Most businesses using bulk phone number lookup are loosing 10%-30% of their potential results because of poor setup in the spreadsheet file they submit for the search. For every one hundred phone numbers they search for, they get 10-30 blank results they shouldn’t get. This lower hit-rate costs businesses time, money, and opportunities. Luckily the fix is simple.
I handle customer support for BellesLink’s bulk search, which includes phone number lookups. Everyday I help customers who got poor results from a bulk search analyze the spreadsheet file they submitted and look for problems that caused a low hit-rate. I’m going to tell you how to setup your spreadsheet file so you can avoid the most common problems I see.
Garbage In, Garbage Out
In the early days of computer science, garbage in, garbage out described the concept that flawed, or nonsense input data produces nonsense output or “garbage”. The same is true for any kind of bulk people search like phone number lookups. Garbage data is the number-one cause of low hit-rates in a bulk search.
Bulk searches, like BellesLink’s batch people search, are automated. Once the file is submitted there’s no person scanning for obvious errors in the names and addresses being searched. There’s no person making a best-guess at mis-spellings and missing numbers. While our search software is excellent at making accurate matches, the law of garbage in, garbage out still rules the day.
Garbage Data Problems
Here are some common types of “garbage” data that cause a phone lookup to come back empty.
One of the required columns of data is blank in a spreadsheet row.
A missing zero from the beginning of a zip code is very common type of incomplete data. Later I’ll show you how to solve the “leading zero problem.”
Names are the source of the most common formatting problems. The image below shows the name incorrectly formatted as a married name. More about proper name formatting below.
Column names, particularly “first name” and “last name” can get switched and cause nearly all the searches to fail.
If your list quality is poor you may be searching the wrong combinations of names and addresses.
Of course, there are other problems which can cause a bulk search file to fail, but those are the most common.
But why are bulk searches so finicky about the data? Why can’t a bulk search just give me its best guess when some of the data is garbage?
To understand the answer to that question we need to understand the difference between bulk lookups and manual searches.
Bulk Phone Number Lookups vs. Manual Searches
To run a bulk lookup of phone numbers you’ll first upload a spreadsheet with your data. That spreadsheet is just a collection of single name and address combinations, each in its own row. The software’s job is to find the single best match for each combination. It’s a one-to-one match—one search, one result.
Let’s say some of the data was garbage and the zip code was missing a digit. Because of the error there were two records in the database which had an equal probability of being a match. What should the software do? How should it decide which is best? Should it return both results?
When the bulk search software can’t make an absolute match it returns no result. Even if the database contains several near matches. The problem is context. Remember, all the information the search software has to go on is a name and address with a bad zip code. It doesn’t know all the things you might know that would lead to a correct guess.
Manual searches are different and the difference is you. Instead of using a bulk lookup you could do the same search manually. The BellesLink Person Search will return the same information, and more, that a phone number lookup does. Though you would input the data yourself, one search at a time.
If you input the same name and address with the bad zip code the search would return a list of results with the best match being listed first. A manual search delivers a one-to-many match—one search, many results.
It’s your brain that can do what the bulk search software can’t—look at all the near-match results and choose, based on the context of the information, which is the best guess. If only something as smart as a human brain could be behind the software of a bulk phone lookup.
We’re working on that. : )
Best File Setup
Ok. We’ve talked a lot about the things that can make a search wrong. Let’s talk about how to make a search right.
Six Essential Columns for Bulk Phone Number Lookups
The setup of you batch spreadsheet file starts with the six essential columns your data needs in order to be searched. Nearly all companies, including BellesLink, that provide phone lookups based on a name and address search will require these columns of data.
- First Name
- Last Name
- Zip Code
All these columns are required for every spreadsheet row being searched. If any are missing the search aoftware will not make a best guess and the search result for that row will come back empty. Remember it’s one search, one result, so review your file to find any empty columns.
Formatting Names for Bulk Search
As I mentioned before, the mis-formatted names are the most common cause of a bad search result. When setting up your bulk search file, pay extra attention to accuracy and consistency of the data formatting in your first name and last name columns.
