It’s a fact that just about any modern business, regardless of the industry, is fueled by data. And perhaps the most vital piece of receiving and storing that data is assuring its integrity. For example, before an accident report can be imported into a claims processing system, certain data points must be validated. In some cases, the data then needs to go through a verification process as a further safeguard. “Validation” and “verification” sound so similar, it’s not surprising that the terms are often confused for one another—or even thought to be the same thing. As you’ve no doubt guessed by now: they aren’t.
What exactly is data validation?
In short: validation is looking to see if the data entered into a document matches an expected format or set of criteria. For example, you might need to confirm that the “5” you see entered into a form is actually a “5” and not an “S”. Or that zero is a “0” and not the letter “O.”
Data validation is also the process of confirming the proper order of numbers and letters. For example, consider a supplier code for customers in which there are first 4 letters, followed by 5 numbers. Validation confirms that the characters come first (and there are 4 of them), and the correct amount of numbers comes next. An International Bank Account Number (IBAN) consists of country codes, account numbers, and bank codes: a mix of numbers and letters that must be validated to ensure those characters are in the correct order.
Another example of validation is one in which there are multiple fields, an address form, for example. If the zip code is entered where the city name should be, that form cannot be validated.
What is data verification?
Verification, on the other hand, confirms the accuracy of data by cross-checking it against a third-party source of data. In other words, the data is not only correct in format and can be read, but it is “true.”
Consider an insurance company using ScaleHub’s claims processing service. As our crowd contributors validate uploaded information (repair estimates, proof of claim, drivers license, etc.), we engage with a specialist crowd that has access to a database of approved suppliers, provided by the insurance company. This specialist crowd can confirm that yes, the supplier name has matched with one in the database. The data has been verified.
How software and managed services help to speed up your data verification and validation
Because of advances in AI and optical character recognition (OCR), software can make a somewhat reliable first pass at “reading” data. It’s relatively straightforward to teach the format of an insurance company’s case or customer number to a machine-learning algorithm. But even an AI-powered OCR reaches accuracy limits when, for example, low-quality scanned images or hard-to-decipher handwriting come into play. And it’s these kinds of exceptions, when data can’t be read (and therefore neither validated nor verified) automatically, that require time-consuming manual processing by humans.
For organizations that need to process more data than their staff can, or should, handle, outsourcing is often considered the best option to seamlessly scale workloads and keep SLAs. Ideally, a service provider combines state-of-the-art technology with humans in the loop to offer optimum data validation at scale and with fast turn-around times.
At ScaleHub for example, our collective intelligence solutions bring together the best of AI and HI (human intelligence) to ensure unparalleled speed and accuracy to both validation and verification tasks. The data our AI extracts and classifies is handed off to a global, 24/7 available crowd to confirm accuracy and handle exceptions. To ensure data privacy and security, this data is presented to crowd contributors in decontextualized “snippets” for validation. (To learn more about how this process works, watch this short explainer video.)
For verification, the process can be customized to happen according to any other external need or regulation. Many organizations, and especially those from the medical or financial sector, must remain HIPAA compliant. Also, in some cases, specific expert knowledge is requested to validate or verify specific data. For these cases, ScaleHub uses so-called specialist crowds with the required credentials and expertise. (To learn more about our various types of crowds watch this short video.)
Now that you are more familiar with the difference between data validation and data verification, let us know if we can support your organization’s document and data processes with either (or both) and get in contact with us to learn more.
