Do You Suffer From Bad Data?
According to Kissmetrics.com, bad data costs marketing and sales departments about $32,000 and 550 hours per sales rep. This translates to an incredible amount of wasted time and money across a company due to bad data.
What is bad data?
Bad data is any data that is inaccurate, outdated, irrelevant, statistically insignificant, or just plain wrong. For example, if your email database for your customers is filled with email addresses that are no longer active, or have been discarded, then this would be bad data. One source reports that 40 percent of the people who use email change their email address at least once every two years.
Bad data can also come in the form of low quality research studies, such as relying on consumer stated preferences versus observed results, or using insignificant data observations (one of my personal pet peeves!). Another example is that historical competitive pricing audits, regardless of how comprehensive, are often a poor indicator of what customers will be willing to pay in the future.
Irrelevant data can also be a problem. This is because collection of huge datasets not correlated to the decisions at hand is simply a waste of resources. While the success of data powerhouses like Amazon have glamourized the collection of broad uber datasets, the majority of companies would be better served by pursuing a targeted data strategy.
Why is bad data such a problem?
Planning growth efforts around bad data is like trying to navigate a ship with a south-pointing compass. It’s going to fail. Some execs may believe that any data is better than bad data. However, that is not necessarily true because you can waste a lot of resources chasing the wrong strategy or customers. Spending a tiny bit up front to reduce uncertainty and risk can save a lot of resources later.
Having accurate data is so important, that one source indicates that 54 percent of marketers say that the single most challenging obstacle to their success is bad or incomplete data.
These days, databases can also create an ethical, security and legal risk for your company. When collecting data, instead of asking “Why not?”, ask yourself “Why?”
What can be done to cleanse bad data?
Here are some strategies that can be used by your company to limit the amount of bad data and to increase the amount of good data that you collect.
1. Inaccurate databases
First and foremost, make it easy for your customers or users to update their data by themselves. This is one of the most inexpensive ways to keeping databases accurate and updated on a timely basis.
Second, share data between departments such as marketing, sales, and customer service operations. Doing this can help leverage the entire company’s resources to verify that data is correct.
Third, scan your data for anything suspicious or obviously incorrect, such as duplicate information, or other errors. Data cleaning software and outside data matching services can facilitate the process of periodical regular scans to help ensure that your data is clean and up to date.
2. Poor Research
To prevent poor research data, make sure that studies are correctly designed and that results are interpreted appropriately. Study design and sample quality should be carefully planned. Make sure your team understands the difference between qualitative and quantitative research and what types of insights can be answered with each. Quantitative studies require statistical modeling skills. When in doubt, bring in a specialist.
3. Irrelevant Information
Instead of casting the widest possible net to get the most data possible, focus only on collecting high quality data that is tied to business decisions. Maintaining massive amounts of ambiguous data can oftentimes be more trouble than it’s worth. At best, it takes time and energy and pulls away your focus from your core business decisions. At worst, it opens you up to inappropriate risk for your business model.
So, next time you’re tempted with the latest “sexy” data or research tool, ask yourself if you’ve fixed the basics. Otherwise, you may find yourself with bad data that is interesting to read, but not operational to use. Instead, identify the crucial information that you need to run your business. Where do you have the greatest uncertainty in your planning process? What pieces of information can be monetized for the greatest financial impact? Then, pursue this data and avoid collecting large amounts of useless distracting data.
Is it time for you to rethink your data collection and maintenance strategy?