Why You Should Always Evaluate Data Quality

June 8, 2021

Having good quality data is crucial for data-driven decision making, especially in the age of big data. If you build on campaigns on unreliable data sets, you run the risk of getting lower conversion rates and returns than you were expecting.

In this post, we’ll discuss the cost of bad data, things to look out for when purchasing data, and the importance of quality data.

Why You Should Always Evaluate Data Quality

The Cost of Bad Data

The cost of making business decisions on bad data ends up being more in the end than what it would have been had you used high quality data in the first place. Let’s dive a little deeper into this: 

  1. You’re paying for all of the raw data, even though you won’t use half of it. This means that you are paying nearly double what you should for data that you don’t even need! 
  1. Not only are you paying nearly double what you should for the actual data, you’re also going to end up paying double for data storage. Once again, you’re doubling your costs… when only half of that data is going to be useful.
  1. Not only will it cost five times more to process extra data, but you also have to consider the data scientist and team of analysts. 

Bad Quality Data = Skewed Results

When your data quality is poor, you end up making data-driven business decisions that may be skewed. For example, let’s say that you are a retailer that is looking for a way to personalize a new campaign.  You decide that the best strategy would be to conduct an analysis of your audience post-COVID lockdown to see how much their preferences have changed. 

You notice that a lot of your customers are visiting Staples Center. Based on this data, you determine that they are likely primarily sports fans and create a campaign based on those insights. However, you have made assumptions about your target market with raw, low quality data that you are using. 

If the data had been higher quality, you would have been able to determine that those customers who were visiting the Staples Center that particular weekend were actually going for a Disney on Ice show. Instead of being sports enthusiasts, they are actually parents  who are interested in family-friendly events. Because of this new discovery, you decide to adjust your campaign to reach parents rather than sports fans. By using high quality data, you are able to increase foot traffic to your store and reach the right audience.

3 Things to Look Out For When Purchasing Data

When you’re shopping for quality data, there are some things that need to be considered. Here are 3 things that you should look out for during the data purchasing process:

Data Sourcing

When there are so many different compliance regulations for the collection of consumer data, as a business, you want to protect yourself from disreputable sourcing strategies. You want to look for regulatory compliance, and ensure that the data being collected is permission-based.

Data Cleansing Methods

We’ve covered why un-cleansed data is such a budget drag, but it’s important to remember that it all boils down to the data cleansing method. Make sure that your location intelligence provider is removing problematic or duplicate data before you purchase it.

Extra Data Filtering

Additional filtering of the data is a step that shouldn’t be missed because it identifies potentially misleading data. This method can help you know which data points are duplicates, erroneous, or fraudulent.

The Importance of Quality Data

The first thing that you should do when looking for a good data solution is to make sure that they are regulatory compliant. This is crucial, especially when the spotlight has been put on how consumer data is used. Ignoring GDPR and CCPA compliance, amongst other local regulations, can lead not only the location intelligence provider but you as well, into legal issues. At Gravy, we take consumer privacy seriously and ensure that our data platform remains fully transparent and compliant with industry and legal requirements.

Good quality data does not include redundant or irrelevant data sets that you don’t need. For example, let’s say that you are looking for foot traffic data for a local mall. There’s a busy highway next to it as well as a large wholesaler across the street. You find that raw location signals pick up location signals at the mall, wholesaler, and highway. This provides you with more data than you need. With the right location intelligence data solution, you can exclude the highway and the wholesaler (extra signals that you don’t need) and only focus on the data that you need from the mall. 

Ideally, your location intelligence provider should filter your data by origin and other key characteristics, and should be able to identify when signals might be corrupt . For example, with other providers, a device may seem to have travelled across the world, making multiple stops in different states before leaving the United States. With Gravy’s Location Data Forensics, you can determine that the device actually never left its original location.

Quality Data, Better Business Decisions

While we have covered a lot of what to look out for when committing to a data solution, it’s only the tip of the iceberg. To learn more about how to evaluate data for quality, watch our free webinar, “How to Avoid Purchasing Bad Data.”

Scroll to Top