Solving your Data & platform to fully empower your customer engagement
Most companies understand that their need to master their engagement with their customers, but fall short how to do so. This article explains the key conceptions, solution architecture and key components. It also guides on the data & technical challenges and helps to identify/ evaluate a most suitable platform.
Overall solution architecture – Key components to be successful
The Customer Data Tech Stack (CDTS) incorporates six key categories:
1) Capturing Customer Interactions: Customer Analytics / Tag Management
A tag management system implemented on the web (Javascript) and mobile apps (SDK) captures data points on customer behavior (page views, clicks, session duration, articles read, videos viewed, etc.). The analysis of this data provides insight into customer journeys. It reveals consumption patterns, which strategically informs the product and content teams.
2) Storing and Querying Data: Data Warehouse
Collect and unify data across multiple sources to build a 360-degree view of customer. All demographic information or records of interaction are integrated from various sources into a single place.
3) Pipelining Data To Warehouse: ELT / ETL
Gather, transform and load data into the data warehouse.
4) Connecting and Orchestrating Data: Customer Data Platform (CDP)
The CDP serves as omnichannel customer data orchestrators. It resolves identities across multiple data sources. It breaks down the customer base into target segments and sends segments to the MarTech stack with prescriptive instructions. The MarTech stack includes various activation and personalization tools (email, targeted advertisements, push notifications, etc.). The CDP helps develop and execute a marketing strategy that personalizes communication across multiple touch points.
5) Exploring and Visualizing Data: Business Intelligence (BI)
BI tools track KPIs in real-time, evaluate performance/ successes of campaigns & initiatives across the organization, and generate custom dashboards and standardized reports.
6) Supplementing Existing Data: Data Enrichment
Data enrichment adds supplementary data sets from second and third parties to gain deeper customer insights. Example: May acquire demographic data about an individual and add to existing customer dataset. Behavioral features are combined with demographic data (income, marital status, home ownership, category purchase intent, etc.).
Putting it to use: Develop strategies around marketing, advertising, product development, and content curation
Personalized marketing
Construct customer profiles for each individual. Segment them into target cohorts that share demographic, psychographic, geographic, and behavioral attributes.
Use microtargeting to send the right message, to the right person at the right time. Use email promotion, push notification on the mobile app, recommendation on the website, etc.
Leverage personalization. According to Havard Business Review, personalization can reduce acquisition spend by up to 50 percent, lift revenue by 5 to 15 percent, and increase efficiency of marketing costs by 10 to 30 percent.
Targeted advertising
Leverage the data management platform (DMP) with lookalike modeling to discover and identify prospective customers who have similar attributes to the customers in the target segments. Then connect the DMP to the demand side platform (DSP), which is an advertising system that allows to manage the purchasing of add slots. Serve targeted ads to specific individuals for promotional and brand awareness. Segments could include cart abandoners, newly signed-up member, high lifetime value customers, etc.
Use the supply side platform (SSP) to leverage your web and mobile apps. Sell ad inventory to other brands and generate revenue. Expand the effort to build a large audience, brand and platform where other organizations want to become part of.
Product development and Content curation
Analyze the customer journey behavior, listen to customer feedback, build right products & features. Continuously innovate and enhance customer experience.
Use content analytics to identify, gather and create the right content to tell brand stories and information.
The challenges with the technology & data solutions at the example of the DCP
The previous sections offered a simplified blue print of a solution architecture and its key components. This section addresses some of the technical and related challenges.
In an ideal world the mentioned platforms and tools mentioned above would be plug & play ready and/ or would have all the capabilities and features necessary to meet the many needs to serve customers.
Unfortunately, this is not the case…
The Customer Data platform (CDP), a key component in the blue print above, is not as ready as most of their respective vendors like to suggest. No vendor product can provide all necessary functionality. Companies will have to add notable systems, tools, applications and technologies to meet their specific needs.
