First Steps Toward Customer Data Integration
By Scott Hambuchen
No doubt you're aware that customer information is potentially your company's most valuable asset. And most likely your business already has some inherent customer data integration (CDI) capabilities, such as basic merge-purge and record grouping.
But how accurate and complete is your customer data? Is it being duplicated through different business channels? Where precisely is the information warehoused? Is it contained in one location—or in several places throughout the enterprise? If you're not certain of the answers to these questions, you could be minimizing the effectiveness of that vitally important asset.
To help you begin moving toward enterprise-wide, up-to-date customer data integration, I've created several guidelines to consider. But before we can outline effective CDI practices, we need to establish what CDI means.
What is Customer Data Integration?
First, let's agree on a simple definition of CDI and why you may need it. According to the national research and advisory firm Gartner, Inc., CDI can best be described as "the combination of the technology, software, processes and services needed to achieve a single, accurate and complete view of the customer across multiple sources of customer data, databases and business lines."
CDI is the consolidation of all data sources used to create a single view of the customer. The core technology layer of a CDI solution includes the cleaning, linking and grouping of customer data, as well as the technology for putting integrated customer data at the points where your business needs it—in the call center, in the mail room or even at the cash register.
It's easy to see how customers can lose confidence in a company when it proves ineffective at the simple task of recognizing them when they enter through different contact points. Implemented correctly, CDI can increase customer satisfaction by maximizing the effectiveness of customer interactions, enhancing the ability to accurately identify new, profitable customers through a variety of channels, and reducing operational and marketing costs by more specific targeting.
Even with a basic understanding of CDI and what it can accomplish, let me add a word of caution. It can be especially difficult for small- and medium-sized companies to implement an effective CDI solution because they may not be able to afford to build the skill-sets needed to implement the processes and technologies.
There are usually two options in setting up a CDI process for most companies: 1) Develop all the necessary systems and expertise in-house (something that even the largest companies have had difficulty doing); or 2) work with a CDI solutions provider. Most large companies—those on the Fortune 1000 list—have tried the first option, and after much frustration, have now turned to the second option. They realized they didn't possess the amount of expertise to accomplish such an undertaking, not to mention the cost of acquiring the right technological solutions, especially in a down economy.
Whichever way you go, keep in mind that data quality is the backbone of successful customer data integration. A CDI solutions provider can conduct diagnostic tests that will quickly give you a snapshot of how good your customer data is—to what degree it is accurate, complete and up to date, and how it compares to benchmarks for data quality in your company's particular industry. If the diagnosis shows your data contains a high percentage of anomalies, consider bringing in outside data that will correct and update the records.
How important is data quality? Look at it this way: If you're working with bad data, you stand to lose a good deal of time and money because your marketing decisions will be made on incomplete or erroneous data.
Seven Keys to CDI
I've outlined the seven underlying areas that can drive the success of your implementation in both the initial and future stages of CDI deployment:
No. 1: Create an enterprise-wide customer information management plan. All key stakeholders—with representation from marketing, sales, customer service, finance, operations and IT, at the minimum—must feed the overall management plan. Gartner Inc. says many companies attempt customer information management plans "… with no idea of what they are hoping to build in the long term. One solution is haphazardly joined with another, initiatives come and go, and soon enthusiasm is waning throughout the enterprise."
Organizations spend vast sums of money on managing and using their customer information asset. As such, this asset should be covered by a comprehensive plan that does it justice.
The goal of the plan is to create a system that enables the company to "recognize" customers. Inaccurate recognition of customers, due to poor data quality, incomplete data and limited accessibility, causes a "ripple" effect across an organization that degrades business performance. Given the increasing volumes of data across multiple systems, lines of business and channels, even small levels of data fragmentation can create significant downstream customer management performance problems. The goal is to prevent this ripple effect from ever happening.
No. 2: Obtain resources to support the plan. Ensure that the resource requirements of customer information management/usage are sufficient to support the enterprise plan. Make it clear this is a specialist area where employee development is critical, and consider using the expertise of an outside consultant/provider to diagnose your company's capabilities.
CDI, by its nature, requires a range of skill sets, which in the open market are in short supply. It requires personnel who are trained in the implementation of customer relationship management (CRM) tools and technology, such as the ongoing management of customer data, from sourcing and procurement through quality control, definition, security, integration, maintenance, accessibility, value measurement and compliance with privacy regulations. Organizations must ensure they have sufficient resources in place to meet current, as well as future, business needs.
