Big Data has been a central topic for corporations for many years
now. Typically, this is associated with how organizations use analytics to
figure out their most valuable customers, or to create new experiences,
services, or products. When devising this strategy, the organization must be
considerate of a few key factors:
- How will the data be used? What is the objective of obtaining this data?
- What story will the data tell?
- What does the data contain? Is Personally Identifiable Information (PII) or Protected Health Information (PHI) included?
- Who within the organization plans to use the data?
All these questions are key to develop the data management
architecture. The Architecture can be divided into three sections:
A. Data Management
o
The way the data is collected and stored
·
Data Security
o
Part of the data management plan, but
specifically focuses on the protection and transfer of data
·
Data Visualization
o
The output/analytics of the data that complete the
story. This involves using the data to influence actions within the company
As a procurement professional, one should consider coaching
stakeholders on adding structure to these three sections before establishing
their “Big Data” plan. When it comes to data management, a company can implore multiple
methods to ingest and manage data. For example, there may be one method for
handling customers that is then used for marketing, and another to handle
product testing data to influence product development. Let’s consider a real
example:
In the Pharma industry, understanding a patient’s lifecycle
journey is often critical to conducting research to produce new medicines for
the market. These companies need to understand how a patient may react/respond
to treatment even when they have not been treated by the company’s medicines. To
paint the full patient lifecycle picture, they need a lot of data from a lot of
patients around the world. The good news is this data is for sale. The bad news
is that the purchasing process can be tricky.
Patient data is protected by HIPAA (The Health Insurance
Portability and Accountability Act of 1996) Laws. This means that it’s unlawful
for a company to buy, use, or track health information that can be directly tied
to a particular patient without their consent or knowledge. But how do we
create lifesaving pharmaceuticals without understanding the people they are
meant to help?
We do something called, “Tokenization.” This allows companies
to aggregate patient data and then anonymize it so it cannot be connected and
tied back to any individual. By not linking this data to a name or person, we
can understand a patient’s medical history without ever knowing the patience. Instead
of John Smith, we now have JS100637. John’s name is never recorded or tied to
the new “Token.” John as a patient may appear in multiple datasets hosted by
various clinical sites that do not communicate with one another. But, by having
a token, John’s information is anonymously stored to eventually provide us with
the data that may create the next big vaccine or cure for cancer.
Big Data faces a lot of hurdles. Humans are resilient and
compassionate. We find ways around the hurdles while also respecting one
another and protecting our well-deserved privacy. In the world of procurement,
we can be the facilitators of this discussion, ensuring our stakeholders consider
each possible outcome and solution to the complex problems they aim to solve. The
relationships that are required in the previous example are vital to building a
stronger data management architecture. There could be one vendor to tokenize
the data, another to establish the data management structure and storage needs,
and a final vendor to address the visualization of the data. All must seamlessly
work together to create a comfortable user experience with optimized efficiency
and productivity.
