5 Steps To Clean Up Your Customer Data
Companies rely on data to make decisions, dirty data can create several challenges here are few steps to clean your customer data.
People hate their data being collected, they rely on data to make decisions. More companies are trying to be data-driven. However, bad data can pose several challenges for companies. Here, Christine Crandell, president, New Business Strategies, provides five steps to fix bad data.
Our entire lives are reduced to a constant stream of data bytes. Where we go by car or foot, what we had for lunch and with whom, to how well we slept (or didn’t) last night. Businesses, as well as consumers, are being digitized. That equals about 2.5 quintillion bytes of data created daily, according to WPDevShed. All that data is a treasure trove mined to make better decisions and respond faster to changing market conditions.
Though we grumble about the lack of privacy and security, we trust data enough to drive our cars, tell us what to buy, and what the next best action with a prospect should be. Companies are striving to be data-driven. Every decision is supported by extensive data analysis and justification. With that comes a sense of assurance.
It works well until it doesn’t. Then the issues become really obvious.
See More: How to Power Innovation With a Cloud-based Data Framework
The culprit is dirty data which carries a hefty annual price tag. According to Experian Data Quality and Gartner, there is a 12% lost revenue and $12.9 million yearly costs. That’s $600 billion in lost revenue annually in the U.S. alone.
Entire technology ecosystems exist to solve for data quality. Unfortunately, poor data quality is not a technology issue but a people problem.
‘Digital Trailblazer,’ Isaac Sacolick’s recent book, states, “You can’t easily move to reliable, real-time data processing when a human is in the middle muddling data from multiple sources.” People focus on completing tasks, and any resulting data is rarely considered. “The dirtiest data is always manual data,” shares Mike Snyder, director of North America sales and revenue operations, ChatLingual. As organizations grow and people come and go, the problem expands. Add in new data streams from websites, event registrations, market automation, and the frequency of incomplete, duplicate, and out-of-date data compounds.
Tom Treanor, chief marketing officer, Snipp Interactive, shares, “Organizationally, focusing on data quality can seem like a lower priority than opening a new market or launching a new product. But data quality improvement is a foundational area that can’t be pushed back endlessly as the problem only gets worse with time and growth.”
The data quality wake-up call for several clients occurred when launching a won/lost or customer survey program. They discovered customers tagged as closed/lost or with a preponderance of incorrect contact information. For others, newly formed MarketingOps or RevOps groups could not make heads or tails out of reports.
Regardless of how the data quality issue is discovered, the key question is how to fix it. The best practice is to implement a process for cleaning and keeping your data clean. Then address the people issue.
Five Steps To Fix Bad Data
Not all data is equal. Jeff Freund, CEO, Akoonu, emphasizes the place to start is to “clearly identify what data areas or elements absolutely must be clean versus those that are nice to have.” Prioritize the ‘must have’ data areas based on severity, business impact, and future need.
Fix critical data areas before tackling historical data. How far back in time should you go? That depends. The cost and time needed to clean all your data outweigh any benefit gained. Treanor advises, “don’t boil the ocean but focus on the most important areas first and get to a solid foundation for future growth.”
With the data areas and time horizon defined, the five-step process to remedy data integrity is:
1. Create a data council
The cross-functional council should regularly review data issues, define the data strategy, and determine ongoing data remediation priorities. The council should also agree on which roles should have access to the data and who should be able to edit what data types.
2. Define the customer data strategy
In the NewVantage survey, only 30% of companies reported having a well-articulated data strategy. A specific role or team should be designated as responsible for the customer data strategy, quality monitoring, and corrective action. Don’t make it someone’s side job; it must be core to their role and responsibilities. That demonstrates to the rest of the organization that data quality is a priority.
The strategy must include an audit of all data sources and determine which data elements are most critical to the business, customer acquisition and retention. Keep an inventory of data feeds and how that data is cleansed before being added to corporate systems.
3. Define manual data entry processes
In partnership with the sales, marketing, and ops, define policies to fix ‘go forward’ data input. It includes requiring specific fields, replacing free-form fields with drop-down menus, and requiring minimum characters for free-form fields. Balance the number of required fields with minimizing the impact on sales time. Use training and guidelines to help users understand the reasons for change.
4. Define marketing data feed policy
Marketing typically owns the bulk of data feeds into customer systems. While not as rife with errors as manual data entry, this data is far from complete and accurate. Marketing owns managing the quality of bulk data feeds and remediating bad data to meet data strategy policies and standards.
5. Ongoing data cleanse and augmentation
After steps 1 through 4, consider automating data cleanse and augmentation with applications such as Introhive and ZoomInfo. These should be routine, rules-based enrichment processes for customer and prospect data.
Automation will not fix all the data quality issues. You may need an external resource specializing in data clean-up to review and cleanse data periodically. Synder uses “leading indicators and dashboards of important metrics to forecast outcomes and determine when to invest time and resources for continuous [data] improvement.”
Depending on the size of your organization, getting your customer data in shape should take about six months. The majority of that time is spent on data cleansing, 30% on new policies and procedures, 15% on training, and 15% on technology automation, according to Nicole Fuselier, vice president of corporate and revenue marketing, Dremio.
See More: What a Successful CDP Implementation Actually Looks Like
People Need to “Get It”
One of the best ways to maintain trusted data is to hire employees that “get” the importance of data integrity. Chris Esposito, vice president of sales and marketing, StudioLabs, advocates that leaders must lead by example, visibly roll out new processes, and stay on top of compliance.
When it comes to customer data, how marketing and sales are measured has a considerable impact on data hygiene. Jeff Marcoux, marketing professor, Oregon State University, shares, “It comes down to how you are measured. Recognize that administrative work and data entry takes time away from selling and closing. So if you’re asking sales to do it, make sure it adds value.”
What steps have you taken to clean up your customer data? Share with us on Facebook, Twitter, and LinkedIn.
MORE ON CUSTOMER DATA
- Why Marketers Should Leverage Behavioral Data To Understand Customer Intent
- Leading First-Party Data Targeting with Consumer Trust
- To Build Customer Trust, Prioritize Data Protection and Personalization
About Expert Contributors: The Expert Contributor program is designed to help kickstart meaningful conversations around the priorities and challenges most critical to C-level executives. The insights and perspectives will help CIOs tackle what’s most important to them. We are always looking for industry thinkers who can help set the narrative for our enterprise audience. To know more about this program, and submit your ideas, reach out to the Spiceworks News & Insights Editorial team at editorial-toolbox@ziffdavis.com.