Text Analyses

Text analyses are based on thousands of texts that psychologists have assigned to certain forms. With the help of “artificial intelligence”, new texts are compared with this treasure trove of assigned information. From this, classifications with certain probabilities can be derived. For example, a phrase like “I want to have things under control” refers to the expression “power motive”. It is logical that a machine can make such calculations faster and more accurate than a human can do.

Another possibility is to feed the “machine” with two information packets to each test person. These are, on the one hand, texts that someone has written about certain questions or tasks – and, on the other hand, the data of his characteristics that were calculated with a valid personality analysis (if possible from different tests). In both systems, the quantity of texts determines the quality of the text analysis.

For the development of ECL (extended computer learning), the IPM authors use this starting point for a further level of consideration. If it is clear from the results from the M.L. and the text evaluations in comparison to the values of the test persons who prefers which words, a prognosis is possible for each word, which says with which probability it speaks for a series of defined values. For example, the word “we” is found more frequently in contact or integration types – the words “I” or “man”, on the other hand, are found more frequently in the texts of individualists or rational types.

Now and in the coming years, different machine learning systems and algorithms, such as ECL, will support each other in the OPEN IPM system. But even today, remarkably detailed results are being achieved.


The above example “I want to have things under control” shows these detailed results:
Here the IPM scale 0..200 with the mean value 100 over all texts of a certain category is used. A transfer into other scales is possible and also intended for the API.

Normal values

As standardized input, ECL uses the summarizing evaluation of all available texts from different sources, such as image descriptions, textual answers to various questions and an extensive prose collection of modern authors, including press and advertising texts. The results for a newly given text thus indicate the extent to which the words used deviate from these average values.

In the example above, bottom line, comparatively many words are used that are typical for people with a greater focus on “continuity” and very few that are typical for a result focus.

Situational results

During the development, the question was examined to what extent the emotionality of the language and thus the characteristics to be derived vary from context to context. In a test, different scenarios were asked, for example “What do you want to achieve professionally?” versus “What do you like to remember when you think of a nice holiday? As expected, the choice of words was clearly influenced by the context of almost all test persons.

Further scenarios were examined. Consultations on technical issues were recorded in a test environment. In other projects, typical job descriptions and a catalogue of job descriptions were analysed. In each case, deviations from the standard values were more or less pronounced. A statement like “I want to have things under control” is more likely to apply to a professional situation than to an enjoyable holiday.


These experiences led to a context-specific evaluation. The “average values” determined from a series of tests for a given context are checked for their average deviation from the “standard values”. Cumulated average values (CMA) are formed for each characteristic within a pipeline, which are then used for a statistical correction.


The Weightet moving average method can be added to this for specific projects. The amount of information (number of characters) is used as the weighting factor. This example shows the evaluation of the communication with a customer:


Integration into existing software solutions

Text analytics can be used as a web service. A request is sent with a text. Either a determination of the standard values is commissioned or a statistical adjustment to a previously agreed pipeline is requested. Unless otherwise agreed, the response consists of the 24 values. These are either processed further by the existing software solution or serve as parameters for retrieving further evaluations or result presentations.

Profile comparisons

“One thing is not enough for everyone. See everyone how he does it…” (Johann Wolfgang von Goethe)

The request of a software asks: What is the right thing for me? And what’s the best way to do it? Of course, the response from the profile comparisons does not answer prosaically, but with interpretable data.

Let’s start with the technical structure. The prerequisite for a profile comparison is that there are profiles. These are either determined in the “analysis” step or lie, such as “professional profiles” in an IPM library. Then these typical queries can be made:

  • Compare = Compare the profile to ID-x (e.g. superior) with that of ID-y (e.g. employee)
  • Select = Search for profile ID-x (e.g. job profile) the best 10 matching from the series ID-1 to ID-n (e.g. applicant)

The technology doesn’t care about the content. It is only a question of answering the user’s questions (request) with facts (response). For example, the contents can look like this:


The next step is to interpret and present the results, for example:

Interpretation: The offer fits 80% to the customer. Their needs for implementation (e.g. speed), individuality (e.g. exclusivity) and empathy (e.g. environmental friendliness) are greater. The need for integration (e.g. fashion) is lower. It makes sense to adapt the offer.

Relate values to one another

The web service responds to the request with comparative values for the 24 types of potential analysis. The values relate to different emotional or decision-making levels.


