Address Validation – US and its various iterations have been verifying, cleaning, standardizing and appending address data for over 20 years! We are constantly looking for ways to add value to the service, update our algorithms and make the service perform better.
This blog introduces a new feature to Address Validation – US as well as some other recent additions
High Risk Addresses
The main addition to Address Validation – US is the concept of High Risk Addresses. This update allows us to flag certain addresses as “Potentially” risky addresses to send a valuable package to. They may be an indication of bad or fake data or they may signal a potentially fraudulent attempt to gain something of value while trying to remain anonymous. For the most part these are good addresses and wouldn’t otherwise be flagged or marked as anything but good addresses. Despite being a good address though, there is a reasonable likelihood that they are not the intended or expected recipient. The system we have incorporated into Address Validation – US will allow us to add new risky address types over time but for this iteration we are flagging Freight Forwarders, Prisons and Hotels.
Freight Forwarders are (mostly) legitimate services that can be used by fraudsters. Freight Forwarders deal with importing, storing and shipping of goods on the behalf of customers. They act as a middleman for getting goods from one place to another. A fraudster might use a freight forwarder to set up a delivery of a valuable item, gather the item, default on payment and disappear into the night. We can’t know if an address submitted that’s linked to a freight forwarder has fraudulent intentions, but we can give our clients the tools to make better decisions when figuring out how to deal with these types of addresses. This address:
10006 Secretariat Dr Goshen, KY 40026 is linked to Welco International Services, Inc. in Goshen KY.
If this was submitted to our address verification service we would return a flag in our DPV Notes output: “The input address matched a freight forwarder”. We are not claiming this to be a fraudulent address, just suggesting the potential for it and that the user should take care in sending anything valuable to this location.
Hotels, like Freight Forwarders could be used by a fraudster to hide their true location and identity. It may be a legitimate delivery to a Hotel but any delivery sent to a room that may or may not be a long term resident is a potential risk. This address:
10007 Princess Palm Ave Tampa, FL 33619 belongs to a Quality Inn & Suites in Tampa FL.
If this was submitted to the service, we would return the flag: “The input address matched a hotel”. We do not generally validate room numbers in a hotel, so a record marked with a DPV flag of 3 (meaning the room number is not valid) would definitely be highly questionable. Adding room 400 to the check would cause the Hotel flag and a DPV 3 to trigger. The suggestion here being that the requestor is in a room in the hotel.
Finally, Prisons may not have quite the same potential for fraud that Freight Forwarders and Hotels might have, they could still lead to some potentially costly deliveries if the intended recipient should not be receiving a package. This address:
1000 Diamondback Rd OK 73772 belongs to the Diamondback Correctional Facility near Watonga OK.
If this was submitted to our address checker service, we would return the flag: “The input address matched a prison”.
All of these new flags in the DPV Notes response can be used to trigger additional reviews by our users prior to sending the package or storing bad data into a CRM. Again, while we cannot say that these are bad addresses, it is certainly valuable to know that they are linked to freight forwarders, hotels and prisons.
It might be valuable to pair the identification of a potential high-risk address with another service, Lead Validation, which looks at additional elements like name, address, email, phone and IP address to determine if a lead is good.
Other new additions to Address Validation – US
High risk address identification is the latest update to the service, but there have been other updates before that that are interesting. We have made some new efficiency changes to the services in the background that will help with performance, it might not be noticeable except for really complicated or messy addresses because for a clean address it balances out well with the new checks.
The most interesting new addition though is to the services ability to parse and handle complex single line addresses (addresses that appear all on a single line not broken up into Address1, Address2, City, State, Zip). We have made significant improvements to our address parsing algorithms that allow us to better deal with parsing addresses that either have extra items in them or misspellings. In the previous iteration, we would struggle with addresses that had missing elements (city, state or zip codes) and other errors/issues like misspellings, extra unexpected characters or overly long non-essential data points.
This address: 148 Aple Ck 43612 should be 148 Apple Creek in Toledo OH, and previously failed because of the combination of a poor Address1 input and important missing elements like city and state. If either was present, we would have been ok but the combination was too messy to fix. After the update we are able to handle the poorly spelled case with missing data points no problem!
This address: 35 N Indin Vly Ct Little Egg Harbor Twp NJ 08087 is on Indian Valley Court in Little Egg Harbor Township NJ. A long complex city name in addition to multiple spelling errors make both parsing and validating a big challenge. We are now able to handle this one as well with no problems with the new build.
If you would like to learn more about our address validation tools, please feel free to contact us, we are always happy to help.
