We applaud “data-driven” business decisions. How can we do otherwise when it seems the right thing to do? Wait. Wrong mindset.
Before we make it a habit to just nod our heads every time someone invokes their precious data, our instinct should always be to question whether the data is good enough to base strategic decisions on.
After all, there’s such a thing as dirty data which costs the U.S. economy a whopping $3.1 trillion annually, according to IBM.
On the homefront, individual businesses could be losing as much as 12 percent of their revenue due to bad data, based on a separate study by Experian.
Don’t think it can’t happen to you
Lest you think you’re immune to the dirty data epidemic, all companies unwittingly host bad data in one form or another. In fact, Salesforce noted that 91% of CRM data is incomplete and that 70% goes bad or becomes obsolete every year.
Horror stories abound. In addition to its bottom line impact, bad data mangles a company’s operational efficiencies and can trigger catastrophic effects on its reputation and growth curve.
For example, misclassification of customers at one commercial bank resulted in erroneous exposure estimates. Inaccurate data led the bank to believe that it had a well-diversified client base, concealing risks of overexposure to the natural gas industry.
When the sector suffered a market contraction, the bank incurred tremendous losses because it had an unusually large number of accounts in the southwest. Something that could easily have been avoided had the bank held accurate data.
What is dirty data?
Organizational success depends on a consistent stream of smart decisions. Smart decisions always depend on the right information. This makes data (re:business intelligence) among the most valuable assets a company can hope to have.
Unfortunately, most data comes raw and requires conscious effort to transform into strategy-grade information. High-quality data are the stuff that drives companies towards success and growth.
Dirty CRM data is a costly dead weight that can drag a company down and should be dealt with relentlessly to suppress the negative impact.
Types of dirty data
Dirty data refers to information that is inaccurate, fraudulent, invalid, duplicate, untimely, or incomplete. Inaccurate or erroneous data are valid data that provide wrong information due to misspellings, typographical errors, numerical errors, inaccurate contact info, and other factors.
- 1. Fraudulent data are false data that have been intentionally entered by humans or sophisticated bots in your CRM, primarily to undermine your competitiveness.
- 2. Invalid data are information entered in incorrect fields, records that crash your software, or information that your CRM cannot process properly due to incorrect formatting.
- 3. Duplicate data commonly refer to duplicated records of customers logged under several names, addresses or accounts, or in separate but unsynced software platforms.
- 4. Untimely data are information that are no longer current or updated and may be inaccurate.
- 5. Incomplete data are records that lack the relevant information for one or more data fields.
The impact of bad data
Having an ocean of data at hand is commendable, but only if your organization has the means and the mindset to keep a high bar on data quality. After all, information is the building blocks of sales and profits.
That goes only for timely, relevant and accurate information. In contrast, bad data can impede revenue growth, dent business reputations, and cause operational inertia. Teams may be forced to waste 50% of their time looking, verifying and correcting data. Time they could otherwise spend learning new skills, building new relationships or closing deals.
But things can get a lot worse than just operational hiccups. Incorrect laboratory diagnostics can kill patients. Wrong design specs based on inaccurate and manipulated test data forced airbag manufacturer Takata to eventually file for bankruptcy. Inaccurate credit assessments due to fraudulent financial statements can expose banks to potentially disastrous lending risks. Obsolete information that irked iPhone owners demolished Apple’s early ambition to dominate digital mapping and geolocation via Apple Maps.
The list goes on. Data quality is a problem every enterprise faces and the potential impact of dirty data has been chronicled for years:
- 1. Bad data leads to higher maintenance costs and higher resource usage.
- 2. Incomplete or inaccurate data lower metrics on customer satisfaction and retention.
- 3. Inaccurate data causes messaging/emailing delivery errors, leading to higher spam reports and opt-outs.
- 4. Bad data unnecessarily prolongs sales cycles and compounds related costs.
- 5. Dirty data mess up sales and distribution channels.
- 6. Bad data undermines sales performance metrics and creates tons of issues with reporting accuracy.
- 7. Incomplete, invalid or untimely data lowers overall productivity and cost-efficiency.
- 8. Low-quality data can lead to regulatory penalties due to non-compliance issues especially in highly regulated industries such as banking and pharmaceuticals.
- 9. Inaccurate data can negatively impact a brand’s online reputation and retard revenue growth in affected segments.
- 10. Bad data leads to bad reports, bad forecasts, bad decisions, and bad bottom lines.
How to keep your data clean
All businesses are exposed to dirty data, whether these data enter the system as accidental human errors, intentional flaws, floating data as a result of software migration, or simply raw, unstructured information. As Salesforce noted, even valid and accurate data will eventually become obsolete later on and need to get updated.
Given the consequences of inaction, cleaning your enterprise data and developing a corporate culture that values high-quality data cannot be overstated. Here are some ways to do both:
- 1. Establish the case for high-quality data and secure support from top management and staff alike.
- 2. Prioritize data quality assurance in all aspects of the organization. Formulate and implement a comprehensive policy on data quality.
- 3. Set a data quality baseline and let everyone start scrubbing existing data to meet the standard.
- 4. Identify points in the workflow (such as manual data entry) where higher incidences of inaccurate data are likely to occur.
- 5. Set manual and automatic procedures to screen out or minimize incomplete, inaccurate or duplicate records.
- 6. Automate data entry whenever possible to reduce the impact of human errors.
- 7. Maintain data quality assurance as an ongoing commitment and a corporate mantra.
- 8. Get data entry right the first time, moving forward.
Not all data-driven decisions are good, simply because some data are dirty. And they’ll remain so unless we start cleaning up our CRMs. For many companies, maintaining high data quality will require a ton of effort but it’s the only way to future-proof your profits.
FREE WHITE PAPER: Top 84 Chief Marketing Officers
Gain access to the world's most influential Chief Marketing Officers.
Latest posts by Max Altschuler (see all)
- The Most Common Problems With CRM Data (and How to Fix Them) - July 26, 2017