Businesses are relying on Big Data to gain a competitive advantage. The concept of Big Data is broad one and I consider it an amalgamation of different areas that help us try to get a handle on, insight from and use out of data. Pat Gelsinger, President and COO of EMC, in an article by the The Wall Street Journal said that Big Data refers to the idea that companies can extract value from collecting, processing and analyzing vast quantities of data. Businesses that can get a better handle on these data will be more likely to outperform their competitors who do not.
When describing Big Data, people typically refer to three characteristics of the data: 1) Volume: the amount of data being collected is massive; consider that 90 percent of all data in the world today has been generated in the last two years; 2) Velocity: the speed at which data are being generated/collected is very fast and needs to be analyzed as it comes in to identify fraud; and Variety: the different types of data like structured and unstructured data (e.g., images, videos and text from call center conversations). Recently introduced by IBM, Veracity describes an important fourth characteristic of data. To be of use to business, data must reflect reality (e.g., accurate, valid) and be trusted by the users.
Three Big Data Approaches
Brian Gentile, CEO of Jaspersoft, argues for a solution-oriented approach to understanding the value of Big Data. Before selecting a vendor, you first need to understand the problem you are trying solve. Keep in mind that Big Data is not just about analyzing data quickly; it is about analyzing data intelligently, perhaps with some theory-driven analyses. Gentile's three Big Data approaches include:
- Interactive Exploration - good for discovering real-time patterns from your data as they emerge
- Direct Batch Reporting - good for summarizing data into pre-built, scheduled (e.g., daily, weekly) reports
- Batch ETL (extract-transform-load) - good for analyzing historical trends or linking disparate data sources based upon pre-defined questions. Sometimes called data federation, this approach involves pulling metrics from different data sources for purposes of understanding how all the metrics are related (in a correlation sense) to each other.
So, be sure to understand that Big Data is not just about quick analysis of your data. It is also about integration of different sources of data.
Value from Analytics
In a late 2010 study, researchers from MIT Sloan Management Review and IBM asked 3000 executives, managers and analysts about how they obtain value from their massive amounts of data. They found that organizations that used business information and analytics outperformed organizations who did not. Specifically, these researchers found that top-performing businesses were twice as likely to use analytics to guide future strategies and guide day-to-day operations compared to their low-performing counterparts.
The MIT/IBM researchers, however, also found that the number one obstacle to the adoption of analytics in their organizations was a lack of understanding of how to use analytics to improve the business. There are simply not enough people with Big Data analysis skills. In fact, McKinsey and Company estimated that the US faces of huge shortage of people who have the skills to understand and make decisions based on the analysis of big data.
The MIT/IBM researchers also found that six out of 10 respondents cited innovating to achieve competitive differentiation as a top business challenge. Additionally, the same percentage of respondents also agreed that their organization has more data than it can use effectively. Clearly, companies want to differentiate themselves from the competition yet are unable to effectively use their data to make that happen.
Disparate Sources of Business Data
Businesses have many different sources of data, each self-contained and built, if not for a singular purpose, at least to address problems in a specific business area. These data reside in four different data silos:
- Operational: Operational data contain objective metrics that measure the quality of the business processes and can come from a variety of sources. Hardware providers use sensors to monitor the quality of their implementations. Customer Relationship Management (CRM) systems track the quality of call center interactions (e.g., call length, response time).
- Financial: Financial data contain objective metrics that measure the quality of financial health of the company and are typically housed in the company’s financial reporting system.
- Constituency (includes employees, partners): Constituency data contain both attitudinal metrics as well as more objective metrics about specific constituents. Human Resources department has access to a variety of different types of data, ranging from employees' performance histories and completed training courses to survey results and salaries. Partner programs track partner information, including attitudes, financial investments, and sales growth.
- Customer: Customer data contain attitudinal metrics. Large enterprises rely on their Enterprise Feedback Management systems to capture and analyze data from such sources as surveys, social media and online communities.
Data Integration is Key to Extracting Value
Data integration is a difficult problem. Within a given company, data are housed in different systems. HR has their own system for tracking employee resources. The call center tracks data on their CRM system. Finance tracks their data on yet a different system. What approach can companies take to integrate all their data? In an interview, Anjul Bhambhri, VP for Big Data for IBM, talked about how business can solve their Big Data integration problem with respect to data silos:
"My response and suggestion - and we've actually done it with clients - has been that, you leave the data where it is. You're not going to start moving that around. You're not going to break those applications. You're not going to just rewrite those applications... just to solve this problem. Really, data federation and information integration is the way to go. Data is going to reside where it is." - Anjul Bhambhri, VP for Big Data, IBM
The problem of Big Data for businesses is one of applying appropriate data federation and analytic techniques to these disparate data sources to extract usable insight to help them make better business decisions. Companies who can extract the right insights from their business data will have a competitive advantage over others who cannot.
I believe a useful way to approach this data problem rests with a customer-centric approach of data integration and analysis. Next, let us turn to the field of customer experience management (CEM) to see how the application of Big Data principles can help companies gain insight from their business data.
