Personality Insights for Selling

By Robert I. Potschka

I show you how one of the largest, if not one of the worlds most recognizable of computer companies increased positive engagement rates by 78% using Personality Insights for Selling.

This book is the go-to resource for Sales Professionals, Sales Leaders and outlines a different approach to selling that for the most part, has been kept secret.

Included are Insights developed by psychologists that have never appeared in print. Part resource, part strategy, part persuasion techniques, they will validate what you thought you knew about you, and you of them.

If you have ever struggled with getting a prospect to respond to a message you have sent, then get ready for a whole new way to sell.

What you are about to read will be one of the most important things you will ever discover…Personality Insights for Selling!

• Language: English
• Publisher: Robert I. Potschka
• Release date: Jul 1, 2021
• ISBN: 9798201131159

Book preview

Chapter 1:  Start

WHEN I FIRST STARTED working with Artificial Intelligence, it was in the early stages of becoming mainstream. This was way back in 2013 when Microsoft launched its Language Understanding Information Service also known as LUIS.  As an early adopter, I joined their beta team to provide feedback and helped them design the system that it is today.

My first Artificial Intelligence model that I developed could determine if a sentence were Positive, Negative or Neutral. I integrated that with a handful of other Natural Language Processing models to support a Chat Analytics system I developed for the second largest employer in Canada. We analyzed 9,000 conversations a week, extracting negative sentences to identify weaknesses in a handful of products which in most cases included recommendations made by users, that helped us drive improvement across the Enterprise.

It was an exciting time, and I was hooked on all the possibilities.

The creation of an Artificial Intelligence model typically requires thousands of sentences that are mapped to one or more intents. Imagine two columns in a spreadsheet where one contains sentences (training data) and the other indicates the personality (intent) of the person that wrote it. 

Microsoft’s system allows you to import that ‘spreadsheet’ into a Model Builder. Selection of a button in the tool teaches an Artificial Intelligence model how to analyze sentences it has never seen before. In this example, it instructs the model to return a personality (intent).  It is then capable of making a guess on how close it resembles a sentence from its training data, which is accompanied by a score, indicating its confidence in a decision.

It is just like you and me. If we are learning something new and we were not quite sure of the answer, we might make a guess. If the score is below 50% an Artificial Intelligence model might return You might have an Openness Personality or if its above 50% Heck you have an Openness Personality.

Chapter 2:  What do you really want to do?

IN 2014 I DEVELOPED an online tool known as Text Analyzer which could analyze large amounts of text. It incorporated a handful of Natural Language Processing models that would parse texts and generate various results for offline analysis. One model could determine the part of speech of a keyword (Verb, Noun, etc.), another the theme or subjects in a sentence, extract keywords and identify various phrase types.

In 2017, I received a request from a start up in California to perform keyword analysis on unstructured texts that they provided. To suit their requirements, I had to create a custom version of Text Analyzer. 

At the time I didn’t know the real intention of the project. I completed the analysis and delivered the results on schedule.

While discussing the results, they explained, that what I had provided, would not really help them. I asked, ‘What do you really want to do’? 

They discussed my request and one of the partners said, ‘well he is Canadian, I think we can trust him’.

They wanted to determine a persons Personality from Written Text!

At the time, Personality determination from automated methods was unheard of.

I floated the idea of building an Artificial Intelligence model instead of using Natural Language Processing.  Having a bit of a background in Artificial Intelligence I knew a thing or two about it :-)

To build the Artificial Intelligence model, we had to feed its brain...

They had previously asked a group of about a thousand Sales Professionals to provide written descriptions of what they do at their company.  Those were the sentences I had analyzed previously, to try and find patterns.

What I did not realize at the time, was that they had asked them to take a long form manual test to determine their Personality.  

Psychologists validated their Personality from their written job descriptions by running a manual keyword analysis to ensure that their test results were accurate.  

You see, they could determine a person’s Personality by their written text; but it would take them weeks to manually run a keyword analysis for each person.

There must be a better way...

I loaded specific sentences written by Salespersons into a spreadsheet and tagged them with their respective intent (personality) and then loaded them into Microsoft’s Language Understanding Information