The debate rages on: Is it better to license a business list database with the most comprehensive coverage or to prioritize quality? Is the best database really the one with the most records?
Good questions. Some database providers focus primarily on quantity, assuming that more data is more value. Others may not capture all businesses, but have a very high accuracy on those records they do have. When taking an in-depth look at your database provider, you should consider the 6 “V’s”, often used to describe “Big Data”: Volume, Variety, Velocity, Veracity, Value and Validate. These parameters help measure the overall usefulness of a business database.
How can you make sure you are getting the best business database?
Here Are 6 Questions to Ask Your Data Provider:
1. How Much of Your Data Comes From a Single Source?
Beware when a large number of these records come from a single source. These records are usually “quick adds” for padding total numbers and have not been fully vetted for accuracy. Records which can’t be verified by any other means are typically a “shell” company, has moved or is actually out of business. Business databases that are created for B2B Marketing should minimize how much noise there is in the data to improve the Veracity measure. By having multiple sources to validate a record -- the Variety measure is also improved.
2. Do You Verify and Enhance?
Simply having data for the sake of having it is not enough. The record needs to be telephone and/or web verified to confirm that the business is truly operational and that business is actually being conducted at that location. Plus, this verification process can allow the vendor to enhance the data by gathering additional information. Remember to ask how many people are involved in the verification and enhancement of the data. Validation from multiple sources improves the Value measure of each business record in the file.
3. Are There Software Processes in Place to Validate New Records?
By running the name through validation processes, your vendor may determine if the business: A) is a duplicate, B) is Out of Business, C) has Moved, D) had a Name Change, etc. These Software Matching Processes will kick out suspicious records for manual review to maximize comprehensiveness and quality. Ask to see some examples of suspect records that have been kicked out of a recent data quality process. If a vendor can’t illustrate their value by showing you examples of these processes – be prepared to have a lot of “noise” in the data that will impact the quality of the data you receive.
4. Do You Perform Data Updates On a Regular Basis?
Even if your data is compiled accurately, verified multiple times and enhanced by other sources — your database WILL be out-of-date in about 6 months. It is essential to have a data update process in place on a regular cycle that can detect moves, executive changes, businesses that have closed, etc. In addition, it is important to have a historical file of out-of-business records to reference for business hygiene. The proliferation of digital devices such as cell phones and internet social communities has led to an unprecedented rate of data creation and change. This Velocity measure drives a growing need for more frequent and more comprehensive update procedures.
5. Do You Rely on Web Scraping for New Record Compilation?
While web scraping is great to identify unique businesses that might be missing from some sources, there must be additional processes in place to keep those records verified and updated as well. The best approach is to combine web scraping with original source compilation from sources such as telephone companies, utility providers, Secretary of State filing and any other sources that identify the activity of a business coming into and going out of business on a weekly basis. This multi-source compilation process will also yield the highest number of records, the Volume measure, since it draws upon a wide variety of data sources.
6. Do You Provide Samples to Customers to Validate the Quality of Your Data?
Always ask for a sample of the data. It’s great to ask for all businesses in a ZIP Code you are very familiar with so you can determine if there are missing businesses, businesses that are in the list that are actually no longer in business, or gross errors in the size of the business or even the type of business they are in.
Bottom Line? Hold your data provider accountable for the business database they deliver. By asking the right questions and setting the proper level of expectation – you will get a database that is robust in quantity and quality.
When you have a business database that adheres to the 6 “V” measure, you will see a decided increase in your ability to utilize the full features of your data. No longer will you simply have “a list of names” to be used strictly for direct mail or telemarketing. Instead, by insisting on a full-spectrum and high quality business database, you will yield a multi-function marketing and business analysis tool.
Go ahead: Ask for better data.