Skills gap

This page helps administrators and managers identify skill gaps and take actions to reduce them.

Skills gap Home Page

This page displays the skill selection fields and user filters to create a new Skills gap analysis, if the user has not saved any analyses.

Otherwise the page displays the skill gap analyses recorded by the user and a "Generate a new analysis" button.

In case of recorded analysis, each analysis carries the following information:

• the user-customized name
• the number of skills analyzed
• the number of users in the analyzed area
• the date the analysis was modified (the creation date if it has never been modified)
• Two KPIs:
  1. Progress bar: This KPI represents the average of declared skills (or one of their versions) at the expected level (if there is one) by all active employees in the scope studied (filters applied), versus the total number of skills studied.
  2. Donut: The donut indicates the number of active collaborators in the scope of study who have declared at least one of the skills in the job description or one of its versions at the expected level (if any). The % is rounded up to the next whole number, if between 0% and 1% it is rounded up to the next tenth.

On each of the analyses it is possible to:

• Edit Analysis: Opens the analysis editing modal
• View Analysis: Opens the analysis report page
• Share analysis: the URL is copied to the user's clipboard
• Delete analysis

Create a new analysis

When clicking Generate a new analysis (or Edit an existing analysis) a modal window appears for the user. It allows them to configure the skills and filters for the analysis.

The editing modale of an existing analysis allows to set a specific name.

Skills selection

Via the search bar

The search bar allows you to select skills one by one via auto-complete :

• only heads embedded with the "in the statistics" attribute in the platform language are suggested (skills without the attribute in stats cannot be selected)
• if a version of a skill "in the statistics" is typed in full (exact match) then the version can be selected from the drop-down menu. However, it is the head that will then be used for the analysis.

If multiple versions of the same skill are entered, they will be listed as is and will all be replaced with their heads once the analysis is generated.

Once the first skill is selected in the drop-down menu, it is possible to select a second skill by typing a keyword in the same way in the search bar and so on.

Via tags, job descriptions, open positions or missions

A drop-down menu allows you to select skills with the "in statistics" attribute.

• with tags (if the filter is enabled in the user filter matrix for the user). A drop-down menu displays the tags that contain skills in the statistics.
• with job descriptions (visible if there is at least one active job description category opportunity in the platform). A drop-down menu with a search bar displays the job descriptions in the application (active or draft).
• with open positions (visible if there is at least one active open position category opportunity in the platform). A drop-down menu with a search bar displays open positions in the application (active or draft).
• with missions (visible if there is at least one active open PROJECT category opportunity in the platform). A drop-down menu with a search bar displays the active missions in the application (active or draft).

We display the different types of opportunity and the opportunities without taking into account the visibility rights setting of each dataset (job_descriptions_referential, mission_referential, open_positions_referential or trainings_referential).

There is no dynamic link between the skills of an opportunity and those of a Skills Gap analysis: changes to skills made subsequently to an opportunity will not be reflected in the Skills Gap analysis.

Only integrated, heads with the attribute in the statistics are added to the search. A skill version loaded in the skill gap search will be transformed by its head in the generated result.

Once the skill level limit is reached, it is no longer possible to add new skills.

Up to 50 skills can be selected, unless the platform limit setting is > 50.

If more than 20 skills are selected, then a disclaimer is displayed and warns the user that the radar and the skill declaration evolution graph will only display the first 20 skills in the search.

If the total number of skills selected exceeds the limit allowed by the configuration, then it is not possible to generate the analysis. The user is prompted to reduce the number of skills selected.

It is not possible to generate an analysis if no skills are selected.

Selection of minimum levels

It is possible to select a minimum expected level for each selected skill. To do this, and at the level of each selected skill, a chevron allows you to select a level on a gauge. The levels defined for the skills selected via the job sheet button are automatically returned.

The selected level can be reset by clicking on the rotating arrows icon.

If the same skill is selected multiple times with different levels, the highest level will be kept.

Indication of major skills

A major skill is indicated by a dot to the left of the skill name. It is indicated like this on all skill gap analysis pages.

Selection of a users

Only filter values corresponding to active users are available.

With user filters

The available filters are identical to the filters available on Skills Statistics

Note: the scope of selected users is determined when the analysis is created. This scope can change if user situations change later on, for example due to deactivation or leaving the scope.

By default, all selectable collaborators are included. Using the filters available in their organization — Family, domain, job, country, city, and more — the user can narrow down and target the population precisely.

A counter shows the total number of selected users and reminds the user which filter category or categories are in use.

Selecting users by file import

To use file import, click the Select by uploading a file tab.

The expected file format is an Excel document containing one column with the header "email".

A sample file is available for download directly from this interface to help format the file correctly. Then upload your Excel file containing the email column by drag and drop or with your OS file browser.

Canceling the selection: you can go back by deleting the previously uploaded file. This resets the selection and lets you switch back to user filters.
Conversely, removing all previously selected filters lets you return to the file import tab.

