Typos, typoes or typo's? No matter how you spell the word, typos are a problem. Even the most accurate typist occasionally misspells a word, leaves off a letter at the end, or just types a different word than intended. Typos in a text to your friend aren't a big deal. They'll figure out what you mean using context clues. But in a business data scenario, if you're following the logic of traditional address matching, where an address is either an exact match or not a match, you can't account for different formats or misspellings.

We recently released a free ebook, 4 Steps to Unlocking Latent Revenue and Driving ROI with Your Address Data, in which we show you how to use your company’s address data to unlock latent revenue and drive ROI. In our last article, we went over address parsing. Today, we’ll explain address normalization and why this matters for data cleansing and analytics. Address Normalization (or making an address look normal) is the process of formatting an address so that it matches the proper address format of the national postal authority, such as the USPS for those in the United States.

Our global economy comprises many industries—healthcare, financial services, insurance, commercial development, and more. But one thing all companies have in common, no matter the industry? They’re all chasing a bigger and better return on investment (ROI). While there are many drivers of ROI, have you ever considered the role that address validation plays? Clean address data is essential for many business functions. Consider eCommerce, one rapidly growing example. Forecasting shows this retail channel will account for 22 percent of all global retail sales by 2024.

There are many problems you may run into if you're working with address data. In our recent webinar, our own Adam Charlton went over some of the challenges he's encountered the most, and how to deal with them. Here’s some of what he talked about. You can view the full recording at the link below. Standardization Standardizing addresses helps solve many of the problems inherent with data blending (combining two or more address databases together). In order to properly blend different address databases together you need to either have an external ID correlate the two sources, or have identically formatted address fields.