Image for post
Image for post
Eli Rodrigues | Senior Digital Experience Manager | everis Brazil

Watson, Tell Me Who I Am!

Believe me, gone are the days when it was necessary to interact with a person to discover the details about their personality, needs and values.

Previously, so-called psychological tests (i.e. DISC, Palographic, graphological, Rorschach etc.) offered impressive results, but they required the person voluntarily answered a series of questions.

Nowadays, 1,200 words, which can be collected from newspaper articles, emails, blogs, social networks and even audio clips, are enough. It’s the missing piece for one-to-one marketing, a full-fledged system for bids, customer service, and customized experiences.

In this article, I include examples of personality tests, explain the overall functioning of a Watson Personality Insights (PI) API and I conclude by talking a little about the tool’s applicability.

Profile Analysis

To test the concept, I selected a few books from the public domain and applied their content to PI, which returned results in less than a second, even for texts with almost 1,000 pages.

In addition to the examples presented below, I also ran tests for Karl Marx, Shakespeare, H.G. Wells, Nietzsche and Oscar Wilde (spreadsheet available here).

Napoleon Hill — The Law of Success

Image for post
Image for post


Openness: imaginative, in search of creative experiences. A liberalist prefers to challenge traditional authority and values.

Conscientiousness: sets challenging goals and works to reach them. Takes rules and obligations seriously, even if they are inconvenient.

Introversion: prefers a full agenda of activities. Is reserved, quiet and prefers to be alone, though he feels comfortable leading groups.

Can you check with Eli to confirm? An extrovert/extroversion does not prefer to be alone.

Agreeableness: selfless and cooperative, an easy person to please, avoids confrontation and thinks it’s wrong to take advantage of others. Has a high self-esteem, is empathetic and believes in others.

Emotional Aspects: not easily provoked, is calm and sure of themselves. Has control over desires and deals effectively with unexpected events.

Needs: feels they need to keep things organized and under control.

Values: searches for new experiences.

Leo Tolstoy — Anna Karenina

Image for post
Image for post


Openness: prefers familiar routines, is aware of their feelings and how to express them.

Conscientiousness: doesn’t spend a lot of time organizing day-to-day activities and has trouble with repetitive tasks.

Introversion: is reserved and enjoys a quiet life[AGH1] .

Agreeableness: prefers not being the center of attention and doesn’t trust others easily.

Emotional Aspects: cares about what others think and sometimes thinks about things that make them unhappy.

Needs: prefers the stability of known and true activities and for the most part appreciates the views and feelings of others.

Values: doesn’t seek self-improvement and is not open to change.

Mark Twain — The Adventures of Huckleberry Finn

Image for post
Image for post


Openness: has a vivid imagination and is intrigued by new ideas.

Conscientiousness: happy with their accomplishments, they have difficulty with repetitive tasks and often doubts ability to reach their goals.

Introversion: serious, reserved and enjoys a slow-paced life.

Agreeableness: Is comfortable being in the limelight.

Emotional Aspects: Tends to be calm and sure of himself.

Needs: Appreciates order and organization.

Values: Doesn’t seek out personal pleasure or gratification.

How the Personality Insights API works

Image for post
Image for post

Scientific basis

This service is based on a personality model called Big Five, which separates a person’s personality into the following aspects:

  • Openness — Evaluates whether the person is open to experience a variety of activities.
  • Conscientiousness — Tendency to act in an organized or reflective way.
  • Introversion — Search of stimulus in the company of others.
  • Agreeableness — Person’s tendency to be compassionate and cooperative with others.
  • Emotional Aspects — Extent to which a person’s emotions are environmentally sensitive.

With regard to needs, 12 categories are included, which are based on Kotler and Ford’s work in the area of marketing, while the values section is based on Schwatz’s model in the area of psychology and seeks to determine which factors influence decision-making.

Linguistic inquiry and word count

The API uses a linguistic inquiry and word count technique to associate the frequency of certain words with a dictionary that contains the personality categories.

The dictionary was based on scientific research, but the key elements that generate the indexes are property of IBM, who carried out tests on more than 500 people to calibrate the model.

“Some words are often reflective, like: work, family, friends, health, money, feelings, achievement and positive and negative emotions.” (IBM Watson Personality Insights: The science behind the service)

Similarly, the statistical models of values and needs were designed through empirical research, comparing written texts with survey forms filled-out by approximately 1,000 people.

Results Format

The API output is a file (JSON or CSV), which shows the indexes and their respective values. In addition to the Big Five, subcategories were added that allow for more in-depth understanding.

Image for post
Image for post

To interpret them, they must consult the tables available in the Watson Developer Cloud, which gives more detail about the relationship between the principal/secondary features and help to understand the facets of each item individually.

Image for post
Image for post


Although our first impulse is to run the test on our own personality, we soon discover we’re more or less who we imagined to be and we look for more profitable applications.

The application IBM quotes is (obviously) discovering a person’s consumption preferences, including: buying preferences for music, health and physical activities, movies, reading, volunteering, concern for the environment and entrepreneurship.

In the Consulting world, I see a more applicable concept: finding out which customer profiles are more interested in a particular line of products, brand or feature and predict the best way to interact with them.

Soon we will go into a store and not only will we be greeted by our names, but we will also be treated the way we like. More reserved customers will be able to walk around the store in peace, while more outgoing customers will be greeted with a big smile.

The ads will also be based on similar mechanisms, eliminating mass stimuli, such as Trivago advertisements and the online courses’ war on all TV channels (Have you ever searched for a hotel on the internet?).

It’s unclear how much companies will influence or be influenced by our buying tastes, but it certainly will be a very interesting experience to live through. Are you in?


Exponential intelligence for exponential companies

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store