Matching clients to their data is one of the most basic requirements of any data analysis. Although this may seem straightforward, it can become an extremely complex and subtle production issue, and is a common source of unnecessary cost, delay, and confusion in many matters.

A practice we regularly encounter that is extremely detrimental to tracking clients’ data is when employee names are substituted for unique employee identifiers. This can hinder even small cases where every person has a different name.

Names are always entered into data systems by a human. If a worker’s name is input multiple times across systems, then mismatches due to human error are almost certain in larger datasets.

While the human brain can easily review and understand that Jason Kandel is probably the same person as Json Kandel, a computer cannot easily exercise the same discretion. There are “fuzzy logic” tools designed to help resolve these types of matchups, and AI is getting increasingly better at it, but all the results still need to be individually verified, which is a time-consuming process and is not guaranteed to catch every potential issue.

These are the types of mistakes you don’t find out about until the case is over, the awards are out, and clients start calling to ask why their settlement check isn’t as big as their friend’s. We will always recommend fighting for unique identifiers in your production, especially with a large class.

In addition to being full-service wage and hour expert witnesses, we have a lot of experience working through data issues like these. We can help your team draft rules for production that address them before they happen, or bail you out if the data is already in hand. Reach out any time. Fairvaluemetrics.com

Below are real issues that we have encountered over the years.

• Simple misspelling of a name between systems (Leonhard Euler vs. Leonard Euler)

• Extra spaces between first and last name (Isaac Newton vs. Isaac Newton, identical to the eye but not to a computer)

• Last name first and first name last (Riemann, Bernhard vs. Bernhard Riemann)

• Introduction of middle initials in some systems (Carl Gauss vs. Carl F. Gauss)

• Middle initials with a period vs. no period (G. H. Hardy vs. G H Hardy)

• Maiden name mismatches (Ada Byron, Ada King, and Ada Lovelace are all the same person)

• Strangers with the same name combined into a single person (the badge records of one John Napier merged with the paystubs of an unrelated John Napier)

• A Jr. and Sr. being combined, even harder to fix when they share an address or phone number (Johann Bernoulli Sr. and Johann Bernoulli Jr., same name, same house)

• People with similar names having their production mixed up (Blaise Pascal vs. Blas E. Pascal)