Read More Cybersecurity
WhenDeepSeek, the new Chinese AI technology, launched last week, as the most downloaded app in Apple’s US Apps Store, it caused—to appropriate Hunter S. Thompson’s famous title—fear and loathing in the global financial and technology sectors. In a dramatic one-day selloff, the technology-heavy Nasdaq Composite slid 3.1% and the S&P 500 sank 1.5%, while DeepSeek’s principal competitor, Nvidia lost more than $590 billion in market cap.
But the dark cloud for business results has a silver lining for business presentations because DeepSeek’s technology uses a process called distillation that provides an object lesson for how you can organize your content and develop your stories.
A Wall Street Journal article explained that AI from leading companies like OpenAI and Anthropic “teach themselves from the ground up with huge amounts of raw data—a process that typically takes many months and tens of millions of dollars.”
The parallel in story development is that business people have huge amounts of raw data in their minds that they have amassed over the course of their careers and the new information at their current company. When they start to develop their stories, all that information comes rushing forward in voluminous bulk. Think of this as the forklift approach to story development. If all that information is not distilled, the story becomes a data dump that inevitably overwhelms the audience.
This approach is often summarized with the tale of the person who, when asked the for the time, tells you how to build a clock.
This dysfunctional process is further complicated when the presenter starts to develop the story by writing full sentences and/or designing slides; adding even more bulk to the raw data with considerations of word choice, punctuation, grammar, syntax, font style, color, size, etc. You get the picture.
The Wall Street Journal article went on to explain that DeepSeek’s distillation is a new system that “learns from an existing one by asking it hundreds of thousands of questions and analyzing the answers.” Translating that to story development takes four simple steps:
Set the context. Decide the goal of your presentation and analyze your audience. This simple step helps focus what content you actually need while reducing the volume of the raw data. It also gives the lie to the concept of the “company pitch.” One size does not fit all.
Brainstorm. Do the data dump in your preparation not your presentation. Get all the ideas out of your mind and onto the external medium of your choice, be it a legal pad, Post-its, whiteboard, or computer screen. Be sure to scribe those ideas in as few words as possible. Think of each idea as a headline. In crafting headlines, you avoid the descent into the deadly weeds of wordsmithing.
Cluster. Find relationships among the many ideas and form them into clusters. You’ll find that ideas have natural affinities. Again, this reduces the volume of the raw data.
Distill. Limit the total number of clusters to no more than six. Five is better, four is better still. Less is more.
Sequence. With only four to six clusters, you can then proceed to organize them in a logical structure that makes it easy for you to tell and—most important of all—easy for your audience to follow.
The result may not be a revolutionary as DeepSeek, but your audience will be grateful to you.