Get Started

What Makes an Intelligent Organization? Insights From Morgan Stanley’s Chief Analytics and Data Officer

Data Basics, Scaling AI Marie Merveilleux du Vignaux

Organizations often say they want to be data-driven, analytically empowered, or even an AI organization. But what does it actually mean to be an “AI organization?” Have you ever heard anyone say they wanted to be a fax-driven organization or a phone-driven organization? Probably not. And that is exactly how Jeff McMillan, Chief Analytics and Data Officer for Morgan Stanley Wealth Management, explains how it sounds when organizations say they're aiming to become an "AI organization.”

He explains that AI is a tool just like mobile and the internet — a tool that we use to do better. It should not be a goal in itself. What’s important is to be an intelligent organization. In this blog post, we will focus on Jeff’s view of the intelligent organization which he shared during the just-released EGG On Air Episode.

→ Watch Full Episode Now!

What I would rather talk about is being an intelligent organization. In the world that I live in, that’s what I want to deliver.”

Jeff McMillan on EGG On Air
Elements of an Intelligent Organization

Jeff McMillan paints a clear picture of what a smarter organization looks like. An efficient, more intelligent organization has the following characteristics:

  1. Purpose: The intelligent organization sets clearly defined goals.
  2. A focus on empowering people to make the best decisions: The intelligent organization ensures that even the most junior person in the company has all the information they need to best serve clients they are engaging with.
  3. Continuous improvement: From a business perspective, an intelligent organization is constantly using data and analytics to learn what is working versus what is not. It is honest and open and uses data to drive decisions.
  4. A focus on scaling access to knowledge: Despite the power of Google and other search capabilities, it is difficult to get answers. Organizations rarely tap into their emotional capital effectively. The intelligent organization structures its knowledge and data so that it is accessible to all individuals in the company — that’s how you can make the most of your emotional capital.
  5. An organizational structure to manage it: The intelligent organization has a clear organizational structure including well-defined leadership, roles, responsibilities, and it aligns compensation structures against this structure. This effort is not just about one team — it’s about organizing the whole company around better decision making.

Challenges to Becoming a Smarter Organization

Being an intelligent organization is not easy. Organizations face a lot of challenges when trying to become smarter. Jeff notes three of the most common ones:

  1. Improving access to data: Jeff explains that most organizations don’t have a single curated high quality data source. They tend to have disparate data sources. And they don't have the infrastructure or even a strategy around bringing that together.
  2. Improving data quality: When it comes to data quality, the biggest difficulty in improving it is that the cost center that has the data quality problem and the cost center that can solve that problem are not the same cost centers. That means the ones who can solve the issue have no incentive to do so.
  3. A common lack of empathy: The reason many people fail is often not because the data was not presented accurately. It is because it was presented without empathy. For example, one might not have included another into the conversation early enough or explained the context clearly enough. Jeff always says to never show data to somebody’s boss before that individual has a chance to look at it and react. It’s even better to let that individual bring the data to their boss themselves as we want them to own the data and drive the decision. It’s all about empowering people!

How Morgan Stanley Does It

  1. Support employees: Jeff listens and focuses on the problems that employees have instead of trying to solve the problems that he thinks they have. He meets with employees to understand what they are trying to solve and uses his data to help them.

What we do is we help people make decisions.”

2. Understand what clients want and deliver that capability: Most just want access to their data! They don’t necessarily need fancy machine learning models — just deliver what they’re asking for. In many cases, they actually have the skills and capabilities necessary to reach their goals — they just need you to support and lead them through the process.

3. Create an organizational governance structure: We need to have data quality infrastructure, metrics around accuracy, definitions of what quality means to our organization, etc. We know who is accountable for that accuracy and have operational people monitoring it on a daily basis as well as issue management control, etc.

The most important thing we do every day is ensure the accuracy of the input. If you are not investing in that data quality/data governance infrastructure, you’re going to fail.”

You May Also Like

Who Should Deploy My Data Science Models?

Read More

Navigating the Data Provider Jungle

Read More

How to Measure AI Maturity & Value

Read More

Machine Learning Model Management: What It Is and Why We Need It

Read More