本文源引自David Iscove 在《MarTech》2024年4月16日的刊文，采用中英双语排版，由ImmersiveTranslate提供翻译支持。
吸引当今的客户就像试图击中一个移动的目标。需求和期望每时每刻都在变化。通过可预测的渠道进行的通用舞台表演不再能吸引因选择而麻木的观众。为了在无尽的选择中赢得忠诚度，品牌必须通过当下的相关性和预期来传递价值。
CX today requires looking beyond the funnel
如今的客户体验需要超越漏斗
For years, we’ve used linear models to map the customer journey, assuming a predictable path from discovery to loyalty. However, rising consumer expectations and rapid technological changes have disrupted the orderly framework.
多年来，我们一直使用线性模型来绘制客户旅程，并假设从发现到忠诚的路径是可预测的。然而，不断上升的消费者期望和快速的技术变革扰乱了有序的框架。
Modern customers’ needs and expectations now shift situationally moment-to-moment, not in predictable linear phases conforming to neat traditional funnels. Think of the massive popularity of the modern subscription product model — we can try out the full features of a product for a term before even committing to make a purchase. Churn is easier than ever, so we must focus on loyalty and retention even ahead of conversion.
现代客户的需求和期望现在随时都会发生变化，而不是符合简洁的传统渠道的可预测的线性阶段。想想现代订阅产品模式的广泛流行——我们甚至可以在决定购买之前试用产品的全部功能一段时间。流失比以往任何时候都容易，因此即使在转化之前，我们也必须关注忠诚度和保留率。
On top of that, as customers, we willingly demand instantaneous personalization in real-time, not generic staged engagements that fail to adapt based on individual context. The explosion of the direct-to-consumer model has been fueled directly by a brand’s ability to deliver personal product offerings to its customers.
最重要的是，作为客户，我们愿意要求即时的实时个性化，而不是无法根据个人情况进行调整的通用分阶段参与。品牌向客户提供个性化产品的能力直接推动了直接面向消费者模式的爆炸式增长。
This new paradigm demands brands meet customers with relevance in the moment while continuously optimizing those engagements through contextual insights and signals. AI and automation make this achievable.
这种新的范式要求品牌在当下满足客户的相关需求，同时通过情境洞察和信号不断优化这些互动。人工智能和自动化使这一目标成为可能。
The new CX imperatives for brands (Content + Signals):
品牌的新 CX 要求（内容 + 信号）：
- Always on: Deliver situational relevance powered by understanding each customer as close to a segment-of-one as possible.
始终在线：通过尽可能接近地了解每个客户来提供情境相关性。 - Always listening: Rapidly interpret signals to predict the next best actions rather than reacting post-behavior.
始终倾听：快速解读信号以预测下一个最佳行动，而不是在行为后做出反应。
This approach to hyper-personalized content velocity and predictive intelligence presents the next era of intelligent customer experiences. And it is powered by the connectivity of data, insights and triggered actions.
这种超个性化内容速度和预测智能的方法代表了智能客户体验的下一个时代。它由数据、见解和触发操作的连接性提供支持。
Enabling always-on relevance
启用始终在线的相关性
Situational content alignment refers to the ability to instantly tailor messaging, offers and creative to each individual customer based on understanding their current real-time context.
情境内容调整是指基于了解每个客户当前的实时环境，立即为每个客户量身定制消息、优惠和创意的能力。
At a high level, this can be achieved through four sequential areas of focus:
在较高层面上，这可以通过四个连续的重点领域来实现：
- Collect. Collect and integrate diverse data types to build comprehensive customer profiles.
收集。收集并整合不同类型的数据以建立全面的客户档案。 - Map. Create dynamic personalized content models mapped to micro-segments.
地图。创建映射到微细分的动态个性化内容模型。 - Create. Generate tailored content assets to align with each visitor.
创造。生成定制的内容资产以适应每个访问者。 - Analyze. Optimize content through real-time performance analytics.
分析。通过实时性能分析优化内容。
Let’s explore these in detail, understanding their objectives, the technology platforms that empower them and how AI can be leveraged within each to enhance the solution.
让我们详细探讨这些，了解他们的目标、支持他们的技术平台以及如何在每个平台中利用人工智能来增强解决方案。
Collect 收集
Personalization requires a deep level of understanding of your customers and their behaviors. To understand our customers, focus on collecting and integrating diverse data types to build comprehensive customer profiles, encompassing their behaviors, preferences and interactions across various touchpoints.
个性化需要深入了解客户及其行为。要了解我们的客户，请重点收集和整合不同的数据类型，以建立全面的客户档案，涵盖他们在各个接触点的行为、偏好和互动。
This is where customer data platforms (CDPs) are essential in their ability to offer a centralized repository to aggregate data from a myriad of sources (e.g., CRM, web/email/mobile analytics, social media management, ecommerce, POS, IoT, customer service and chat, survey tools, etc.).
这就是客户数据平台 (CDP) 至关重要的地方，因为它们能够提供集中存储库来聚合来自无数来源（例如 CRM、网络/电子邮件/移动分析、社交媒体管理、电子商务、POS、物联网、客户服务和聊天、调查工具等）。
The CDP is the dumping ground for all customer touchpoints and arguably the most essential tool required to deliver a brand’s ability to embrace customer centricity.