Separate Full Names
All phone lookup tools I know require a separated first name and last name. Because this is such a common challenge for our customers, BellesLink created a batch file uploader that can separate full names into two separate columns for you.
Separate Married Names
There is no way for the software to know which first name of a married name to use for the search. Separate these and choose to search one name or place each name/address combination in it’s own spreadsheet row.
Separate Partner Names
Like married names, there’s no way for the search software to know what do with partner names in the first name column. Separate these and place each name/address combination in its own row.
Tip: With a little bit of practice you can learn to use any spreadsheet application to separate names. Here’s one tutorial. The same company makes excellent software add-ons that make separating name data painless.
Remove Businesses, LLCs, and Family Trusts from Your Phone Number Lookup File
Many real estate investors use phone number lookups to get in-touch with property owners. Many properties, even residential, are owned by businesses, LLCs, and family trusts. These types of records cannot be searched with batch lookup tools, so they should be removed from your phone lookup spreadsheet.
You can search businesses, LLCs, and family trusts with using the manual searches in BellesLink.
To search a business or LLC:
- Use a Premium Business search to find the registered agent of the company.
- Do a Premium Person search on the name and address of the registered agent.
To search a family trust:
- Use a Premium Property search to find the property owner.
- Do a Premium Person search on the name and address of the owner.
- Alternatively, you can do a Premium Person search on the last name of the family trust and the property address.
Searching businesses and family trusts is a manual process that takes a bit of detective work. With practice you will learn to how to do it and your hit-rate will improve.
List Sources and Search Accuracy
In my list of common data problems I included wrong data as type of garbage data. This one is tough because you won’t know your data is wrong until you search it. If you do a bulk phone number lookup with a clean, well setup file and you get a hit-rate of less than 60%, the problem is almost certainly your data.
Purchased lists are more likely to contain old data and bad name and address combinations. List you build yourself through research or from your own customer base will have a higher data quality. Wrong data is the epitome of “garbage in-garbage out” and there’s no way around it.
Fixing the Leading Zero Problem in Zip Codes
It’s common to open your spreadsheet of name and address data in a spreadsheet application like Microsoft Excel and see that many of the zip codes are missing a digit. A zip code which should be 08807 is instead 8807. This is called the leading zero problem and it’s easy to fix. Here’s a tutorial on how to do it.
Deleting Extra Columns from Your Spreadsheet
Most bulk search services have a spreadsheet template they require you to use. The template will include the six essential columns I listed earlier. Most of the time, though, the spreadsheet you’re working from has other columns, too.
To run your batch you’re going to have to save out a separate CSV file and delete all the extra columns. That stinks. Now you have to manage two files and make sure nothing gets mixed up between them. For folks who aren’t fully comfortable working with spreadsheets, this can be the source of a lot of data errors.
At BellesLink we solve this problem by allowing you to upload any spreadsheet format. It doesn’t matter how many columns it has. In our batch file upload wizard you’ll choose which columns in your file should be used for each of the six essential columns a search requires. The other columns in your file will be ignored. You don’t have to delete extra columns and you don’t have to deal with the hassle of managing two spreadsheets.
Getting Good Bulk Search Results
Problems with data formatting and data quality have the biggest effect on the accuracy of your results when you do a bulk phone number lookup. You can get 10%-30% more results and overall more accurate results if you keep in mind the old computer science rule of “garbage in, garbage out.”
As you’ve seen in this article, most of the data formatting issues are simple ones. Develop a process for reviewing and error checking your bulk search spreadsheet before you submit. The BellesLink batch uploader includes a review screen to help you catch obvious errors. No matter what website you’re using for your search, use those review tools whenever you can.
Remember what you can’t search in a bulk phone number lookup, businesses, LLCs, and family trusts. Eliminating these records types will immediately improve your search hit-rate and lower your cost. Look for a future post on how to manually search those types of records.