In my last post, we explored the integration of entity resolution technology as a core component of a master data management (MDM) application, and I raised the question as to whether the rampant phase of acquisition and integration of entity resolution tools companies into MDM solutions providers implied that the utility of entity resolution was limited to MDM.

We can revisit some of the original uses to seek inspiration. One example is householding, a process originally intended to identify multiple individuals living at the same location, which makes use of entity resolution to look for enough similarities among a collection of records to demonstrate that they represent people living within a single household. In this case, entity resolution is used to link similar records based on home location. I could suggest that the evolved version of householding is more general – looking for record subsets in which there are recognized similarities and in which a single entity has decision-making control for the subset.

Another example is duplicate elimination, a process of looking for sets of records within a single data set that represent the same real world entity. Entity resolution is used as a data cleansing technique to match records that are apparent duplicates so that all but one can be removed from the data set.

There are many operational uses of identity resolution that are inlined into business processes requiring determination of unique identity, even in the absence of a master repository for that data. A good example in the banking industry involves compliance with Know Your Customer laws that require banks to execute due diligence processes to ensure against doing business with persons of dubious credentials or behavior.

So there is an argument for entity resolution surviving outside of the “realm” of MDM. But in practice, what does that really mean?