CDP 是所有客户接触点的垃圾场，并且可以说是提供品牌以客户为中心的能力所需的最重要的工具。
By using AI/ML algorithms, CDPs can significantly enhance the automation of data cleansing, refine data quality and implement preventative measures to ensure clean and accurate data.
通过使用人工智能/机器学习算法，CDP 可以显着增强数据清理的自动化、提高数据质量并实施预防措施，以确保数据干净和准确。
This saves time and manual effort in aggregating data from the external capturing systems and reduces human error and speeds up the data preparation process.
这节省了从外部捕获系统聚合数据的时间和人力，减少了人为错误并加快了数据准备过程。
Dig deeper: AI-powered features to look for in customer data platforms
深入挖掘：在客户数据平台中寻找人工智能驱动的功能
Map
AI-powered CDPs process immense datasets to surface insights that can refine customer profiles into smaller and smaller segments (“micro-segments”), surpassing what could be achieved through manual interpretation.
由人工智能驱动的 CDP 处理大量数据集，以得出洞察，从而将客户档案细化为越来越小的细分（“微细分”），超越了通过手动解释所能实现的效果。
AI can detect nuances of customer behavior across profiles and then automate the segmentation process with granularity and extreme depth, predicting the most effective content for each micro-segment and moment.
人工智能可以检测客户行为的细微差别，然后以粒度和极端深度自动执行细分过程，预测每个微细分和时刻的最有效内容。
With micro-segments identified, the decision on what content to deliver to that individual customer profile is made using tools such as a dynamic content optimization (DCO) platform. We can assign specific content strategies that resonate with each micro-segment at various stages of their journey.
确定微细分后，可以使用动态内容优化 (DCO) 平台等工具来决定向个人客户资料提供哪些内容。我们可以分配特定的内容策略，与每个微细分受众在旅程的不同阶段产生共鸣。
Applying dynamic content models to micro-segments sets the rules for our personalized content strategy, enabling unprecedented levels of customization in responding to individual customers.
将动态内容模型应用于微细分市场为我们的个性化内容策略制定了规则，从而在响应个人客户方面实现了前所未有的定制水平。
Assemble 集合
DCO creates personalized content by using real-time data (visitor activity) and specific user information (profile match from the CDP). It follows content strategy rules defined in the DCO and combines them with pre-designed, flexible and modular creative templates. This approach ensures the delivery of highly relevant and engaging content to individual customers throughout different touchpoints and stages of their journey.
DCO 通过使用实时数据（访客活动）和特定用户信息（来自 CDP 的配置文件匹配）创建个性化内容。它遵循 DCO 中定义的内容策略规则，并将其与预先设计的、灵活的模块化创意模板相结合。这种方法可确保在整个旅程的不同接触点和阶段向个体客户交付高度相关且引人入胜的内容。
DCO leverages machine learning algorithms to decide which creative elements (ad copy, images, CTA) to adjust based on the firm understanding of the visitor’s identity compared against profiles in the CDP, all within milliseconds.
DCO 利用机器学习算法，根据与 CDP 中的个人资料相比对访问者身份的深入了解，决定调整哪些创意元素（广告文案、图像、CTA），所有这些都在几毫秒内完成。
This advanced communication and decisioning is only possible through connected systems driven by machine learning, which can interpret advanced consumer signals, then deliver contextually relevant, hyper-personalized content.
这种先进的沟通和决策只有通过机器学习驱动的互联系统才能实现，机器学习可以解释先进的消费者信号，然后提供上下文相关的超个性化内容。
Assess 评估
Measuring in-market performance is essential for improving how content is delivered. It helps enhance decision-making, monitor results and make continuous adjustments to optimize engagement.
衡量市场表现对于改善内容交付方式至关重要。它有助于增强决策、监控结果并不断调整以优化参与度。
This is where machine learning shines in its ability to self-assess and improve based on a feedback loop. Real-time interaction management (RTIM) systems and integrated analytics tools equipped with AI algorithms assess how customers engage with content.
这就是机器学习的亮点，因为它能够根据反馈循环进行自我评估和改进。配备人工智能算法的实时交互管理 (RTIM) 系统和集成分析工具可评估客户如何与内容互动。
This real-time analysis enables immediate adjustments to content strategies based on what is most engaging to customers, ensuring that marketing efforts are always aligned with customer expectations and preferences.
这种实时分析可以根据对客户最有吸引力的内容立即调整内容策略，确保营销工作始终符合客户的期望和偏好。
The journey toward always-on enablement is comprehensive, requiring a thoughtful integration of technology and data across customer identification, content creation, delivery and analysis. Still, if orchestrated strategically, you can unlock the potential to deliver truly personalized, dynamic customer experiences at scale.
实现永远在线的过程是全面的，需要在客户识别、内容创建、交付和分析方面对技术和数据进行深思熟虑的集成。不过，如果进行战略性策划，您可以释放大规模提供真正个性化、动态客户体验的潜力。
Listening with predictive intelligence
用预测智能聆听
Historically, brands reacted to customer behaviors only in hindsight once an outcome had already occurred, missing opportunities in the moment. AI changes this by quickly interpreting signals to understand emerging needs immediately.
