Posted by Brian Berg Google+
No matter how well the mailing list originated, there will always be a number of inaccuracies with every mailing list. So when a list broker says their list is updated every 30 days, what exactly does that mean? How do deceased records remain on a file for years? Is income range accurate? Why is it that the list isn’t 100% deliverable? The above questions cause a great many mailers concern. The simple fact is that no one list is entirely perfect at any given time and there are multiple reasons for inaccuracies. Let’s start with the most common. At any given time, a portion of the population in moving to a new location. Is that number 20% on average in any given year? This really depends on the market you’re mailing.
All things being equal, the younger 20 year old renters are far more transient than the senior homeowners. So even when a list is “updated” every 30 days, there will be a number of people who have relocated since the last time we checked but the list will contain their old address. Another reason is that unless a person fills out a National Change of Address card, it will take some time before we learn that this person has relocated. Again, some of the population that has moved complete the NCOA card before they relocate, while others wait for a month or two to fill theirs out. Compilers can also identify movers based on other new mover flags such as the relocation card of a magazine subscription. If you consider how some lists are compiled, you’ll realize that some people who’ve died years ago continue to pop up on a mailing list. This occurs because unless their name has been removed from public record, chances are; they will continually be added to the database. These name and address can be added to the DMA Deceased Database. Most all legitimate database compilers will pass their database against this deceased file and remove those records each time the list is “updated”. In addition to the names and addresses being inaccurate, there are also elements of data that suffer from minor inaccuracies.
Consider a number of “inferred” elemental data such as Estimated Income or Estimated Home Value. Though pretty accurate within a range, these numbers are in fact estimated. There’s no way to predict the value of a home. There are also interest categories which try to identify people with an interest in one particular hobby/behavior or another. Typically, these interests’ categories come from reliable self reported sources. Though self-reported data is considered accurate, when the person reported their interest could have been long ago. To control this, it’s always good to ask if it’s possible to restrict the interest select to being captured within the last 24 months. This keeps the data integrity in tact. One point I want to make here is that though people relocate, the interest category goes with them. So the next time you get a request asking, “when was the list updated”, explain that no matter how well the database is, there will always be degree of error.