and account numbers

Descriptive data

Descriptive data builds on identity data and gives you a fuller picture of your customers. What counts as descriptive data varies from one type of company to another.

For example, a car dealer might collect details about a customer’s lifestyle, while a diaper company would collect details about the number of children in the customer’s household. Descriptive data includes:

• Career information, such as previous employers, industry, income and job level
• Lifestyle information, such as the type of home the customer lives in, vehicles and pets
• Family information, such as number of children and marital status
• Hobby information, such as magazine subscriptions and gym memberships

Quantitative or behavioral data

Quantitative data lets a business understand how each individual customer has interacted with the organization, whether through specific actions, responses or transactions. Quantitative data includes:

• Transaction information, such as the number and type of products bought or returned, the number of abandoned carts and order dates. It also includes RFM analysis: recency (how recently did this customer buy?), frequency (how often does this customer buy?) and monetary value (how much does this customer spend per purchase?)
• Email communication information, such as emails opened, emails clicked and email replies
• Online activity information, such as website visits, CTR, product views and engagement on social media
• Customer service information, such as contact dates, details of each inquiry and details of the service representative

Qualitative data

Qualitative data gives context to a customer profile and shows the personality behind the data. This type of data captures any motivation, opinion or reaction a customer expresses, whether or not it relates to the company. Qualitative data includes:

• Customer motivation, such as: How did you hear about us? Why did you buy this product? What made you choose this product over others?
• Opinions, such as: How would you rate this product? How would you rate our customer service? How likely are you to recommend us?
• Attitudes, such as favorite color, animal or food

As you can see, a CDP collects and organizes many different kinds of data. It is worth keeping in mind that the specific data points and data categories will vary depending on the business and the industry it operates in.

The upgraded CDP: marketers expect CDPs to go beyond data aggregation and power omnichannel activation

According to research, only 10% of CDP users feel that first-generation products meet their current needs, and 99% do not believe their current CDP will meet their future requirements.

The study was commissioned from Forrester Consulting by Zeta Global (NYSE: ZETA), a cloud-based marketing technology company that helps businesses acquire, grow and retain customers. It found that most CDP users believe today’s CDP solutions fall short of expectations and of what they were supposed to be capable of. These users want a stronger CDP, one that goes further in aggregating and managing data and also carries out key marketing functions such as personalization and cross-channel activation.

According to the study, 66% of respondents believe adopting a CDP is a strategy or initiative that matters to their organization, yet only 10% of today’s CDP users feel their product meets all their needs. That figure drops sharply to 1% among those who believe their current CDP will meet their future requirements. Data management matters, but marketers also say they need to analyze data, gather insights, personalize messages, adjust strategies and campaigns, measure performance and activate data across every marketing channel. More than half of respondents have run into technical challenges with their CDP and need more expertise from the vendor on integration, deployment and execution.

“While first-generation CDPs gave marketers the ability to manage, enrich, store and access first-party customer data, the growing complexity of today’s marketing environment has exposed a widening gap between what users need and what CDPs currently do,” said Christian Monberg, Chief Technology Officer and Chief Product Officer at Zeta Global. “We believe today’s marketers need an upgraded CDP that can go beyond data management and activate across all channels, generate insights and measure results, comparable to integrating a modern marketing cloud.”

The study confirms that to meet top strategic goals, a CDP needs integrated capabilities that join customer data management with marketing execution, such as personalization, a unified dashboard and real-time data processing. It also stresses the role CDPs play in top-level outcomes: higher revenue, better acquisition and retention rates, higher customer satisfaction, more personalized interactions and greater customer trust.

Significant overall growth

According to the CDP Institute’s January 2022 industry update (covering the second half of 2021), growth came largely from established vendors, which accounted for 85% of new employees and 88% of new funding. That reflects the effort these vendors are making to become leaders in the CDP category.

Over the past 5 years, larger companies have grown faster than smaller ones, but there is still no single vendor that dominates the market.

Only 4 new companies were added in the reporting period, a low number considering the average of 15 in previous periods. In addition, 3 small companies were acquired and one went out of business.

By the end of 2021, the CDP industry had added an estimated 1,375 employees compared with the previous reporting period (CDP.com, January 2022). Specifically, the top 25% of companies by headcount grew their number of jobs by 14%. The rest of the industry grew by only 4%.

Looking at growth by when vendors entered the market, the traditional core vendors (founded between 2008 and 2012) account for 28% of the industry. Vendors that arrived after 2014 make up a quarter of the market but are smaller and have less funding.

Another interesting point: US-based CDPs make up 45% of the market (64 vendors) and received 75% of the funding. EMEA comes next (with the largest CDPs founded in France, the UK and the Netherlands), followed by CDP vendors from APAC. APAC vendors are younger companies that usually offer delivery CDPs.

Big funding and consolidation

Funding in this period reached $853 million, more than in any previous period. Most of that money went to established vendors. 5 deals in the period involved companies with more than 200 employees, including 4 vendors that each received more than $100 million. Another 6 smaller deals added up to $117 million, which shows there is still support for smaller vendors.