从历史上看，品牌只会在结果已经发生后才对客户行为做出反应，从而错失了当时的机会。人工智能通过快速解释信号以立即了解新出现的需求来改变这一点。
For example, an ecommerce company can now analyze every step millions of shoppers take across multiple channels — from browsing categories to social shares and then predict the likelihood and timing of a single individual customer’s propensity to make a purchase.
例如，一家电子商务公司现在可以分析数百万购物者在多个渠道中采取的每一步——从浏览类别到社交分享，然后预测单个客户购买倾向的可能性和时间。
Or a streaming service can determine subscribers at risk of canceling even before they threaten to quit. This is achieved through four primary areas of focus:
或者，流媒体服务甚至可以在订户威胁退出之前就确定他们是否有取消的风险。这是通过四个主要重点领域实现的：
- Collect. Aggregate and unify data from diverse sources to create a comprehensive view of customer interactions and behaviors.
收集、聚合和统一来自不同来源的数据，以创建客户交互和行为的全面视图。 - Analyze. Identify patterns and insights within the collected data, understanding customer preferences and behaviors.
分析。识别收集的数据中的模式和见解，了解客户的偏好和行为。 - Predict. Forecast future customer actions, determining what they might need or do next based on historical data.
预测。预测未来的客户行为，根据历史数据确定他们下一步可能需要什么或做什么。 - Trigger next-best-action. Act on predictive insights to engage customers with the right message at the right time.
触发下一个最佳操作。根据预测性见解采取行动，在正确的时间向客户提供正确的信息。
Collect 收集
As discussed earlier, CDPs are the central repository for collecting data from various sources. They provide the foundational data necessary for AI algorithms to analyze. Without this comprehensive data collection, AI wouldn’t have the information needed to identify patterns or make accurate predictions.
如前所述，CDP 是从各种来源收集数据的中央存储库。它们提供人工智能算法分析所需的基础数据。如果没有这种全面的数据收集，人工智能就无法获得识别模式或做出准确预测所需的信息。
Analyze 分析
After collecting data, AI and machine learning algorithms analyze it swiftly to identify subtle patterns in customer actions. This process is often much faster than humans alone could detect trends across millions of data points, enabling the identification of patterns that might not be immediately obvious.
收集数据后，人工智能和机器学习算法会快速分析数据，以识别客户行为中的微妙模式。这个过程通常比人类单独检测数百万个数据点的趋势要快得多，从而能够识别可能并不立即明显的模式。
This step is where the “connecting the dots” happens. AI algorithms sift through the collected data to understand customer behaviors and preferences, which is critical for making accurate predictions about future actions.
这一步是“连接点”发生的地方。人工智能算法筛选收集到的数据以了解客户的行为和偏好，这对于准确预测未来的行为至关重要。
Predict 预测
Predictive analytics tools use the insights generated by AI algorithms to score the probability of future customer actions. These tools can predict which products a customer will likely buy next, when they might make a purchase, or if they’re at risk of churning. This predictive capability is key to moving from a reactive to a proactive approach to customer engagement.
预测分析工具使用人工智能算法生成的见解来对未来客户行为的概率进行评分。这些工具可以预测客户接下来可能会购买哪些产品、何时可能购买，或者是否有流失的风险。这种预测能力是将被动的客户参与方式转变为主动的客户参与方式的关键。
Trigger (next-best-action)
触发（下一个最佳操作）
Marketing automation and journey optimizer platforms use predictions to automate personalized marketing actions. Based on the results of the predictive analytics, these platforms can trigger targeted emails, personalized product recommendations or customized offers to customers at the optimal time.
营销自动化和旅程优化平台使用预测来自动化个性化营销行动。根据预测分析的结果，这些平台可以在最佳时间向客户触发有针对性的电子邮件、个性化产品推荐或定制优惠。
This is the execution phase, where insights and predictions are translated into real actions. Marketing automation platforms ensure that the recommendations are delivered to the customer through the right channel and at the right moment, completing the cycle of predictive engagement.
这是执行阶段，将见解和预测转化为实际行动。营销自动化平台确保在正确的时间通过正确的渠道向客户提供建议，从而完成预测参与的周期。
Delivering always-on, always-listening customer experiences
提供始终在线、始终倾听的客户体验
Over time, AI continually gets smarter by learning from outcomes to refine predictive models and signals. This means you can pivot experiencing design from reactive to proactive using AI to understand emerging desires during micro-moments of engagement.
随着时间的推移，人工智能通过从结果中学习来完善预测模型和信号，从而不断变得更加聪明。这意味着您可以使用人工智能将体验设计从被动式转变为主动式，以了解参与的微时刻期间出现的需求。
An always-listening strategy represents a holistic approach to customer engagement, predicated on listening, predicting, engaging and optimizing with unparalleled precision.
始终倾听的策略代表了一种全面的客户参与方法，其基础是以无与伦比的精度倾听、预测、参与和优化。
