One Model Embedded Insights (1+1=3) - Photo by George Becker from Pexels

Product Update - Embedded Insights

As you saw in the last One Model product update we are really focused on delivering a compelling capability for Storytelling with Data. Our storyboarding capability is designed to harness the power of One Model’s amazing Data Orchestration and OneAI platforms to unify your data and provide predictive insights to drive better decisions and business outcomes.

Key storyboarding capabilities we covered in the last release included new data visualizations, new data visualization options/control, filter search, geo-mapping and the ability to Export to PowerPoint. 

In solving people analytics challenges one of the most useful tools at our disposal is correlations.  Understanding if two variables are significantly related is critical to drive understanding and well designed interventions. However, something I rarely hear is someone say: "I got into HR because I really love statistics".

pexels-george-becker-374918-Jun-03-2021-06-18-34-90-AM

An important part of our Storyboard vision is to democratize AI and provide statistical decision support across the entire One Model platform. In the last update we shared details of our real-time forecasting within Storyboards and building on this, we have been working on delivering a wide range of statistical services in the hands of all users.

With One Model you have long had the ability to create scatterplots to visually understand the relationship between two variables and identify notable clusters or trends. Within a scatterplot you now have the ability to run a Correlation or show a Line of Best Fit to determine the statistical relationship between two metrics to answer questions such as “What is the relationship between employee engagement and retention?”

Taking this to the next level, we have created an explanatory layer that interprets the statistics of the correlation and provides a simple explanation for anyone to understand the results.  The explanation appears as text sitting above the chart and describes whether or not there is a correlation, to what degree, as well as the likely significance.  Additionally, if you click the info icon you will be able to get more details about the actual correlation and significance values, the formula and definition of the metrics in play and details of how to understand correlations and significance. There is also a link to an even more comprehensive guide in our help portal. This is a game changer for HR, managers, or anyone using data to make decisions.

For the Data Scientists and Statisticians out there, OneAI uses an automatic heuristic to determine which correlation metric to display to the user. This heuristic chooses between the Pearson Correlation Coefficient, the Spearman Ranked Coefficient, and the Cramer’s V Statistic. If both incoming datasets are numeric, and the relationship between those datasets is linear, then the Pearson Correlation Coefficient will be displayed to the user. If the two incoming datasets are numeric but not linearly related, then the Spearman Ranked Coefficient is used. If the two datasets are categorical, a Cramer’s V Statistic is displayed. We have also added an option to specifically pick what correlation will be used for a given visualization if you want to override the automatic heuristics from One Model.

These are just a couple of examples of how One Model is taking Storytelling with Data to the next level with embedding insights.  We have a lot more capability like this coming in our product roadmap and we can’t wait to get it into your hands.

Here’s a quick video showcasing the latest Embedded Insights within One Model.

Embedded Insights - Tony Ashton

 

One Model provides a people analytics platform that pulls together all your workforce data into one place for faster, better, evidence-based workforce decisions through flexible storyboards and advanced machine learning.

Asset 1

Subscribe to
Our Newsletter

Get the latest insides,
news and updates delivered straight to your inbox.

PREVIOUS ARTICLE NEXT ARTICLE