You should not use demographics when creating products or services. Reason #4 of 6: Academics also have bad habits of inserting demographics about very small samples. Don't accept the bad habit. Ask WHY? What is the purpose?

You should not use demographics when creating products or services. Reason #3 of 6: Economists (who habitually work with population data) have a bad habit of inserting demographics about very small samples. Don't accept the bad habit. Ask WHY? What is the purpose?

You should not use demographics when creating products or services. Reason #2 of 6: Demographics bleed over into propaganda, to make people think of a group as bad, show contempt for them, or give them unjust treatment. Demographics are truly dangerous in this context.

You should not use demographics when creating products or services. Reason #1 of 6: Thinking styles are not population-level (unless you work in government)

One reason why you should not use demographics when creating products or services: Intro: There ARE places where demographics are valid, such as in population level tracking.

😍Enrollment open: Finding Thinking Styles (6 weeks) starts 21-Aug, aimed at Australia, Asia, New Zealand lnkd.in/dZiJjxxg (4 price tiers) People don't all think alike. Don't expect your AI-solution to satisfy each person. What are the 3 or 5 or 9 thinking styles to train them on?

People are not all alike. So why do orgs still have the mindset of making one solution work for "everyone?" For example, A/B testing results in one answer. youtu.be/HTHHMzJDbHk

Many teams touch on cognition (thinking, emotional reactions, personal rules) in their research, but typically do not guide decisions by nor measure by it. Teams' processes are like wanting to run an ultra in the mountains on steep, rocky trails but only training on flat pavement. It doesn't work.

Measure the outcomes a person has. Their outcomes on their own goal (aim, purpose). Measure it using their cognition, and suddenly the whole "WHY" question gets answered clearly. (Plus everyone loves to see these measurements improve over the months and years.)

What is collaboration on a team (or across teams)? Collaboration requires: - Erasure of traditional power-based reward systems - Skill to hold ambiguity and keep iterating until the team gets deeper

Characters instead of personas: To tell a good story you need characters and a plot. Make up a cast of characters to use over and over, like episodes on TV. Each character represents one of the thinking styles.

When you use thinking styles, the point is to stop judging other people's way of doing something. You never try to change someone's thinking style. Instead you support the thinking style.

Don't make this mistake. Don't try to understand the whole person you are supporting with your solution or policy. Instead, only focus on the slice of what they are thinking about which you are positioned to support.

🧵The word "scientific" gets used in business as a synonym for "trustworthy" or "valid." Doing experiments like AB tests is not science. Hard science is practiced upon objects with no consciousness or agency, objects that behave the same each time.

Now we (will) have time to do the work of understanding people's cognition so that we can properly train our models. Skylines and thinking styles let your team stabilize your direction and explore paths no one else is exploring.

🧵 Thinking styles: Groups of people whose cognition is similar as framed by the person's purpose. (Not made up! Only a cast of characters is made up.) Thinking styles emerge from the listening sessions.

🧵 The value of a company like a B-Corp (Benefit corporation) is measured in part by how much their solutions help society and help the planet. Orgs in this age of AI can also add a measurement of how much they help individuals do what the individual intended to do.

🧵 If we want to admit to our assumptions, here is a common one. Teams divide behaviors by demographics, such as income level. The behaviors, instead, are guided by people's cognition.

When it comes to AI-generated images, Jessica Soto and I have developed a set of personal rules. In the spirit of helping us navigate this new age, we thought we'd share them in case they are helpful.

🧵 I love reading fiction. 💕 I remember one year I got to take a vacation, my suitcase was filled mostly with paperback books, not much clothing.

🧵 What would happen if your org could grow their market simply by recognizing (and supporting) different thinking styles? AI can help you ... if you help AI understand the thinking styles.

🧵 AI responds to human thinking. Building solutions with AI requires understanding how people think. Understanding how people think is what Data Science that Listens does. We've done this for 30 years. And it is NOT about people's opinions, preferences, and explanations.

What if the means of interacting with a product, service, or policy can change according to your thinking style? For example, a washing machine. Modern machines have all sorts of settings, but they're all there. It's not designed for your particular way of thinking.

🧵 Thinking styles are the only strategic practice that makes human variety visible and actionable — before ideation begins. Thinking styles surface the cognitive diversity that demographics and preferences miss.

🧵 "If you want to move the needle, you have to understand what your customers are addressing and measure the value for them. That's what that harm scale is about - actual value. ... Going back and mending existing solutions is very innovative. And it allows you to grow within the market you know."

🧵 Make time to forget about your product (or service, or policy). Forget the thing you're creating/polishing and spend a while remembering what people are trying to address. They might use your solution; they might use other solutions. They definitely use their memory.

🧵 In my own lifetime I've seen three different fashions of capitalism. So I know it'll change. (Quality metrics were popular once. Then innovation and distruption were all the rage. Now it's all about cutting expenses (labor!) and charging extra for features that used to be part of the package.)

Average user: A myth. You have a variety of people in your audience. One solution only fits half (or less) of your audience. –– Get the "I SPY: Data Science that Listens" game here: 🔍 indiyoung.com/wp-content/u...

🧵 Meetings or working sessions: which do you do? Meetings typicaly cause more work to be done afterward. By contrast, working sessions (or collaboration sessions) have "no homework." All the work is finished in the session by the people present.

Many teams limit their thinking by assuming they are only making one solution that must work for "everyone." And that solution gets implemented according to the thinking style of the team. So many other approaches get ignored.