No. 3: Build measures of data quality. You need these measures to take remedial action when necessary. Data quality cannot be emphasized enough, so at this point we need to discuss it in a little more detail to get an idea of how data quality can and should be measured. There are four components of data quality:
* Data completeness—The percentage of all possible data sources and coverage across all data fields a company has accumulated in its decision-support and operational processes.
* Data accuracy—The overall exactitude of the data content and how well it contains identifying contact information and known data (internal or external) associated with each customer record. The question to ask, once your company has all the sources and fields identified and populated, is how accurate is that data content? In most companies the answer varies considerably with each source of data.
* Grouping accuracy—The accuracy with which a company can consolidate data from disparate sources. Once a company has all of the data sources identified and has performed the necessary cleanup to ensure they are accurate, how well can the company identify and group multiple occurrences of the same customer to provide a comprehensive, accurate customer portrait?
* Data access—The speed with which a company can integrate its data and provide that data in a usable form. If you're trying to beat a competitor to the market with a new product or service, how quickly can you put your data into play? Or, how quickly can you identify a current customer and treat him or her accordingly?
No. 4: Define your own "single view" of the customer. Whether your industry is financial services, retail, automotive, communications, travel or any other, this process starts with accurately recognizing customers through the use of names and addresses; demographic data; customer interactions; previous marketing efforts; risk scores; profitability (determining whether the customer is bringing profit to the business); and external data, which brings in more marketing information and credit risk knowledge.
Keep in mind that this single view is often a moving target, and the key is to build an understanding of what is actually required by your company, what is doable and at what price.
No. 5: Collect transaction history data. Hold and provide appropriate access to a minimum of three years of transaction history in a form that enables detailed analysis. Three years typically represents two or more sales cycles (except for some industries, such as automotive, where six years of data may be required).
Data should be held at the transaction level and should include a unique customer ID along with the date, product, volume, value and channel/outlet for each transaction. For analysis purposes, ZIP codes and customer segmentation-related fields also should be accessible.
No. 6: Use customer data to understand customer worth, lifetime value, preference and retention. An organization should be able to determine the worth of individual customers by combining sales margin, sales and marketing costs, and logistics and services. Armed with this information, an organization is able to make clear decisions about marketing activity.
Your company should also recognize the potential length of lifetime for new and existing customers, which ideally should be translated into an allowable cost-per-sale. Customer preference data is compiled by inviting customers to advise your company on their preferences around communications frequency, channel and time, as well as information sharing. This can be a powerful retention tool and a significant contributor to reduced operating cost.
And speaking of retention, a reason for every customer loss should be sought and stored in the customer database. Given the bottom-line benefits of retention over ongoing acquisition, any supporting information for retention is valid. Similarly, data events such as price inquiries, changing order patterns and lapsed accounts can be used as possible predictors of defection.
No. 7: Understand privacy and its implications in all relevant geographies. CDI technology makes it possible to quickly and accurately link related customer data, an ability not unnoticed by those who are drafting an expanding range of federal, state and global privacy regulations. For many reasons, including avoiding damaging negative publicity and maintaining your customers' trust, your business must monitor and adhere to all laws and regulations regarding privacy.
The best practices in this area go well beyond monitoring and implementing legislative imperatives. Companies can and should become strong consumer advocates. Our experience has shown that when given more control over their data, consumers become less concerned and tend to focus on the accuracy of the data (even to the point of volunteering information) rather than choosing to opt out of the database. And when the substantial benefits of having customer information are clearly explained and provided to consumers, most will see the value of the process. As customers enjoy the tangible results of the process—lower prices, more relevant offers, and improved products and services—their loyalty to the company is strengthened.
It's Your Move
At first, implementing customer data integration may seem like a daunting process. However, you should be encouraged by the vast majority of companies that have established CDI processes who now wonder how they ever did business without it. CDI is a proven commodity—time-tested, successful and measurable. With CDI in place, you will enjoy increased revenues, significant cost savings, higher efficiencies in marketing and, most important, happier customers.
SCOTT HAMBUCHEN is client services group leader for Acxiom Corp., which creates and delivers customer and information management solutions, including market-leading CDI services solutions, that help clients maximize the value of their customer relationships. He can be reached at email@example.com.