PROJECT TEMPLATE V1.2
Seizing an AI Opportunity
You’re on your path to AI leadership. Before you go back to the office, I want to make sure that what you’ve learnt here helps you in your job.
A. Could you identify a few opportunities to use data analytics to improve your team’s performance? For example, you can think about a usual bottleneck you encounter or one of its main KPI’s where your team underperforms.
Remember, these opportunities can be tasks of classification or prediction (or generation?).
(Aim to write down at least 3 opportunities, but feel free to write down as many as you want!)
1.
2.
3.
It looks like this is going well. Now, choose the most promising among these opportunities.
(C) A. Lanteri, Seizing an AI Opportunity v1.2 Sources: Columbia University (2018), Harvard ManageMentor (2021), MIT CSAIL (2017) 1
B. Using the matrix below, classify the opportunities you’ve identified above, according to whether you can expect a big or small business outcome (if it helps, you can reevaluate this point after you answer point D. below) and the degree of complexity (e.g., number of elements, different data points, stakeholders that need be involved…).
HIGH COMPLEXITY
LOW COMPLEXITY
Remember to look for a quick win, initially. You can always scale your efforts and ambitions. So, you should probably start with an opportunity in the bottom-right quadrant.
Choose only one opportunity for the rest of this exercise. (You are welcome to repeat this process for your other opportunities later if you wish).
There are a few more questions for you to answer.
C. Which template for value creation will you use?
Real-Time Responsiveness (react to changing conditions)
Hyper-Customization (tailoring to unique stakeholders)
Insight (reveal invisible connections between inputs and outputs)
Prediction (anticipate future contingencies and behaviors)
(C) A. Lanteri, Seizing an AI Opportunity v1.2 Sources: Columbia University (2018), Harvard ManageMentor (2021), MIT CSAIL (2017) 2
BIG PAYOFFSMALL PAYOFF
D. If this project is successfully implemented, what would be the value for your team? For your company? And for you personally? What are some measurable improvements you will strive for?
(e.g., better overall performance, lower chance of mistakes, avoid boring tasks, reduced costs, accelerated time to market, improved access to distribution, improved stakeholder satisfaction, higher chance of bonus/promotion, personal development, …)
Company:
Team:
You:
Client(s):
Customer(s):
Third-Parties:
(You can include more stakeholders in this analysis.)
E. What data would you need for your project? Consider first the information a human expert would normally use to make the same decision. Then consider other possible data that might be or not be useful but is normally avoided – perhaps because it’s too complex or expensive to collect and analyze…
(e.g., data about customers’ behaviors, their preferences, their social links, your supplier’s offerings, number and frequency of usage, state of consumables, conditions of use…)
Also review Harvard ManageMentor (2021), Surface the Data you Need.
(C) A. Lanteri, Seizing an AI Opportunity v1.2 Sources: Columbia University (2018), Harvard ManageMentor (2021), MIT CSAIL (2017) 3
F. How will you obtain this data? Do such data already exist? If so, where are they? If not, how can they be collected? If it cannot be collected, can you find any credible proxies? How will you convince your clients or partners to give you access? Clients may be more willing to share data with a company they trust if they see relevance or value to them.
G. How can you assess the quality of your data? Is it complete, unique, correct, consistent and timely?
Also review Harvard ManageMentor (2021), Ensure your Data is Accurate.
H. What kind of data analytics models will you need?
Descriptive (what has happened?)
Diagnostic (why has it happened?)
Predictive (what will happen?)
Prescriptive (how can we make it happen?)
(C) A. Lanteri, Seizing an AI Opportunity v1.2 Sources: Columbia University (2018), Harvard ManageMentor (2021), MIT CSAIL (2017) 4
I. Remember that data is only an input. In order to make a decision based on data, you need to generate an output of some kind. What output will your data generate? In other words how will you generate information that you do not know by using data?
(e.g., input: weather in a location > output: customer willingness to pay for beach resort; input: Google searches for apartments in a location > output: real estate prices…)
J. How will you use this data. In other words, are your data actionable? Look for data that is not only interesting, but insightful enough to drive decisions and actions.
(C) A. Lanteri, Seizing an AI Opportunity v1.2 Sources: Columbia University (2018), Harvard ManageMentor (2021), MIT CSAIL (2017) 5
K. For this project, who would you collaborate with [inside and outside your company]?
L. How could you run a small-scale test of your project, with little risk of backlash in case of failure?
M. Whose authority will your project question or undermine? Who might try to oppose your project? How can you obtain their support?
(C) A. Lanteri, Seizing an AI Opportunity v1.2 Sources: Columbia University (2018), Harvard ManageMentor (2021), MIT CSAIL (2017) 6
N. What leadership style do you need to embrace? Is your current leadership style appropriate to this digital transformation opportunity?
Now, finally, look at the last reflection question and answer it individually.
O. What will you do differently tomorrow at work and/or managing your team?
(Consider an actual behavior, not just a generic mindset or attitude.)
Thank you! It’s been a true privilege meeting you on your digital leadership path.
Best Regards,
Alessandro
(C) A. Lanteri, Seizing an AI Opportunity v1.2 Sources: Columbia University (2018), Harvard ManageMentor (2021), MIT CSAIL (2017) 7
