More companies than ever are interested in using mobile location data for market research, and to inform audience segmentation and business operations. As business interest grows, more businesses are entering the location data market. The problem is: most of these businesses are simply reselling raw location data to unsuspecting buyers.

For any company considering a location data purchase, it is critical to understand the difference between processed location data, such as that offered by Gravy Analytics, and raw location data.

At a glance, raw location data looks compelling: it’s less expensive to purchase than processed location data, and there is much more of it to work with. But companies considering a raw location data purchase should not underestimate the true cost of raw data, as it is more difficult to work with and will add significant complexity to any project where location data is used.

Raw location data is problematic, rife with fraudulent devices or signals, and many duplicate signals. About 45% of the data that Gravy receives is removed during data processing. Without this important step, our customers would have to deal with nearly twice as much data — much of which adds no extra value. Bigger is not always better: considering that our daily data deliveries range from 69.5GB to 155.5GB, the cost of storing and managing all that unnecessary extra data can add up quickly.

If you are not working with processed location data that has been cleansed and validated, you will need to figure out how to filter out any questionable data before you can start to use the data in your analysis. You will need to have a data scientist on staff help you identify and remove fraudulent signals and devices from the dataset, and to identify and segment out low-precision signals that are not suitable for your project. Gravy’s chief data scientist estimates that data preparation requires at least 5 times the number of working hours needed to perform the analysis itself.

When you work with processed data, the data does not need to be evaluated by a data scientist every time before analysis can start. Processed data is delivered to you ready-to-use, meaning that you do not have to do any extra data preparation work. Gravy makes it easy for a data analyst to choose just the subset of data that is relevant to his or her analysis. No data scientist is needed: a mid-level data analyst will have all the necessary know-how needed to do the work themselves.

Given the complexity of cleansing and validating raw location data, it can be tempting to just work with the data as it exists. After all, won’t it just even out in the end? Unfortunately, this approach is particularly problematic because location data is not predictably wrong. While 90% of location signals at one place may be accurate, in another place less than 5% of signals may be accurate. Similarly, some devices will generate fewer, high-quality signals, while others will generate thousands of low-quality or suspicious signals. The volume of fraudulent activity also ebbs and flows. Raw location data is not predictable, so any analysis performed using the data will be deeply flawed, at best.

When you choose Gravy Analytics for your location data partner, you have the assurance that you are working with only the best quality location data and datasets. Every location signal we process—more than 15 billion signals daily—is individually verified by our AdmitOne engine with Location Data Forensics technology. Most importantly, the data you receive complies with all applicable laws, including CCPA and GDPR.

If you do decide to work with a partner selling raw location data, be sure to understand the complexities of working with this data, and the true cost of cleansing and verifying the data on your own. If you are planning to use location data in an upcoming project, we invite you to evaluate our processed datasets to see the difference that our AdmitOne engine with Location Data Forensics technology makes. Schedule your no-obligation consultation today.