This complexity makes it possible to use the comparison functions also for other personality systems or typologies.

The matches or deviations can either be interpreted by the application software itself or, in the next step, control which results IPM should deliver.

The profile comparison focuses on two essential things: the best fit and the efficient complement.

Experience has shown that profile deviations of less than 20% are tolerable in most cases when persons or offers are matched, and that a similarity can be said. Larger deviations lead to a different way of thinking, feeling and acting depending on the depth (see picture above).

This being different has great advantages in working with other people, because different competences complement each other. Also in partnerships the “foreign” brings something fascinating with it – if there are enough similarities as a basis for the relationship.

Good fits

Profile comparisons are used for the fit.

  • We want a professional task that completely fulfills us.
  • We want to offer something the customer likes.
  • We look for the optimal way to learn or teach.
  • We reduce personality-related wastage in marketing.
  • We are looking for the best possible employee.

When our inner values, goals and aspirations deviate from what the situation offers us, stress arises. We want to have a balance in ourselves and to our fellow human beings and strive for balance. If this is not possible, it can endanger our health or reduce the necessary profits in relation to our business environment.

Good supplements

For the inner balance and for the interaction with other people, however, it is also true that potential arises from opposites which, when combined, have a strong synergetic effect. When we talk about personality traits, we mean these different forces that move us in different directions. A part in us wants to go high (red) and its counterpart (green) needs to be firmly anchored in the ground. The red one alone would leave the earth, the green one would sink into the ground.


If we accept both sides, then red supports with assertiveness and green with risk assessment. This principle of mutual support also applies to the other colours. The real and successful life is colorful.

Profile comparisons also serve the purpose of optimisation.

  • We stabilize developments.
  • We combine rationality and emotionality.
  • We want to use diverse competences.
  • We learn in a holistic way.
  • We avoid objections and resistances.

The examination of a profile shows which aspects are on average less pronounced for all comparison profiles, so that additions are useful.

Areas of application

The comparison functions are intended to provide information, answer questions and help to check decisions. Here are a few examples:

Which professions suit me?
How do I learn best?
Which offers offer me which benefits?
How can I integrate into the company?

How does my partner fit in with me?
How can I support my children, friends or partners?
What do I and my loved ones need to be satisfied with themselves and each other?

Which applicant fits the job well emotionally?
Which questions should I ask in the interview?
How can superiors and employees improve their communication?
Which training measures suit which employee?

Which employee fits the team?
What competencies are missing in the team?
How can cooperation be improved?
How do I lead the team as a whole and the employee in particular?

Which customers do we address with this product?
What kind of communication suits the target group?
How do which contents affect which customer types?

How do I have a successful sales conversation with this customer?
Which arguments are positive for this customer?
What objections are to be expected?
Who can best take care of this customer?

Sometimes the answers result from the comparison of two profiles, often it is useful to relate several profiles to each other and sometimes it can be helpful to reduce the number of potential comparison profiles by a selection.

Not every software package will want to cover this itself. Therefore, prepared results can also be “ordered” via interfaces instead of data. The results can then alternatively be displayed in a separate browser window, in a frame or as an Excel table for download. More information can be found in the article “Results”.


Integrate Questionnaire

The task of a questionnaire is to collect information from which certain characteristics can be inferred. The IPM questionnaires use either ranking methods, text analysis or a combination of both.

IPM uses an analysis generator for the texts of the questions and the assignment of items. This makes it quick and easy to adapt to customer requirements. 1 to 30 questions can be defined according to the ranking procedure or as text entries.

Each questionnaire type can be integrated into software environments. The call, for example by an HR or CRM application, is made as a “single sign-on” with the transfer of parameters such as the desired questionnaire version, the ID and optionally name, gender, contact type and language.

In a separate browser window or within a frame defined by the application, each question is displayed after the call. If not set otherwise, the analysis is performed immediately after the last question and the result is displayed.

As an alternative to this procedure, the questions can also be programmed on the application side. In this case, the software sends the user’s input via web service for immediate analysis.

Ranking procedure

On the left side texts or pictures are displayed one below the other. The user has the task to move these items into the right fields, sorted according to his own preferences. For example:


Text questions

The user should answer open questions in writing for the potential analysis, or copy existing texts, such as job descriptions or e-mails, into the text field for external profiles.

