Our Lead Validation services take Name, Email, Address, Phone Number, IP Address, and optionally business names, in order to identify the best leads and fight fraud. We are constantly improving our algorithms, adding new test paths and datasets. Both Lead Validation and Lead Validation International use the same underlying set of tools, data and services except for the additional international datasets the international service has access to.
Both of these services have had some major updates in the last several months, as well as a few exciting things to come. This article will provide a overview of these upgrades.
Recent improvements
We added the ability to detect vulgarities in the business name. Names and emails were previously being checked, and now businesses are checked as well. The quality of vulgar name detection has also greatly improved.
The garbage detection algorithms of emails and phone numbers were revamped to be more accurate. These algorithms check for unlikely patterns and obviously bogus data inputs. In the case of phone numbers, we have improved our level of detection, while for emails, we use even more flexible and accurate criteria. These updates also include more grey area partial points or penalty type results.
Business to Email logic has been completely reworked, and now extends to work internationally. However, the main update is to the algorithms and tests comparing the email to the business. A combination of a known good email with a match of a business name to either the mailbox or the domain can be a good indicator of a solid lead, but it can be extremely challenging to match them, since both business names and emails can be freeform. This new update does a good job of partial name matches or initial matches.
A few examples of this might be:
Big Bad Business, Inc. would match bobsmith@BBBInc.com
Computer Science Experts would match bobsmith@compsci.com
In a similar fashion, we continued to improve our matching algorithms in other categories as well, such as business name matching phone contact name. As with businesses and emails, the same kinds of major differences can exist in the phone records. This algorithm has been reworked to also consider variations of abbreviations and initial version of the name.
Lead Validation has also added foreign character detection, to detect names in other language sets such as Russian names in Cyrillic or Japanese names in Kanji. In the past, some of these names were identified as garbage and given an unfair penalty, given that Lead Validation International is an international service. The name still has to make sense in the context of the lead, however. For example, an American lead with a given personal or business name in Cyrillic does not make sense and will continue to be penalized. This is developing technology and will continue to evolve.
A smaller update was also added for IPv6 style IP addresses, and basic syntax checks have been added for IPv6. While Lead Validation is not performing more advanced IPv6 location tests, prior to the update, IPv6 formatted addresses were returning syntax errors. IPv6 formatted addresses should expect average scores in the 70’s for the IP component. Addresses not adhering to either IPv4 or IPv6 formats will still receive more severe penalties for invalid syntax.
What is on the horizon for Lead Validation?
There are always lots of new things being added to the Lead Validation services. Algorithms for improving matching between data such as names, business names, emails, and phone numbers are constantly being enhanced, but we have three interesting projects being incorporated into these services in the near future.
In the next month we should be adding a new update that will allow companies to track emails and IP addresses that are hitting the system too fast. For example, we might add optional penalties for the same email or IP address registering multiple times on a site, to help companies identify potential fraudulent usage.
We are working on translation software to further improve the validation of leads with foreign language set characters. These are becoming more and more common with international leads, and while we can now work with these foreign character sets, being able to translate them will only improve that. For example, most email addresses do NOT include foreign characters. This would make a match with a name or business name more challenging. However, with a translation there is a chance that we can make additional comparisons that were not possible before.
We will also be working on improving notes for components like phone number and email, to help clients make better decisions on certain fields when a lead is deemed back. For example, a lead with bad data including a garbage phone number such as 666-666-6666 might receive a low score but have a contactable email address. For example, Bob Smith and the email BobSmith@Gmail.com seem to be connected, and with an identification of a good email mailbox, may actually be reachable. But if they submitted garbage data for the rest of the inputs, it would be hard to give the lead a good score. However, saying this is a bad lead but it has a potentially good email is still something many companies can make use of, and we will attempt to provider cleaner ways for companies to use these details.
Lead Validation and Lead Validation International are continually adapting services. Feel free to contact us to learn more about how these services might help you and your business.
