What is Decision Intelligence? Many Concepts and ChatGPT as an Expert...

For my ninth #DecisionModelInnovation newsletter, I got inspired by a couple of new studies published by #technology & #consulting companies. Some of these reports/studies provided interesting inputs, while other reports were the usual 'marketing material'...you know what I mean.

The topic of these reports & studies I am referring to?

DECISION INTELLIGENCE...

My major issue with some of the recent reports about DECISION INTELLIGENCE (DI) was the confusion and ambivalent descriptions of what DI actually is. Until today, we neither have an agreement in academia nor the industry about why and how DECISION INTELLIGENCE is different or similar to other concepts. Below, I have listed a few definitions of people & organizations that have spent a little more time thinking about DI than the average technology or consulting company currently jumping on the bandwagon.

德勤 : A pleasant surprise from the consulting sector

In a very recent study (2023), Deloitte published a very insightful report about DECISION INTELLIGENCE, briefly summarizing their core understanding of DI as follows:

What I genuinely appreciate about this short statement is the inclusion of the term 'humanity' along with science & technology. Many concepts of DECISION INTELLIGENCE focus very much on data & algorithms and the possibilities of decision-making automation but often ignore the challenge of (cognitive) biases not only when it comes to the gathering and processing of data to turn them into insights but also the interpretation of insights to make a different type of decision - where AI/ML and big data have a supporting role only.

谷歌 's Cassie Kozyrkov: A Pioneer in Decision Intelligence

One of the leading authorities in DECISION INTELLIGENCE (Engineering) is Cassie Kozyrkov, who defined DI in her 2019 Introduction to DI as follows:

Gartner Group: The one that got DECISION INTELLIGENCE into the board room

While many consulting & technology companies have done a lot of good work to push the term 'DECISION INTELLIGENCE' into the general public (of executives), Gartner has probably been the key driver behind the rise of DECISION INTELLIGENCE since 2020 when they first included the term DI in their Top 10 Data and Analytics Technology Trends for 2020 report. Today (2023), their definition of DI is as follows:

I could go on with half a dozen other relevant descriptions of DECISION INTELLIGENCE, but you get the point that there are plenty of different interpretations and definitions currently flooding the market.

This inspired me to conduct a brief (and not very scientific) survey via a LinkedIn poll to understand better how the #corporate sector understands DECISION INTELLIGENCE. Unfortunately, these #LinkedIn polls don't allow you to offer very elaborate answers to choose from (30 characters, to be precise). However, these 30 characters were enough to ask for four different understandings of DI in a simplified manner:

• DI as "DATA-DRIVEN Decision-making"
• DI as "DECISION-DRIVEN Data Analytics"
• DI as "DECISION-based VIEW of the Company" (complementing the Resource-based View, Market-based View and Institutional View of the company; i.e. #StrategyTripod)
• DI as "something else" (and people could comment on their understanding).

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

The results after one week were as follows:

As you can see, the most chosen option was understanding DI as 'DATA-driven Decision-making'. While this wouldn't have been my primary choice I talked to some of those who had chosen this option and asked them why.

Interestingly, many of them told me that d) they were not experts for DI, so they wouldn't choose "something else", c) they didn't really understand the notion of "Decision-based view of a company", so that wasn't an option either (That's something I understand as you need to be familiar with key strategic concepts and ideas like the #DecisionFactory to make sense of this definition), and finally b) DI as "Decision-driven Data Analytics" which they understood has having a focus on data ANALYTICS rather than decision-making. So, they chose option a) highlighting the notion of DECISION-MAKING (although data-driven).

This explanation opened my eyes to how the corporate sector can perceive different (brief) definitions of DECISION INTELLIGENCE and where experts can get it wrong with how they understand DI if they don't broaden their horizon.

That's when I had the idea to turn to #ChatGPT (#GPT4) as an expert and let it chose one of the options (along with an explanation of why).

Imagine you are an expert for "Decision Intelligence" and you have to opt for one of the following options describing "Decision Intelligence" best:

• Option 1: Data-driven Decision-making
• Option 2: Decision-driven Data Analytics
• Option 3: A Decision-based of the Company
• Option 4: Something else. Please justify your selection in detail.