For the self-profile, several tasks are set one after the other, such as:

Inside-questionnaire(2)2.pngAt least 50 characters must be entered or copied into the field before the next question or result is called.

Foreign languages

Most questionnaires are available in German, French and English.

All languages that are supported by Google Translate can be used to enter text through text analysis.

Typically, German or English questions and hints are used to control text analysis. If translations are available, a French, Italian, Russian or other language user interface can also be agreed.

Browser window or frame – questionnaire and result display

The questionnaire is processed either in a new browser window or in a frame of the controlling software solution.

If the integration does not provide otherwise, the results are displayed to the user at the end of the questionnaire processing and a PDF document is offered for download.

Various formats are available for the presentation of the results, which can largely be adapted to customer requirements. Here is an example:


White label solutions and integration into other questionnaires

The questionnaires can also be designed and controlled on the pages of the software solution. This is useful, among other things, if different questions are to be combined with each other, as is useful, for example, in market research.

In this case, the parameters of the rankings (e.g. question 2, positions 3,4,5,1,2) are transferred via an API. IPM then answers with a type number and the analyzed values.

Interactive Advice System

In principle, the same procedures also apply to the IAS system. The call is made with the customer ID and the agent ID because various statistical usage reports are generated.

Usually, the IAS is part of a CRM system and is started when the data for a customer is called and closed again when processing is finished. Sometimes an icon in the CRM mask will also be used to start the IAS.

The IAS is almost always designed individually for a customer service project. All input and output elements can be freely defined. Here is an example:
Inside-questionnaire(6)6.pngThe analysis is repeated with every click and a communication type is calculated. The results are immediately displayed in this mask, so that the agents are shown arguments and suggestions for objection handling that match the course of the conversation.

The IAS Editor / Generator is used for the design of these masks, the required analytics and the control of the display (e.g. which button should be displayed or hidden in which situation with which effect?). Thus, adjustments, for example as a reaction to the feedback in the projects, can also be implemented on demand.


The data for all profiles are stored in a database that can be addressed via a project number and a profile ID. Here is an excerpt:


This profile data forms the basis for displaying the results for one or more profiles as tables, graphics or text/images.

The source of this data are personality profiles and text analyses. In principle, a transformation of existing profile data or typifications from other personality tests is possible – if the respective categories have been coordinated and the algorithms for the transfer have been defined beforehand. This then applies in both directions (from a personality test to IPM and vice versa).

Texted: Library

In a library that has grown over the years, thousands of text blocks with many pictorial representations are waiting to be called up by an API.

This library grows as new requirements are added and realized from different projects. Here is a small excerpt:


These blocks are called by “sections” which are addressed in the menus of the software solution or via search commands. For each of the menu items and icons shown in the next picture, the material is available in the library, which is compiled according to the result type and presented on the screen:


TEXTED: Editor

Here you can edit the descriptions or recommendations that are derived on a specific topic from the data of the profile or from the comparison of two profiles.

The profile data determine the contents of the recommendations, which are described in the editor in general as well as in relation to a motivation type or to dominant basic needs. In the example above, the texts “General” were combined with “Green” and “Yellow”. The algorithms for this are also stored in the editor.

Both the general descriptions and the detailed results can be linked with in-depth information, PDF reports for download or images. These links are also defined in the editor for the building blocks. This results in this structure:

Portrait version, e.g. employee
	Recommendation group, e.g. relationship to superior
		Section, e.g. Reporting
			General Text
				Optional Supplements (Links)
					inline image
					Type-related description (for 39 result types)
					Supplementary text(s)
					Supplementary image(s)
					Supplementary PDF document(s)
				Search strings for the recommendation search
			Recommendation modules for the 6 basic needs (optional) each with
			these contents
				Special texts (for enforcement, affiliation, etc.)
					Optional additions (links)
					Supplementary text(s)
					Supplementary image(s)
					Supplementary PDF document(s)

The supplementary information is maintained in a separate area of the library so that a link can be used by different building blocks.

 Additional text, e.g. "Keywords
			Images, e.g. visualization "Collaboration 
			Documents, e.g., "Understanding Buy Signals."
			Contents, e.g. "Tasks" (texts for 39 motivation types)

The API is then used to transfer the relevant content with these additions. On the side of the application, these are usually offered via icons or to the user.

In another section of Texted, the texts resulting from the comparison of two profiles are edited.

