Scenario planning is a disciplined way to think about the future using plausible stories. The Large Language Models (LLMs) like ChatGPT are the perfect tool to generate those stories. Especially because they might be wrong but sound plausible.
Source: “Scenarios in Strategic Planning: Full Guide with Examples” bscdesigner.com
When I was writing about scenario planning in the domain of strategic planning, my definition was:
Scenario planning is a disciplined way to formulate strategic hypotheses in the context of existing driving forces and their uncertainties.
To keep it simple, scenario planning is about these three components:
- A plausible story about the future
- Early-sign indicator of the expected uncertainty
- Prevention or response plan
I’m pretty sure that AI chats based on LLMs like ChatGPT are good for generating plausible stories, but let’s try it for other aspects of scenario planning.
PESTEL Analysis
A PESTEL analysis (analysis of the external environment and its driving forces) typically precedes scenario planning. In the most basic case, it’s about doing a Google search and finding the trends that might hit your organization.
Source: An Example of Using PESTEL Template for Strategic Planning. bscdesigner.com
Let’s ask this question to OpenGPT:
“What are the trends that will impact most organizations next year?”
Here’s what AI chat says (sorry for sharing yet another screenshot from OpenGPT):
A screenshot from OpenGPT (The trends that will impact most organizations next year).
As expected, the answers were pretty obvious. I hoped to find the Extreme Weather Events or Climate Change among the trends, but it looks like the way the question was asked (focusing specifically on “organizations” and “trends”) confused the AI.
Let’s continue with “Increase focus on sustainability.” I’ve recently shared my version of how to align the strategies of organizations and Sustainable Development Goals 2030, so it will be interesting to see what plausible stories AI can generate for this scenario.
Plausible Story
After some attempts with the chat, I’ve found the formula that generates results in the needed format for most of the trends. Here is an example of the question:
Imagine a situation where sustainability reporting becomes mandatory for European companies. What impact would it have on the operations, talents, customers, finance?
The answer was:
A screenshot from OpenGPT ( Imaginary situation that sustainability reporting becomes mandatory for European companies)
To summarize the answer:
- Impact on operations: new processes needed to collect data and report on it
- Impact on talents: bringing on board people with the relevant skills
- Impact on customers: the need to communicate about sustainability efforts (the AI is optimistic and doesn’t mention the greenwashing if you don’t ask about this challenge)
- Impact on finance: additional costs + financial challenges for small businesses
The answers are far away from predicting the future (we are not doing this in scenario planning) but fit pretty well in the idea of “plausible story”.
That’s pretty close to the semantic level stories used by the Shell company, a recognized pioneer of scenario planning.
The choice of the projections (operations, talents, customers, finance) was not random. My intention was to make the story resonate more with the perspectives of the Balanced Scorecard.
Early Sign Indicators
Now the difficult part — the early sign indicators.
The question I formulated:
“What are the early sign indicators that sustainability reporting becomes mandatory for European companies and how can those early signs be quantified?”
A screenshot from OpenGPT (Early sign indicators that sustainability reporting becomes mandatory)
Let’s try to find some viable indicators among the answers:
- The number of countries that adopt new legislation — I would not call it an early sign indicator, but it’s a good lagging indicator
- The number of investors who request sustainability information from companies — that’s my favorite, actually in my article, I was arguing that having a quantified alignment with SDGs will make conversations with investor stakeholders easier.
- The number of sustainability-aware consumers. In this case I additionally asked the chat to be more specific about the ways to quantify it. The suggestions were to use data from surveys, sales data, online searches, social engagements. These are interesting suggestions, but I’m not sure there will be enough data to work with.
- % of companies in the industry that are reporting on sustainability. A kind of social proof indicator — nice.
Reformulating questions in different ways, I was able to get some better answers focusing on media coverage and public discussions of new legislation projects.
Response Plan
In the final step of scenario planning, we need to develop a prevention or response plan.
My question was:
“How can a software company be prepared for sustainability reporting becoming mandatory for European companies?”
The AI chat, in this case, gave a general answer about the need to study the new legislation, create reporting procedures, tweak sustainability-related features in the software, and update communications with the customers.
Well, that’s not the level of detail that you want to have for your response plan of some important uncertainty, but it’s definitely a good starting point for the discussion with your team.
Conclusions
Scenario planning is not about predicting the future — this business tool works with plausible hypotheses or plausible stories based on existing driving forces. With that application area in mind, AI tools like ChatGPT are a great alternative to gathering information with search engines.