This is what ChatGPT answered:

I would opt for "Option 4: Something else". The most accurate way to describe Decision Intelligence is as a multidisciplinary field that encompasses elements of data analysis, decision theory, machine learning, and system thinking to improve decision-making processes. This discipline provides a framework that enables decision-makers to understand, analyze, and use the interconnections between various elements in a complex system to make decisions.

Justifying my selection:

Option 1: Data-driven Decision-making - While Decision Intelligence does involve using data to make decisions, it is more than just that. It's not only about making decisions based on data, but also about understanding the complexity and interdependencies of the decisions being made, considering the context and different possible outcomes, and learning from past decisions to optimize future ones.

Option 2: Decision-driven Data Analytics - Decision Intelligence does include aspects of data analytics. However, it's not just about using data analytics to make decisions. It also includes understanding the broader system in which decisions are made, considering the ethical and societal implications of decisions, and using a range of methods, including machine learning and complex system modeling, to improve decision-making.

Option 3: A Decision-based View of the Company - This option is unclear, but Decision Intelligence is not specific to any one company. It is a general approach that can be used in any decision-making context. [This explanation makes sense as "Decision-based View of a Company" is a term that has been hardly used so far, although it goes back to 1984 when JW Fredrickson describes such a concept in an Academy of Management Journal article.]

By opting for "Option 4: Something else", I am acknowledging that Decision Intelligence is a multidisciplinary field that involves more than just data and analytics. It includes a holistic understanding of the decision-making environment, considering the potential impacts and feedback loops, and using machine learning and other methods to improve the decision-making process. It's about making smarter decisions by understanding the broader context, potential impacts, and interactions of those decisions. It's a field that combines technology, human psychology, and system thinking to improve how we make decisions.

I then decided to go one step further and 'force' ChatGPT to choose among options 1 to 3.

Great choice. Now, if you don't have option 4 (Something else). Which of the other options would you choose? Please justify your selection.?

This is what ChatGPT answered:

Given the constraint, I would choose "Option 2: Decision-driven Data Analytics".

While neither of the remaining options perfectly captures the essence of Decision Intelligence, Option 2 somewhat better approximates the central idea. Decision Intelligence isn't merely about being data-driven; rather, it's about how we use that data, and more importantly, how we navigate the myriad decisions that data can inform. This necessarily involves data analytics, but the focus is the decision-making process itself and how data informs that process.

In the context of "Decision-driven Data Analytics", one could infer that the decision-making process is driving the need for data analytics, which is closer to the spirit of Decision Intelligence. This implies that we first identify the decisions that need to be made and then use data analytics to guide those decisions, which is an element of Decision Intelligence.

Remember, however, that while this choice may be the best among the provided options, it still doesn't fully encapsulate the scope and breadth of Decision Intelligence, as this discipline also incorporates understanding systems, human judgment, learning from past decisions, and considering the ethical and societal implications of decisions.

I am not arguing that ChatGPT is correct, but it simply showcases how people perceive DECISION INTELLIGENCE differently. In many circumstances, they mean the same but are misled by slight differences in wording or phrasing. Our own DECISION INTELLIGENCE NAVIGATOR at SatSure tries to integrate the key concepts as we have seen them in all definitions listed above:

1. Decision Context: Be aware of what kind of Decision you need to make and what open questions you need to be answered.
2. Framework Proficiency: Use the best of humanities, natural and social sciences to identify the most suitable frameworks, models or formulas to structure the answers to the open questions.
3. Intelligence Access: Use all kinds of technologies and approaches along the #datavaluechain (from new sensors to advanced machine learning) that modern #datascience offers to fill the frameworks, models or formulas with #data, #information or #knowledge.
4. Decision Proficiency: Avoid #cognitive #biases when turning insights (i.e. decision-relevant knowledge) into decisions - leveraging the most advanced insights from #neuroscience, organizational behaviour etc.

I've included a more detailed description of the Decision Intelligence Navigator HERE.

I want to thank?SatSure?for its continued support of this newsletter.?SatSure?is an excellent example of how?Decision Model Innovation?can lead to competitive advantages for organizations and their clients!