Comparison, e.g. applicant
	7 Comparison results:
		Equal/Low (there are no higher basic needs)
		The need for assertiveness is greater
		The need to belong is greater
		The need for security is greater
		The need for individuality is greater
		The need for knowledge is greater
		The need for empathy is greater
			For each of these comparison results:
				Descriptive text
					Supplementary text(s)
					Supplementary image(s)
					Supplementary PDF document(s)

Due to this structure, only supplementary competences resulting from the two comparison profiles are reported. If, for example, the applicant’s need for enforcement is significantly lower than expected in the job profile, the comparison “job profile” reports what results from this, i.e. should be clarified in an interview if necessary.

Customizing the recommendation texts

The modular and freely editable design of the Texted system allows customizing down to the last detail.

Integration of the results

A simple form of integration are links with which IPM functions can be called. This applies to an existing profile or a list of profiles. A separate browser window or a frame defined by the application opens.

The content described in the following, such as text modules or images, can also be retrieved from an interface (API, web service) and then displayed according to the user’s own design specifications.

Menus and recommendation search

In the API Editor the menus are defined, if the result representation is to be taken over by IPM. These can be defined together with the users and changed at any time. Here is an excerpt from the “Learning profile” menu:


The respective menu items therefore refer to the sections in Texted and are loaded via the API:


In addition, the user can search for recommendations in the library. This is controlled by keywords that have been defined for the texted sections. This example looks for recommedations to describe the USPs:


Compare two profiles

The library offers comparison modules for different profile types, for example customers, employees, superiors, jobs, learning, offers, partners. Here is a comparison of employees and supervisors:


The results of a comparison are formulated positively, because the question is how or with what one side can support the other. The links provide the user with in-depth information and visualizations in addition to the short notes.

Comparison of several profiles

The matrix function allows a quick comparison, for example when selecting employees or offers. To determine the positions, the dimensions “enforcement vs. security” (vertical) and “individuality vs. affiliation” (horizontal) are determined as X,Y difference values, which are passed to a series of IDs of the application.

In the following example, a product manager is to be selected for a new product “Security Software”. The “emotional closeness” of the candidate “BVQ” is immediately noticeable:

Inside-results(12)12.png Middle positions (deviation from both 0-points max. 20)
In this area there are profiles where the dynamics between enforcement and security, as well as individuality and integration are weakly pronounced. In the best case everything should always be considered equally.
Motivation (quadrant top right)
>The needs for enforcement and integration have a stronger effect here than the need for security and individuality. This results in an effort to motivate colleagues to participate.
Consultation (quadrant bottom right)
In this area, the focus is on “proper” consultation according to defined criteria. All organisational measures focus on their effect on those involved and affected.
Specialisation (quadrant bottom left)
The interaction of “security” (order, sustainability) and individualisation (special features, quality). Team members in this area are reliable and strive to avoid mistakes.
Development (quadrant top left)
Whoever equally strives for individuality and implementation likes to develop something new and special. These are the typical members of project teams in which they see a chance to realize themselves to a certain extent.

Team profiles

When considering the emotional relationships within teams, the focus is on individual characteristics as well as the resulting dynamics.

A “team personality” is first determined for the candidates from the above matrix (how does the team act as a unit on the members and on external reference persons?):


Further criteria are then examined in a PDF document generated for this purpose, for example on the topic “Search for findings” (research tasks, analytics):


Depending on requirements, both the data for the graphics and the result texts, such as the team description, can be transferred in the integration.


Application-specific IPM databases are usually divided into several project areas, for example departments, sales districts or task areas. The selection functions can be specified accordingly.


Alternatively, a standard profile can be used as a default. In the following example, a “customer consultant” is searched for and the database area “Applicants” is to be selected according to profiles that most closely match this pattern.

A typical use is the definition “Top 10” – in this case the restriction to the 10 most suitable candidates:


This function compares the profile data to the basic needs.

Red = Enforcement
Yellow = Integration
Green = Security
Blue = Individuality
Know. = Knowledge
Emp. = Empathy

and shows the percentage variance between the individual profiles and the target profile. If required, these detailed deviations can be clarified in the assessment. A “recommended value” is calculated from the sum of the individual deviations:


Interactive Advice System

To support telephone conversations with customers or applicants, the existing profile data is used to give situational communication recommendations that can be varied in the course of the conversation.

A complex editor is available for these applications. You can find more information on this topic in the “Sales” section of the “Customer Service Center” article.