In a study on customer feedback programs, I found that business data integration played a crucial role in the success of the programs. Specifically, loyalty leading companies, compared to their loyalty lagging counterparts, integrated different sources of business data (e.g., operational, financial, constituency) with their customer feedback data. By linking disparate data sources to their customer feedback data, companies gain insight about what is important to the customers.
When I asked Stacy Leidwinger, Senior Director of Product Management at IBM Vivisimo, about how she views the economic value of Big Data and customer-related data integration across the company, she says:
“This single view empowers the right decisions to be made around such areas as customer service, new product offerings and marketing campaigns, to name a few. Also, it also ensures that those on the front line can have a personalized interaction with the customer providing accurate and fast responses. Past systems have not delivered on this promise as the fight was always where to store the data. Now the combination of leaving data where it resides and ability to run deep analytics on external & internal customer data has changed the game and allowed organizations to service the new empowered consumer.” - Stacy Leidwinger, Sr. Director of Product Management, IBM Vivisimo
Next, let us turn to the formal area of customer experience management to understand how Big Data principles will change how businesses manage customer relationships.
Customer Experience Management
CEM is the process of understanding and managing your customers’ interactions with and perceptions of your company or brand. A CEM program consists of a set of organized actions that support the goal of CEM. While a CEM program has many moving parts, an easy way to organize those pieces into six components of a CEM program (see figure to the right).
A primary goal of CEM is to improve the customer experience to increase customer loyalty (e.g., customers stay longer, recommend, continue buying, increase share-of-wallet). Businesses that have higher levels of customer loyalty experience faster growth compared to businesses that have lower levels of customer loyalty. Businesses who have implemented CEM programs realize that these programs can be data intensive, generating millions of data points about their customers' attitudes, online behaviors, and even their interactions with a given employee, just to name a few. To optimize the value from these data, companies need to apply appropriate analytics to provide insights about how to increase customer loyalty.
The source of data in most CEM programs, not surprisingly, is customer feedback data. Businesses gain customer insight primarily by collecting and analyzing customer feedback data from different sources, including customer feedback surveys, social media sites, branded online communities and emails. Using customer feedback data, companies identify the customer experiences that are closely linked to customer loyalty and use that information to allocate resources to improve those customer experiences, and, consequently, increase customer loyalty.
Integrating different business metrics to understand how they relate to each other is sometimes referred to as the process of Business Linkage Analysis. How you integrate/link your different metrics depends on the problem you are trying to solve or the question you are trying to answer.
For example, here are three popular questions that can be addressed using linkage analysis of disparate data sources.
- What is the $ value of improving customer satisfaction/loyalty?
- Which operational metrics have the biggest impact on customer satisfaction/loyalty?
- Which employee/partner factors have the biggest impact on customer satisfaction/loyalty?
Each question requires different datasets, merged at the right level for the appropriate analysis.
Integrating your Business Data
The figure to the right illustrates some common ways companies integrate disparate data sources. The columns represent the different types of customer feedback sources and customer metrics. The rows represent the other data sources and metrics: financial, operational and constituency. Even though many different data sources can be integrated, I refer to this approach as a "customer-centric" approach because the data are organized to gain insight about the causes and consequences of customer satisfaction/sentiment/loyalty.
Depending on the question are you trying to answer, you will use a combination of different sources of data. For example, when dealing with questions around financial metrics, you can integrate those with customer feedback at the relationship level via relationship surveys and social media sources. For operational or constituency-related question, you will need to consider other data sources and integration level (e.g., link data at transaction level instead of customer level).
The process of merging disparate data silos depends on the question you are trying to answer. You will need to apply appropriate data federation and aggregation processes to build specific data sets for statistical analyses interpretation for each question. For example, you need two differ people studying the impact of employee satisfaction on customer satisfaction requires a different data set than when studying the impact of call center metrics on customer satisfaction.
This entire process of data integration is sometimes referred to as Business Linkage Analysis. The interested reader can explore the outcome of this data federation and aggregation process below. I developed three customer-centric data federation processes and data models to help companies use their existing data to address some of those Big Questions presented above.
1. Linking operational and customer metrics: We are interested in calculating the statistical relationships between customer metrics and operational metrics. Data are federated and aggregated at the transaction level. Understanding these relationships allows businesses to build/identify customer-centric business metrics, manage customer relationships using objective operational metrics and reward employee behavior that will drive customer satisfaction.
2. Linking financial and customer metrics: We are interested in calculating the statistical relationships between customer metrics and financial business outcomes. Data are federated and aggregated at the customer level. Understanding these relationships allows you to strengthen the business case for your CEM program, identify drivers of real customer behaviors and determine ROI for customer experience improvement solutions.
3. Linking constituency and customer metrics: We are interested in calculating the statistical relationship between customer metrics and employee/partner metrics (e.g., satisfaction, loyalty, training metrics). Data are aggregated at the constituency level. Understanding these relationships allows businesses to understand the impact of employee and partner experience on the customer experience, improve the health of the customer relationship by improving the health of the employee and partner relationship and build a customer centric culture.