With this new funding, the top 5 companies have now raised an average of $190 million each. The rest of the market has raised an average of $21 million.

The CDP Institute splits customer data platforms into 4 main types: campaign, analytics, delivery and data. As of January 2022, the most common type was the Campaign CDP, at 39% of all CDPs. Campaign CDPs account for 41% of employees and 44% of funding. The oldest vendors in the industry are mostly Campaign CDPs and Delivery CDPs.

Campaign CDPs still hold the largest share of the market, but that share is gradually shrinking, and the change is working in favor of Delivery CDPs.

Nearly all Analytics CDPs and Data CDPs appeared in 2012. Data CDPs are also growing, but more slowly. These CDPs also tend to be smaller and less well funded (except for those in the US). Analytics CDPs were usually the largest and best-funded CDPs, but they are now below average in headcount and funding.

Strong growth for a key technology

These are some of the highlights of the CDP Institute’s January 2022 industry update. Established vendors are clearly still doing well and growing stronger, but there are a few promising newcomers worth watching because they have raised significant funding. Delivery CDPs are growing strongly while Campaign CDPs are growing more slowly, which points to a shift in the types of CDP on offer, even though nobody expects that shift to happen quickly.

What do experts say about CDP in Vietnam?

Where the CDP came from and why it grew

According to Triều Nguyễn (founder of PrimeData), the CDP came about and grew for these reasons:

• Successful marketing requires several stages to work together smoothly and without gaps: data segmentation, channels, content and creative, automation and so on.
• Each of those stages has matured on its own, for example using Mailchimp for email marketing.
• A CDP brings all of those solutions together in a single piece of software, so they are convenient to use and the data does not get fragmented or inconsistent.
• Budgets are limited, and using too many automation tools drives costs up. A tool that integrates everything, or is built to cut the investment needed, makes the decision easier for customers.

CDP features look different in theory and in practice

• Unify

The key things to consider are customer identity data and access management (the user’s history). Data that cannot be tied to an identity is meaningless. How you identify people, however, depends on each business case. Ask yourself these two questions:

Anonymous identification: what do you do if the user has not logged in?

Identity resolution and cross-checking identity: a single customer view comes from combining channels and services. So if a company has 3 services, how does it confirm that it is dealing with the same customer?

• Audience segmentation

There are 2 main approaches to segmentation:

The first is exporting data from legacy systems and uploading it to the CDP. The drawback is that this only works for existing customers (transaction data).

The second is dynamic segments, which can prompt anonymous users (visitors who have not identified themselves) to take actions in real time.

In practice, these 2 types of segmentation do not yet fit together well. The data in the first group seems very small compared with the second, so it is hard for a CDP to solve this problem properly. Triều also pointed out a common mistake: many people think a CDP can solve every problem. Features such as cross tracking and recommendations existed long before CDPs appeared.

Nam Nguyễn (founder of Opla), who has many years of experience advising on CDPs as a consultant at Oracle and Salesforce, said that each business defines a CDP differently depending on its sales strategy. As CDPs evolve, that definition changes too. What matters, then, is to focus on the customer’s actual problem and adapt the CDP to fit it.

Take an e-commerce marketplace that needs to optimize its conversion rate (CR). There, a CDP may be software that collects data, analyzes behavior and makes recommendations. For an O2O (offline to online) retailer that needs to improve the shopping experience, a CDP is software that collects data, measures and analyzes user behavior, and identifies touchpoints so it can personalize sales advice and customer care.

The typical digital marketing and e-commerce model in Vietnam is still simple: Google and Facebook ads serve as the top of the funnel, and below that comes direct communication (sales and customer care). So simply running web push notifications is enough to lift CR (conversion rate), and there is no need for a tool as complex as a CDP.

Minh Nguyễn (founder of AkaLava), who has many years of experience advising on CDPs, said CDP vendors always open with the same story: customers who have a bad experience will leave. The CDP then arrives to improve the customer experience. From management’s point of view, that is the clearest benefit of a CDP. If, on the other hand, you approach a CDP as a way to cut marketing costs or make advertising more effective, that benefit is very hard to prove.

From the brand side, Thuận Đoàn, Deputy General Director of Kowil Group, said brands worry about falling behind new technology trends such as AI, data, machine learning and martech. Boards have high expectations of these buzzwords and push marketers to find ways to apply them.

The hard part on the brand side is mindset, because people in management roles are used to traditional ways of working. The people working on CDPs, meanwhile, are mostly young and tend to use newer metrics such as CAC, LTV, CR and funnels. The result is that the two sides struggle to find common ground.

Another challenge is that large companies have fairly complex operating processes, which makes it hard to use the system or collect data. Most companies do not want to waste data and would like to own a data lake that holds 3 kinds of data: data they already have, data they are collecting now, and data they could collect in the future. Only once the data lake is big enough do they start analyzing it for customer insights and running automation. Building a data lake, however, is far from easy.