The difficulty starts for many companies already when selecting a CDP. There are many vendors sell "CDP" offerings. Each setting different emphasize and displaying strengths & shortcomings.
One reason is that the CDP is a fairly new concept (although some vendors will tell you vehemently otherwise). Arguably, there is currently not native purpose-built CDP. The majority of CDP offerings are repurposed products that are rebranding and repositioning themselves. Many are built on ten-plus-year-old architectures that served different purposes.
Most current CDP offering come from different starting points and development journeys. There are 6 types that may be be classified into Technical Professional CDPs and Business Professional CDPs.
Technical Professional CDPs
Each of these CDPs have their origin in other systems that in their areas have been providing valuable services that fulfill part of the many needs to serve the customer.
Critiques argue that technical professional CDPs may underdeliver at times on authentic experiences because today’s business professionals require agility, accuracy, and automation in their personalization operations. Technical professional CDP may impede business users from realizing the key outcomes as they hope for near-zero reliance on IT (i.e. agility), correctness of predictive recommendations (i.e. accuracy) and auto-scaling campaign execution (i.e. automation)
Business Professional CDPs
Their focus on enabling data democracy via friendly interfaces helps distinguish them from technical professional CDPs. Their greatest impact is scaling personalized experiences by putting data in the hands of the non-technical masses within an enterprise—such as marketers, CX professionals, service analysts, support analysts, Site personalization professionals, finance analysts, etc.
Other emerging types of CDPs
Another type of CDP is emerging, the Smart Hub CDP. This type is built from the ground up to better fit today's need to serve customers in a better way.
The advantage is that these types of CDP vendors may design their architectures from the ground up according to the latest customer needs/ insight and latest technologies. They do not need to retrofit previous legacy systems and merge architectures of various systems.
Overall perspective on CDPs
Each of these offerings have strong value propositions for certain use cases. None of them currently covers all necessary CDP functionality. All arguably fall yet short in delivering personalization at scale and enabling the agility, accuracy and automation needed.
Vendors are proceeding fast to fill gaps and drive improvements.
Each company needs to diligently evaluate their needs and then prioritize key areas that are most important to them.
This section offers three key criteria and its related capabilities
A) Unify. Unify and match all event-level interactions across online and offline channels, creating a single customer view: 1) Data Collection: Ingest, cleanse, and store 100% of customer data breadth from all online/ offline sources (via batch, API, streaming). Provide full depth of event-level historical data in preferred cloud environment. 2) Identify matching: Resolve, authenticate, and update customer records through configurable, deterministic, and probabilistic matching techniques.
B) Analyze. Democratize insights, increase organizational agility, and augment human intelligence with AI to support 1:1 personalization. 1) Self-Service Usability: Enable business users to self-serve insights, audiences, and targeting with defined permissions and governance controls. 2) Perform advanced analytics and operationalize ML-based models across the entire history of customer behaviors.
C) Activate. Orchestrate, test, and measure personalized experiences across all customer touchpoints. 1) Orchestration: Orchestrate ad-hoc, triggered, or journey-based experiences from a single UI to all marketing, service, and sales touchpoints, in real-time. 2) Optimization: Configure tests across any channel and measure performance using configurable business metrics or machine-learning algorithms that automate experience optimization.
See how it applies to omni-channel customer engagement
In Summary
The needs to serve customer are many and requires holistic solutions to address the main pain points and opportunities. Companies need to constantly monitor/learn/ improve as the engagement with their customers, products & services, business models and technology offerings evolve.
Special credit
This article leverages much of the content & insights of IX.CO, a leading digital media company that serves many top brands around the world. It also leverages much content & insights of ActionIQ, a company that solves enterprise data challenges for G2000 companies.
Legal disclaimer:
This article (like any other of my articles) reflects only my personal views and does not claim to represent my current employer or any other client I have engaged with throughout the years. The article is meant to further discussion only and does not present an offer or advice from a legal perspective.