Once the file is uploaded, the system checks each email against the defined validation rules. Note: email comparison is case-insensitive, so uppercase and lowercase versions of the same letter are treated identically.

If errors are detected, the system displays a detailed list of issues line by line. Form validation is blocked as long as errors are detected in the imported file.

• Wrong file format
• Unknown email
• Duplicate email in the file
• Archived collaborator (archived)
• User is not selectable (search disabled)
• User is not selectable (login disabled)
• User with no associated job description – only job campaigns
• Collaborator with inactive job description – only job campaigns

If needed, delete the previously uploaded file and import a new one.

If several errors are detected on the same line, they are all listed. Multiple errors can therefore apply to one line.

Both methods — filters and file import — remain available, depending on your needs. They cannot be combined: choose one or the other for each analysis.

The "Generate an analysis" button launches the analysis and takes you to the report page.

Skills Gap Analysis Report

Analysis reminder

The analysis report has a banner at the top of the page which shows:

The information selected when creating the report

The number of skills applied and users affected by the analysis.

A chevron allows you to expand the details of the skills and user filters used. If no population filter has been selected, an All employees tab is displayed in the summary of the selected population.

In the accordion, the version and head skills are grouped under the head skill, and an explanatory message with a tooltip indicates that the skills have been grouped. The minimum level set is displayed on the skill chip.

A " Save the analysis " button allows you to save this analysis for later. When you click on it, a modal window allows you to give this analysis a name (limited to 70 characters), this name will be used on the home page to identify the campaign. This name can be changed later by clicking on " Edit this analysis ".

If the skills for the analysis were retrieved via a job description, a tag or an open position, the name of the opportunity or tag will be proposed by default (within the limit of 70 characters).

An " Edit this analysis " button reopens the creation modal and allows you to modify the selected skills or the chosen population perimeter. If the analysis has been saved, the name of the analysis can also be modified.

A " Share this analysis " button copies the page URL to the user's clipboard so they can share it with a collaborator. All campaign information is reusable, except for the campaign name.

When sharing, Skills Gap access rights are required to view the results. The results are conditional on the collaborator scope that applies to the user who opens the link. Additionally, the filters and skills selected by the user are stored, not the results of that campaign.

→ results may therefore vary from one consultation to another and from one user to another.

For each report two tabs are available to consult "Global Analysis" or "Individual Analysis"

Global analysis

KPI progress bar of average skill possession

This KPI represents the average of skills (or one of their versions) declared at the expected level (if there is one) by all active employees in the scope studied (filters applied), versus the total number of skills studied.

KPI donut percentage of owners

The donut indicates the number of active collaborators in the scope of study who have declared at least one of the skills in the job description or one of its versions at the expected level (if any). The % is rounded up to the next whole number, if between 0% and 1% it is rounded up to the next tenth.

Possession Radar by Skill

A radar is available when the number of selected skills is greater than 2. It presents skill by skill the rate and the number of employees who possess the skill at the minimum expected level.

The display of skills is limited to the first 20, beyond which a warning warns the user that the number of skills displayed is truncated.

Each branch of the radar corresponds to a skill of the analysis organized in alphabetical order in clockwise direction. If there is any major skill, they will be displayed in first position.

2 types of analysis:

Differences between expected and actual skills

The concentric circles correspond to the levels defined by the platform configuration. The center of the radar corresponds to level 0 (No declared level), the outer circle to the highest level.

There are 2 "spiders":

• A blue one that corresponds to the expected level
• A black one which corresponds to the average level of employees’ skills.

For each employee, the evaluations for all versions of a given skill are examined, and the highest score is retained. If a manager has validated a skill, the most recent validation score takes priority. The average of the best declared or validated scores from each employee is calculated, then rounded up to the closest whole number to obtain the final group score (for example: 3.2 becomes 3).

The "Include unrated levels" toggle changes how undeclared employee skill levels are considered:

• On: Undeclared levels are treated as the middle level. For example, level 2 when 4 levels exist.
• Off: Undeclared levels are excluded from the calculation.

The toggle is enabled by default.

When clicking on a point, both the average level and the expected level can be seen in the tooltip:

Below the radar, a legend explains how the radar works, with an explanation of the calculation method: if the toggle is activated to take into account undeclared levels, the message is " The average is calculated using the levels declared by your employees, putting the skills without an evaluation at "I'm practicing" level ", otherwise the message is " The average is calculated using the levels declared by your employees ".

Percentage of target skill statements

The concentric circles correspond to the % of employees who declared that they possessed the skill at the minimum expected level: the center of the radar corresponds to the 0% level, the expected target level corresponds to the outermost concentric circle (100%).

The current rate of reporting at a level higher than or equal to the minimum expected level of the skill or one of its versions is indicated in light color (= number of active employees who have reported the skill or one of its versions at a level higher than or equal to the minimum expected level / total active *100).

Largest Skills gap

To the right of the radar, a summary shows the least mastered skills, i.e. those with the lowest reporting rate. In the event of a tie between two or more skills, a sort by ID is done to suggest 5 skills.

