Where does Collaborative Intelligence (CI) align best with your company’s strategic agenda?
Part 1 of a 7-Part Series on Using CI to Improve Company Performance
As I wrote previously, when speaking with company executives regarding how best to use CI to improve their company’s performance, I ask seven clarifying questions, the first of which is: “What strategic goals are you pursuing such that you think CI might help you achieve them?”
The purpose of this question is two-fold: it helps me understand the company’s strategic priorities, and their answers hint at their understanding of CI’s current and near-term capabilities.
Both parts are essential. CI is well-suited to support the achievement of some strategic goals, but not all. And I have found that many executives have a limited understanding or a misunderstanding of CI’s capabilities. This part of their answer highlights where and how I may have to adjust expectations about what is possible.
The most common strategic goal for CI that I have encountered is cost-cutting via headcount reduction. While this is understandable, I think it is short-sighted.
As I have written:
It’s called collaborative intelligence for a reason. The focus of all CI efforts, not just the initial few, should be on implementing workflows wherein intelligent machines support humans or work alongside them to achieve important business goals.
Certainly, some job types and skill sets will be less valuable or rendered obsolete because of CI, but the overall theme should always be that CI-centric workflows:
enhance human performance,
improve work life by offloading routine and uninteresting tasks to machines, or
make previously impossible tasks possible.
When just getting started, before there is any evidence within your company that this statement is true, you must be especially sensitive to this issue. I recommend that your initial efforts avoid projects that result in job losses.
Here’s another way to think about this: Suppose a company could apply CI to enhance human-in-the-loop work processes to create $100M in value and could also apply CI to reduce personnel costs to create $50M in value. Suppose also that this company pursues the headcount reduction first. In this case, it jeopardizes its ability to achieve the other $100M in value because realizing this additional value requires successful collaborations with humans who have just learned that CI is a threat to their future. Good luck with that.
How to apply technology to achieve strategic goals is a well-trodden path, so I want to focus on the second reason I ask this question. But for completeness, here is an unordered list of all the strategic goals that current and near-term CI technologies can help achieve:
Increase revenues
Create new products/services
Enhance existing products/services
Increase the pace of product/process innovations
Reduce costs
Operating Expenses (in general)
Headcount
Improve productivity
Improve organizational agility
Improve customer engagement and customer experiences
Improve operational visibility and decision-making
Retain talent / Improve Quality of Work Life (QWL)
Strengthen competitive position
Develop / Enhance strategic capabilities
Strengthen Activity Systems
My experience to date has revealed that most executives underestimate the current power of CI to their companies' potential detriment (they risk being put at a strategic disadvantage very quickly by those companies that recognize its potential and embrace it quickly).
To highlight this, here is a concrete example from a recent LexisNexis survey of 1,175 UK-based legal professionals regarding their expected impact of Generative AI (such as Large Language Models [LLM]) on various aspects of their work.
My first reaction to this was that the respondents were not conceiving of many other use cases throughout the professional services value chain. My second reaction was that they were grossly underestimating the potential of the use cases they did identify.
While I could comment on every one of these results, I will pick two: “Understanding new legal concepts” and “Coding for internal IT systems.”
The ability of LLMs to understand complex concepts and explain them in easy-to-understand ways is one of their well-recognized strengths. Why lawyers think this capability cannot be applied to new legal concepts is hard to understand.
Using LLMs to generate computer code may be the most widely used application to date. GitHub’s Copilot X is specifically trained to generate computer code. Thousands of research articles have been published on how best to apply LLMs to software development, and many Python (a popular programming language) libraries built on LLMs have already automated many programming tasks. To not rate this use case more highly implies that the respondents have a limited view of what's happening outside their areas of expertise.
As I wrote, gauging executives’ understanding of the strategic implications of CI and the magnitude and timing of its prospective impact on their business and industry is as important as understanding their strategic priorities.