Bold Prediction! In the next three years, the "supercharged driver" of private domain growth will undoubtedly be this!
Recently, while working from home, I attended several online seminars on digital marketing and private domain operations. I had the privilege to observe some private domain marketing case studies from last year, ranging from well-known brands to small and micro brands...
After reviewing these cases, I noticed that most brands are showing a trend of homogenization in their private domain operation models. It's also worth mentioning that after several years of various analyses and sharing, the field of private domain operations has almost nothing left to dissect.
In fact, according to search results from Baidu Index, the concept of "private domain traffic" saw its search index increase nearly 20-fold within two years since it was first mentioned in 2018, reaching its peak in 2021.
However, after May 2021, the index started to show a downward trend. Based on Webpower's industry observations, we have reason to believe that the growth of private domain traffic may have reached a critical bottleneck.
Therefore, we've been contemplating:
What exactly will be the driving factor to trigger another surge in private domain operations;
When various people-centric methodologies and operational tactics are no longer unrivaled, could technological innovation become the key to influencing the development of private domain operations;
Can the rapidly advancing AI intelligence in recent years bring a higher-level breakthrough to private domain operations, which have traditionally been centered around human efforts...
Next, based on our discussions with various tech brands at Webpower, we'll share some insights on the application of AI intelligence in private domain operations.
The article is quite long, so we have organized an outline to help with summarization:
1. Keywords: Big Data, Customer Acquisition
Step 1: 500 Million User Profiles
Step 2: How to Precisely Reach Users
Step 3: Secondary Filtering and Follow-up Contact After Performance Tracking
三，Keywords: AI Outbound Robot (Intelligent Customer Service Robot), New Touchpoints, Refined Operations, Maximizing Customer Value
Two Problems Faced by Brands
How to Solve the Two Problems
三. Keywords: AI Outbound Calling, Common Phenomenon, Technical Advantages
The Most Difficult Points for AI Outbound Calling
The Advantages of Our AI Outbound Calling in Collaboration with Webpower and JD.com
Based on our interactions at Webpower, we know that many technology companies have entered the intelligent outbound calling field. This includes big players like JD Technology as well as many small companies developing their own solutions. As the saying goes, the duck on the river knows when the spring water is warm. When everyone is willing to invest money and time in a certain area, it often signifies huge market potential.
However, in terms of intelligent outbound calling functionalities themselves, after comparison, we find there aren't significant differences. They all automatically dial numbers from a user list and simulate real customer service representatives to conduct conversations.
If your main goal is to get an AI voice customer service system to save on labor costs, I would suggest choosing an outbound calling platform that has high simulation accuracy, strong learning capabilities, quick responses, can handle multi-round conversations, and offers good cost-effectiveness.
These are important criteria to consider when choosing a voice AI customer service system. I believe most companies offering AI outbound calling services provide these features. You can consult with several companies to verify this.
But this is not the main point we want to emphasize.
When evaluating the quality of a company's intelligent outbound calling product, we should not only look at the product itself but also consider how much value it can create for us.
As I mentioned above, a standalone intelligent AI outbound system can only serve as an AI customer service tool, performing simple tasks such as customer follow-ups, notifications, surveys, and promotions. From this perspective, its value is merely equivalent to that of a phone customer service representative.
Therefore, we can see that many companies bundle intelligent voice customer service and text customer service, collectively calling them intelligent customer service or robotic customer service. However, for those of us who have been in marketing for ten years, this kind of simple bundling brings value only in terms of quantity, without a substantial breakthrough—you can still only earn what a customer service representative can earn.
So, starting around two years ago, when private domain traffic and member SCRM (Social Customer Relationship Management) became popular, we began to ponder—if we use the intelligent customer service robot as a new touchpoint, integrate it into the SCRM system, and link it with big data, private domain traffic, and community operations, could it lead to a breakthrough in value?
So, we teamed up with JD Technology to develop an "unified in-house" marketing strategy that encompasses private domain operations and membership management, both of which are areas of great interest. AI intelligent outbound calling is used as a new touchpoint within this strategy, focusing on its key capabilities.
Here’s how we plan to execute it:
1. Keywords: Big Data, Customer Acquisition
JD, being a significant user traffic pool, can create immense value if we can channel its high-net-worth customers to the brand for targeted marketing incubation. The potential value generated in the future will be far beyond that of a mere customer service role.
Therefore, the first step to achieving "unified in-house" marketing was for JD to establish the "Jingxiaozhi" lead generation platform.
The lead generation activities on this platform are mainly carried out in three steps.
Step 1: 500 Million User Profiles
To put it simply, this platform can integrate the big data of JD's 500 million active users, generate tags based on their consumption behavior, create user profiles, and proactively capture users that match the brand's positioning and target demographics.
Step 2: How to Precisely Reach Users
Once the target customers are matched, the next step is figuring out how to reach them. JD's system primarily uses the Dongdong app and SMS to engage with users within the platform. However, it's not as simple as sending mass messages. If it were that easy, anyone could do it, and there would be nothing special about it.
The key lies in the intelligent level of engagement. It's not about blindly sending mass messages and calling it a day. The system can autonomously decide when to send messages, at which stage of the user lifecycle to send them, what to do if the user doesn't respond, and how to personalize messages for different users...