So how you define a CDP does not really matter. What matters is that the business understands which needs the CDP will meet and how it will be used. Each industry and each field has its own use cases, and each of those use cases leads to a new definition of what a CDP is.

• Digital transformation in Vietnam and the growth of CDPs

Trần Ngọc Hải, who once ran a CDP project for 1MG and now holds a data role at Techcombank, has experience rolling out CDPs at the level of a large conglomerate. To be able to adopt a CDP, assess the platform’s capabilities accurately and apply it for partners, he drew up 3 CDP architecture views:

Conceptual view: lists the main capability blocks and what each one does.

Data flow and logical view: explains how each block works, how each capability is handled logically, how data flows in and out, and how the blocks work together as a whole.

Tech stack: working with the tech team and data engineers to decide which technology to use in development.

He once spent 3 months building a CDP model, only to have it turned down because the business could not picture how it would be used in a real case. The hardest part of a CDP is managing expectations. Many people expect far too much from CDP features, when in reality the data they have may not even need AI.

Hải also shared some of the misconceptions many people commonly hold about CDPs:

A CDP does not suit every business. For example, every business likes the idea of a customer 360 view. But this metric does not have much data behind it, or many touchpoints to interact through. In the end it delivers only 1 view of the customer, so the customer 360 view turns out to be an unrealistic metric.

Painting a picture that is too broad or too deep before working on a real problem is a mistake. Many CDP vendors rushed to show off AI and machine learning features before they had real data, and then failed once the actual work began. Take VinID: many transactions in the Vingroup ecosystem are still low frequency (users visit rarely because they do not spend much time in the app), so the amount of data collected is small and the behavior is not varied. As a result, there is not enough input data to analyze for insights.

2 ways to put a CDP to work

Native CDP: built for the tech team, which can write code to connect data flows. For the data the CDP collects, it offers query capabilities to filter that data. Today there are still CDPs with real-time segmentation and CDPs without it.

Marketing CDP: already integrated into a martech platform built for marketers.

Đỗ Hữu Hưng, CEO of Accesstrade, a top affiliate platform in Vietnam, laid out a practical problem: he is ready to pay well for experts who join Accesstrade and help it gain 50,000 new publishers, with new data, in real time. New publishers bring in a lot of revenue, but their churn rate is very high.

The problem is how to get publishers to come back and earn more. According to Hưng, whether you use a CDP or a CXP, the outcome is the same as long as it solves that problem. Any advice or strategy can work if it has a good case study behind it. This is a typical example of the kind of problem that challenges people working on digital transformation in Vietnam. Ask questions such as: Does the client want to invest in technology? Does the client want a CDP? Does the client want to go through digital transformation? Does the client have the budget?

• A big opportunity and an open market, but can businesses actually take on a CDP?

Hưng Lê, CEO of Mobio.vn, has many years of experience working with CDPs. In his view, the market needs pioneers to explore what CDPs can do. Vendors need to choose a direction based on what customers need. Some companies, for example, have moved from email to CDP, which shows that the tools people use shift with market needs and brand needs.

Jackie Long (Thành Long), CEO of Hub.js, said Hưng’s view of CDPs is quite realistic. In his own case, though, he seems to need a DMP rather than a CDP, to lift CR for advertisers.

He felt that Triều’s and Minh’s views on CDPs already cover both the marketer’s angle and the strategic angle. On how DMPs and CDPs are used today, Long shared these use cases:

A use case in retail O2O (using DMP, CRM, OTT)

A use case in gaming, a Game CDP (using a data lake, ClickHouse, analytics, omnichannel campaigns)

A use case in banking, omnichannel (using analytics, CRM, agent eKYC, automation)

A use case in HRM scoring (using talent, profile, CRM, onboarding)

According to Long, CDPs already exist in their most basic form in companies’ marketing and sales work. He has run many experiments using DMPs and campaign automation to collect and mine potential data based on existing scenarios.

To solve the bigger problem that a CDP is meant to address, however, three parties have to work together:

The business: needs to state clearly what the customer is aiming for and understand the real value of each platform (not only the CDP, but also CRM, ERP and martech).

The consultant: needs to listen to the client and choose the most suitable solution, one that can be deployed with the business’s limited resources.

The solution vendor: needs to know which platforms its capabilities allow it to provide, and which platforms it can draw on to meet the customer’s needs.

Managing data effectively with Adtek

A CDP is seen as a solution that helps businesses collect, manage and make use of their customer data effectively. How to actually put a CDP to use, though, is not a simple question.

With that problem in mind, and with the goal of helping businesses use their data to grow, Adtek offers consulting on and implementation of CDP systems for data management. A CDP makes it easier for a business to collect, manage and analyze its data. Adtek works with businesses to mine data from the CDP, analyze it, and plan a strategy for building customer experiences that raise customer lifetime value.