The number of employees who declared the skill or one of its versions at the minimum expected level and the percentage in relation to the studied population are displayed.

Example in the picture, for a studied population of 17 employees

• nobody has declared the skill Customer service or one of it's version at the expected level: the number of users is 0, the gap is 100%;
• 5 employees has declared the skill "Proposal writing" or one of it's version at the expected level: the number of users is 5, the gap is 71% (12 employees on 17 did not declared the skill or one of it's version at the expected level);

Graph of the evolution of the declaration of skills

A graph displays the evolution over the last n days (x-axis) of the number of declarations of the skills selected in the analysis (y-axis) by the population studied.

A tooltip when hovering over the curve gives the number of declarations on the selected date and the change compared to the previous day.

It is possible to choose the study period:

• Custom: Choice of start date and end date
• 30D: The last 30 days
• 6M: The last 6 months,
• 12M: The last 12 months,
• All time: Since the beginning of the history on the platform.

The data displayed is calculated using the previous day's data: it is not possible to have the data for the day when displaying by day, for the current week when displaying by week, or for the current month when displaying by month.

KPIs levels of possession, expertise and evolution over time

A menu allows me to select one of the skills in the analysis in order to have more information on the level of expertise of each skill.

For each selected skill, a widget presents:

1. Number of employees who reported this skill at a level higher than or equal to the minimum expected level (902) compared to the total number of employees who reported the skill (2425)
2. Number of employees who reported this skill at the expected level (902) and the percentage compared to the population of the entire platform (17%)
3. Distribution of the levels of employees who have declared the skill or one of its versions ( donut identical to that on the Skills Statistics page ).
   If multiple versions have been declared by an employee at different levels, the maximum level is counted. If the skills have been validated by a manager, the validated level is taken into account.\
   Each employee data is only counted once.

The cumulative graph of the evolution of the declaration over the last n days of the skill distributed by level of expertise.

If the skill has a level description, hovering the level legend will display it.

Individual report

The individual report tab provides access to the ownership matrix.

The matrix displays all profiles that are visible to the user. The number of private profiles, whose data is not displayed, is indicated.

Columns:

1. Collaborators: Displays the avatar + name + job attribute of each employee
2. Skills : One column per skill studied (displayed in alphabetical order). Next to the skill, a dot indicates the overall skill possession rate.
3. For each employee,
   1. Number and percentage of skills declared at a level greater than or equal to the minimum expected level
   2. if the skill is declared at a level higher than or equal to the minimum expected level: the level declared for the skill is indicated online using a gauge with 1 segment per level as well as a green check.
   3. if the skill is declared or validated at a level lower than the minimum expected level: an orange exclamation mark is displayed
   4. if the skill is not declared: a red cross is displayed.
4. Individual actions:
   1. An action button on each row of the table allows you to contact the employee or their manager using a mailto. If the manager is not visible to the user, the button does not exist in the drop-down menu.

If the user has the rights to give a feedback to the employee, it is possible for him to assess the employee's skill in the skill drop down details.

Exporting the matrix

An export allows you to retrieve all the information from the matrix. The visible data on employees depends on the filters selected in the filter matrix.

The values ​​of declared, validated and expected levels for each skill are also returned.

Learning Tab

Once the skill is selected, the list of training courses relating to the selected skill (maximum 100) is displayed, in all languages, sorted in order of creation from the most recent to the least recent.

AI identifies the most relevant training courses based on:

1. if the skill is directly associated with the opportunity
2. or if the name of the skill is cited in the title or description of the training.

When clicking on a card, the "Staffer + Trainings Skills stats" modal is displayed.

If no results are to be displayed, an empty report is presented with, if it has been configured (see configuration), a link to the client's LMS platform(s) can be configured.

Two factors can create, in a minimal way, data gaps between the indicators of top skills, skill stats, skills gap and find an expert:

The indicators are not updated at the same time

On Top skills:

• The skills, expertise and preferences data are updated every night

• Search data is updated in real time

• The number of employees in the scope is updated in real time

On Skill stats and on Skills gap

• Data is updated in real time

• The evolution of the level of expertise is updated every night

On Find expert

• Data is updated in real time

Why not a real-time update everywhere?

Our goal across all app pages is to have sufficient performance to be able to use the page without irritation. On a page like Top Skills, it's impossible to continuously calculate all the skills/user data needed to populate the various indicators because it represents a very large volume of data, and it would make the page unusable.

The tool used on Skill stats may generate approximations

Skill stats data is managed by an independent 365Talents technology called Elasticsearch.

This technology retrieves user/skill data from several "machines" and compiles them to calculate a result which will be the skill to be displayed and its number of declarations/expertise etc.

In case of data inhomogeneity, Elasticsearch may generate a result that will be slightly out of sync with the result on Top skills.

This is something common, visible on other everyday tools: the Twitter like counter, the number of comments on a Facebook post, etc.

→ These shifts remain rare, and when they happen the shift is minimal, but you have to keep it in mind so as not to be surprised.