Let me give an example:
When the system detects that within six days a user has viewed a specific product from the brand more than twice, it automatically identifies this user as a potential customer and will trigger the sending of a Dongdong message.
Step 3: Secondary Filtering and Follow-up Contact After Performance Tracking
Generally speaking, our first round of outreach, although we strive to refine the granularity, doesn't usually achieve the best feedback results. In reality, the first round of outreach acts as a test, allowing us to identify the most promising potential customers through data monitoring.
At the same time, we shouldn't give up on customers who are not highly engaged, or on dormant and lost customers. Instead, based on the feedback from the first round, we should carry out a more targeted second round of outreach to re-engage and win them back. This approach allows us to maximize the extraction of user value.
For instance, when the system detects that some users have added products to their cart but later removed them, it will automatically trigger Dongdong messages and SMS notifications to try to win them back.
One of the home appliance brands partnering with us on Jingxiaozhi, with an average order value of over 3000 RMB, achieved a transaction volume of over 2.5 million RMB during last year's Double Eleven event. The conversion rate was 1.38%, significantly higher than the industry average.
These are the lead generation aspects of the "unified in-house" marketing strategy, which are greatly beneficial for both membership management and private domain operations. However, the importance of AI outbound calling hasn't been highlighted yet. Don't worry, it's coming right up.
二. Keywords: AI Outbound Calling, New Touchpoint, Maximizing Value
A few years ago, most B2C companies, as long as they had a platform like Jingxiaozhi and managed the initial lead generation well, could make money from either membership management or private domain operations. However, in recent years, brand owners have clearly felt some shortcomings:
The massive amounts of acquired traffic cannot be managed with precise, targeted operations.
The effectiveness of unilateral outreach, traffic diversion, and operations is gradually diminishing.
To address these two challenges faced by brands, we began utilizing AI technology to implement intelligent outbound calling and intelligent customer service robot systems. These systems are designed to act as new touchpoints, parallel to traditional channels like Dongdong, SMS, and WeChat, thereby expanding and supplementing the traditional touchpoints.
How can we make up for the shortcomings of traditional touchpoints?
1.Compared to the unidirectional outreach of WeChat, SMS, and email, AI outbound calls enable bidirectional interaction with users.
2. Utilize the interactive functionality of AI outbound calling to implement multi-scenario marketing initiatives, such as converting private domain traffic (adding followers, store visit invitations); membership management (event notifications, new customer activation, old customer re-engagement); and order scenarios (pre-sale invitations, payment reminders, after-sales follow-up, etc.).
3. Utilize the immediacy of AI outbound calling for high-efficiency user feedback and user analysis, providing a basis for the next marketing decision.
For instance, a certain beauty brand (we can't reveal the name to avoid any issues with the brand), during last year's Double 11, integrated the JD Xiaozhi traffic diversion and AI intelligent customer service robot touchpoints into their existing marketing steps by collaborating with us. In just about 7 days, they reached an additional 370,000 users, had a connection rate of over 51%, and achieved over 3.6 million in GMV. Additionally, they managed to convert 91% of the hesitant users.
So you see, marketing requires a combination of strategies to truly create value that exceeds the sum of its parts. A single small measure alone cannot achieve a quantum leap in value.
三. Keywords: AI Outbound Calling, Common Phenomenon, Technical Advantages
Of course, many companies are now jumping on the AI outbound calling bandwagon, but JD Technology, as a major player, has clear advantages in terms of technology.
Take the AI outbound calling module as an example. As I mentioned earlier, the functionalities offered by various companies are quite similar, and I also talked about the key factors to consider when choosing an AI outbound system: high simulation accuracy, strong learning capabilities, quick response time, ability to handle multi-round conversations, and cost-effectiveness. However, we believe that the two most challenging aspects to achieve are:
1，AI outbound calling has strong learning capability.
2.It can handle multi-turn conversations.
I think many of you have received AI voice calls. In terms of realism, many companies can achieve a tone and intonation that is indistinguishable from a real person. However, if you engage in a detailed conversation and ask questions, the system often reveals its limitations. Many systems can't sustain a conversation for more than a few turns before they start providing irrelevant answers or responding very slowly.
The fundamental reason lies in the limited technical capabilities of these companies. JD Technology's intelligent outbound calling, in our view, has the following advantages over others in the market:
1. Strong Learning Ability: With an extensive vocabulary database and JD's unique R&D capabilities, their AI outbound system can quickly learn new scripts. It has an exceptionally fast response time, being able to recognize up to 8 seconds of customer speech and respond within just one second.
2. Easily Handles Over 10 Rounds of Conversation with Customers: Currently, this is still challenging for most systems on the market, which typically struggle to continue beyond 5 rounds of dialogue.
3. High Realism: Besides mimicking human-like intonation and voice quality, it can even simulate the breathing and pauses of a live customer service agent, ensuring a truly seamless experience.
4.Automatically collects user feedback, and analysis reports are generated automatically.
5、Millions of outbound calls are conducted automatically without the need for human intervention, making it truly intelligent.
Based on the above analysis and industry observations, we at Webpower boldly predict that in the next three years, AI intelligent outbound calling systems will be a powerful driver for the resurgence of private domain operations. While many brands have yet to catch on, especially those in retail, you now know what steps to take, right?
We are Webpower, a digital marketing company with an attitude.
