GenAI for People Analytics
Real life examples.
Welcome to FullStack HR, and an extra welcome to the 89 people who have signed up since last week.
If you haven’t yet subscribed, join the 6300+ smart, curious HR folks by subscribing here:
Today’s FullStack HR is presented by… Amby!
Amby is the go-to recruitment partner for ambitious teams.
With their embedded recruitment service, teams of all sizes can plug in skilled recruiters to ramp up hiring with quality, pace, and flexibility. Their embedded service is full-stack, so whether you need them to manage dozens of recruitment projects from A to Z or build up your hiring playbook from scratch, they’ve got you covered.
No matter your team size, hiring challenges, or industry, Amby helps you find the talent you need with a recruitment model designed for scale.
See why talent teams at Kahoot!, Whereby, Superside, reMarkable, and more use Amby to scale their organizations: Learn more here.
Happy Friday, folks,
The holidays are lurking around the corner and that’s starting to feel!
Short reminder about the webinar on Thursday, and then we dive right into it today.
Last year, when the hype around AI began to swell, many pointed to People Analytics as an area that would be significantly impacted.
However, perhaps I've been looking in the wrong places, but I haven’t seen much about People Analytics yet.
Sure, in my collection of 307 prompts for CHROs, a couple touch on People Analytics.
And yes, several articles allude to potential uses, but there's a lack of real discussion about actual, concrete use cases that detail “this is what people are doing.”
Like many articles about Generative AI, there’s a lot of hope and aspirations (and one or two consultants predicting potential outcomes) but not many substantial use cases. Let’s try to change that.
This won't be a prompt bonanza but rather a summary of what I’ve heard through the grapevine. This includes insights shared by people I've talked to, who have given me permission to share their experiences, with the promise of anonymity for various reasons.
But before we delve into this, let's consider security and sensitivity.
People data can be sensitive, which is probably why we haven’t seen many articles outlining specific applications. Always sense-check yourself before using tools like this.
Talk to Your Employee Engagement Data
The first company is one who had been collecting employee feedback through an engagement tool for years. However, the HR team and managers struggled to derive actionable insights from the data collected.
(Any one relating to this? lol)
Despite having rich data on employee sentiments and engagement, they find it hard and cumbersome to analyze and interpret it the data manually, and the tool they are using isn’t helping either.
Their solution has been creating a custom ChatGPT version specifically designed to interface with their employee engagement tool's database. This CustomGPT is programmed to process, analyze, and interpret the data, identifying key trends and areas of concern.
It’s designed to understand the nuances of employee feedback and categorize it into themes such as job satisfaction, leadership effectiveness, and workplace culture.
Note here tat it only took a couple of hours to get this up and running. And then the custom GPT rapidly processed the accumulated feedback data. It identified a trend of dissatisfaction related to remote working arrangements and a desire for more frequent team interactions, which the company is now addressing. They had a hunch about this, but now they have clear data to back it up.
HR continues to use the custom GPT to monitor employee sentiment, making it easier to identify and address issues proactively, by talking to the data.
It’s still early days, and they've just implemented this, but their hypothesis is that this will lead to improvements in employee satisfaction and engagement.
The custom GPT has already become an invaluable tool for the HR team, enabling them to make data-driven decisions - for real.
Skills Gap < ChatGPT
The second example is perhaps one of my favorite examples of using ChatGPT for a real, concrete HR problem. This software development company aimed to address skill gaps due to the need to constantly upskill their employees.
The company first used ChatGPT to create a comprehensive baseline of the necessary skills for the future. This involved feeding the AI data on current industry standards, emerging technologies, and future market trends.
They also double-checked this with ChatGPT throughout the process, asking ChatGPT if this was inline with what ChatGPT deemed to be skills and knowledge that would be needed in the future.
ChatGPT processed this information and generated a detailed list of essential future skills, such as advanced programming languages, AI and machine learning expertise, and cloud computing proficiency.
Creating Evaluation Forms
The next step involved designing evaluation forms to assess employee skills against the established baseline.
According to the person I spoke with, ChatGPT was instrumental in this process, helping create forms that included relevant skill metrics, proficiency levels, and future-oriented competencies.
These forms were used by HR and managers to evaluate employees, focusing on identifying areas where current skills did not align with the future skills baseline.
Once the evaluations were completed, the results were inputted back into ChatGPT. ChatGPT then analyzed these results, comparing individual employee skills with the future skills baseline.
ChatGPT identified specific skill gaps at both the individual and departmental levels, providing a clear picture of where the company's workforce stood in relation to future skill requirements.
It’s worth highlighting that, according to the HRBP, this was done blazingly fast. He mentioned they could have done this manually, but ChatGPT completed this, including the prompting, in less than 30 minutes.
Generating Tailored Training Recommendations
Did they stop there? Oh no, they didn’t!
Based on the analysis, ChatGPT then generated personalized training and development recommendations. These recommendations were tailored to address the specific skill gaps of each employee or team.
The AI also suggested potential external training programs, online courses, and in-house workshops that could effectively bridge these gaps.
And that’s where they stand today, on the verge of starting to bridge the skill gaps they have now identified.
Speed
These two examples are not unique nor impossible for a competent HRBP to solve. But they require one thing we HR professionals often lack in abundance: time.
In the first instance, we’re talking about less than one day from start to finish.
The latter is slower, but the slow part here involves the manager and HR assessing and talking to people (which, in my opinion, is how it should be).
But the creation part?
We are talking about an estimated one day's work for one person to create this full plan and handle all the data connected to it in a company with several hundred employees.
I don’t know what People Analytics challenges you are facing, but I hope these two examples serve as inspiration for what you can do within the People Analytics field.
If you have other examples of how to use GenAI and People Analytics, please reach out!
Hopefully, we can have non-anonymous examples here as well. :)