Medical, legal, regulatory, and editorial review has always been viewed as a bottleneck for getting critical information to providers and patients quickly.  Processes have evolved, agile teams have been implemented, and workflow tools have been enhanced to enable a faster, more streamlined promotional review process. However, Life Sciences organizations recognize that there is an opportunity to leverage technology to improve speed, quality, and profits even more.

When evaluating technology to improve content quality prior to MLR submission or to speed up manual marketing operations tasks, is it better to utilize a rules engine or machine learning? 

A rules engine is a good fit in the situation where:

1. You have the capacity and expertise for humans to analyze data and write the rules ahead of time for the logic you want to execute.
2. The logic you want to execute is precise and does not change with each reviewer.
3. The data you want to check is simple and consistent.
4. You have the capacity and expertise for humans to govern and maintain the rules if they need to change over time.

Machine learning is a good fit in the situation where:

1. You want the technology to analyze data and generate predictions based on past patterns of MLR approvals.
2. The exact logic you want to execute is not known and may change with different reviewers.
3. The data you want to check is complex.
4. You want the technology to automatically update its predictions based on each new approved data set. 

In summary, rules engines and machine learning each have benefits and can also be leveraged together in certain situations. However, there are cases when you might want to choose one over the other.  In the context of marketing content development and review, MLR feedback is not always consistent, the content is complex, and teams have limited capacity. These criteria make a strong case for machine learning over business rules for speeding up manual tasks and improving content quality prior to MLR.