How Big Data can Advance Customer Experience Management
Customer feedback is just one type of data that need to be analyzed and managed. By integrating different business data silos, businesses can more fully understand how other business metrics could impact or be impacted by customer satisfaction and loyalty. The impact that Big Data integration will have in CEM falls in three related areas: 1) Answering bigger questions about customers; 2) Building companies around the customers; 3) Using objective measures of customer loyalty.
Implication 1: Answer Bigger Questions about Customers
First, I think that the application of Big Data principles can help you ask and answer bigger questions about your customer. A successful CEM program is designed to deliver a better customer experience which translates into a more loyal customer base. The source of data in most CEM programs is gathered through customer feedback tools like surveys and social media sites. Businesses gain customer insight primarily by analyzing customer feedback data with little or no regard for other data sources. By linking disparate data sources to their customer feedback data, companies gain insight about their customers that they could not achieve by looking at their customer feedback data alone.
Businesses can now ask, and, more importantly, answer these types of questions.
- Where do we set operational goals in our call centers (e.g., number of handoffs, length of wait time) to ensure we maximize customer satisfaction?
- How many hours of training do employee need to ensure they can satisfy their customers?
- Which call center metrics are the key determinants of customer satisfaction with the call center experience?
- Where do we need to invest in our employee relationship (e.g., across the employee experience touch points) to ensure they deliver a great customer experience?
- Do customers who report higher loyalty spend more than customers who report lower levels of loyalty?
Companies who integrate their business data to understand the correlates of customer satisfaction and loyalty can better answer these questions and, consequently, have a much better advantage of effectively allocating their resources in areas that they know will help improve the customer experience and maximize customer loyalty and business growth.
Implication 2: Build your company around your customer
Next, Big Data principles can help you create a customer-centric culture. By integrating different sources of business data and uncovering insights about a variety of different metrics, you build interest across different organizations in understanding what is important to the customers. The integration of different business data would necessarily involve key stakeholders from each organization, and the mere act of integration would be a catalyst for further cross-organizational discussions about the customer.
Applying a customer-centric data federation and aggregation approach to business data integration helps senior leaders understand how their organization (and its metrics) impacts the customer.
Additionally, the results of customer research become more applicable to other organizations or departments when their data are used. Expanding the use of customer data to other departments (e.g., HR, Call Center, and Marketing) helps the entire company improve processes that are important to the customer. Here are some examples of how companies are using this type of research to build a customer-centric culture.
- Identifying and building customer-centric operational metrics for executive dashboards
- Removing the noise from executive reports by including only customer-centric business metrics (known to be predictive of customer satisfaction)
- Integrating customer feedback into operational systems (CRM) so front-line employees understand the interactions and attitudes of their customers
- Conducting in-depth customer research using all business data to continually gain customer insight and gain a competitive advantage
Big Data technologies and processes can go a long way in helping you support your CEM program. By taking a customer-centric approach to your Big Data, you will be able to literally build the company (its data) around the customer.
Implication 3: Use Objective Loyalty Metrics
Despite the existence of objective measures of customer loyalty (e.g., customer renews contract, recommends you, buys more), CEM programs rely on customer surveys as a way to assess customer loyalty. Measures of customer loyalty typically take the form of questions that ask the customer to indicate his or her likelihood of engaging in specific types of behaviors, those deemed important to the company/brand.
CEM professionals (me, too) typically use these self-report measures as our only measure of customer loyalty when analyzing survey data. While these loyalty metrics do provide reliable, valid and useful information, you are always interested in what customers really do. By linking up financial data and customer feedback data, you would be able to understand how the customer experience impacts real customer loyalty behavior using objective metrics, like purchase amount, products purchased, products liked, products shared and renewed contract.
End-of-quarter financial reports include customer loyalty metrics (e.g., churn rates, ARPU, repurchase rates) with no information about the factors that might impact those numbers. Traditionally analyzed at the end of the quarter as standalone metrics, these objective loyalty metrics provide no insight about how to improve them. Linking satisfaction with the customer experience to these objective loyalty measures, however, lets you build predictive models to help you understand the reasons behind your financial metrics. This is powerful stuff.
Could we stop using self-reported customer loyalty metrics? It would make the loyalty measurement debate a moot point. I think, though, the use of self-reported customer loyalty metrics will always be used. Survey-based loyalty metrics allow companies to quickly and easily gauge levels of customer loyalty and provide a forward look into the future about customer loyalty.
The era of Big Data is upon us and the Big Data problem for business is one of linking up their disparate data silos with customer feedback data in order to identify the correlates of customer satisfaction and loyalty. A major hurdle in solving this problem involves applying appropriate data federation and aggregation methods across the different data silos. This data federation process results in usable datasets with the right metrics culled from different data sources to answer specific questions or hypotheses. Once the metrics are pulled from their respective data sources, businesses can apply statistical modeling to answer important questions about the causes of customer satisfaction and loyalty.
Big Data principles have a role in CEM programs. Integrating other sources of business data with your customer feedback data can help you extract much more value from each of your data sources. By linking up these data sources, companies will be able to ask and answer bigger customer experience questions, embed the importance of the customer across different organizations/departments and provide the use of both subjective and objective metrics of customer loyalty.