Medium to large-sized firms invest heavily in data management. They lose more in missed opportunities with inaccurate data costing firms in the United States over $600B a year. Data management investments are focusing not only on collecting and storing data, but also on cleaning it up, increasing reuse, and improving its quality. Motivations for these investments include focus on creating hyper-personalized customer experiences, the need to make informed, timely decisions as well as legal, regulatory, and compliance requirements.
Data issues can be encountered in nearly every aspect of a business, from incomplete personnel directories in human resources, to siloed data required for complex risk modeling, to aggregate client total credit exposure across several business lines. In this two-part blog post, we’ll discuss how to improve data quality through the lens of a common business issue: campaign management.
5 Factors that Hurt Your Lead Quality
Companies rely on their list of target leads for marketing campaigns in order to connect with potential prospects and existing customers. Leads accumulated through advertising and organic outreach present a constant challenge with quality and accuracy. It isn’t uncommon to find data quality problems within these lists such as:
Empty fields (e.g. last name, address, email, phone number, zip code, etc.)
Out of date fields (e.g., email address is not current, address change)
Incorrectly formatted fields, which can lead to delivery errors or bounces
Data duplication (i.e., the same person or firm appearing more than once, sometimes with variations, such as David Smith vs. Dave Smith)
Inability to cross-reference with other datasets in your firms, such as transaction history (how much business have we done with this person) or cross-reference with external datasets (e.g. demographics, social networking activities)
These inaccuracies and shortcomings all add up to missed opportunities. They not only reduce the target marketing surface for your firm, but they also make it more difficult to personalize campaign content with effective personalized and highly contextual messaging. Thankfully, there are simple steps every business can take to improve their data management and targeting accuracy. These best practices work for data in campaign management systems and can be applied to many other datasets within the firm.
1. What Data Really Matters?
This first step is critical whether you have mountains of data or a relatively small set. Make sure you understand which data matters in order to achieve your current objective. What do you need the data to do for you? Taking our example of a list of leads for an email marketing campaign, you’ll likely want to focus on each contact’s first name, last name, email address, and job title at a minimum. You may also want to include other relevant history, such as transaction history.
Depending on the type of outreach you’re performing, home and cell phone numbers or physical addresses may not be relevant. Retaining this information is important, but worrying about improving its quality right now would require more investment of your resources than you actually need. Once you have developed a priority hierarchy for data quality attributes in a fit-for-purpose manner, you can move onto analyzing the actual quality of your data within that refined scope.
2. Quantify the Problem According to Experian, 69% of organizations across all industries say inaccurate data continues to undermine their customer experience efforts. In order to address low-quality data, you first need concrete metrics of your organization’s data quality. Start with the basics by using simple scripting or tooling to uncover statistics on your databases such as:
Percentage of the ratio of empty to non-empty fields.
Basic format adherence such as zip code, email address, first and last name.
Timestamp analysis to establish currency: When were records last updated? Are records stale, and therefore need to be removed?
Is the data duplicated? Look for issues with exact duplicates as well as fuzzy logic and machine learning tools that would detect patterns such as name variation (Jen vs. Jennifer, Bill vs. William) and other highly recognized dictionary-based substitutions.
Each of these scans will yield a score on your data quality scorecard, giving you an assessment on the current state of your data and its quality in a tangible, measurable metric.
3. Make the Business Case
Of course, any investment in data management will need a business case to quantify its importance. You set your goal and identified the important data in the first step. Then you measured the problem in concrete numerical terms. Now it’s time to paint the picture for stakeholders. Potential goals may include improving campaigns for higher sales, enhancing customer experience, reducing risk and fraud, or increasing the efficacy of billings and collections.
Be sure to define reasonable improvement targets. You are not going to overhaul your data overnight and, frankly, you don’t have to. You can even get started without spending any funds at all. You’ll be surprised how much lift you can achieve from the resources and tools you already have by assigning clear goals and objectives. Once this initial lift has been achieved, you can look at making investments in data quality calibrated according to estimated benefits.
These three simple steps will go a long way in increasing your data quality without requiring large investments or overhauls of internal processes.
Be sure to keep an eye out for the second installment of this two-part blog series, in which we dive into three additional steps to round out the best practices.
Join the conversation - leave your thoughts at the comments section below.