Building an understanding of your learners allows you to design training experiences that more deeply resonate with learners, address their needs, and ultimately create more impact. We’ve spoken and written a number of times on just how powerful doing an audience analysis and utilizing tools like Learner Personas during your design process can be, but we’ve seen many organizations start and abandon the process because of the time and resources it can take to thoughtfully collect and analyze all of this data.
Luckily, advances in AI have made data analysis faster and easier than ever, meaning that we can quickly gain insights into learner data in hours or minutes rather than days or weeks. This is great news (especially for those of us who loathe Excel formulas) because these insights can help us better tailor our training to our learners unique needs, measure effectiveness, and quickly adjust and pivot our programs based off of our data.
Cost and Time Savings of Using AI
L&D teams are constantly being pushed on to develop and roll out “the next big thing” leaving little time for outcome analysis or research time to help prep for your next initiative. Taking time to understand their learner data may seem costly but ignoring it all together risks an even bigger expense in misused resources and ineffective training programs.
New Machine Learning and AI tools have proven to be adept at analyzing large data sets, coding and categorizing responses, and accurately summarizing trends. What once might have taken someone hours of time sifting through data sheets is often only a well-constructed text prompt away. By using these tools, time can be shifted from analyzing and interpreting the data to deciding how to implement those insights to build better learning experiences.
How Do I Start Collecting Learner Data?
You might be surprised by how much data you already have available to you. If you are using a Learning Management System, you likely have a wealth of learner data including test scores, completion rates, system usage, and more.
In addition, you may have other systems or data sources that you can leverage alongside your learning data such as:
- Sales or call systems
- Safety and compliance issue tracking
- Customer experience surveys
- Operational Performance Metrics or systems (Retail scans per minute, fulfillment rates, etc.)
If you aren’t already collecting learner feedback, consider the following ways to learn more about your learners’ training and job experiences:
- Conducting learner interviews
- Holding focus groups
- Shadowing learners on the job
- Sending out surveys before, during, and after key training programs or periods (like Onboarding)
How to Use AI to Analyze Data
Now that you’ve got your data, there is a little bit of formatting that you’ll want to do to make sure your data is clean and ready to be used by your preferred AI tool.
Review your data for the following:
- Make sure column headers are descriptive and clear
- Remove duplicate data
- Remove any private or sensitive data and be sure to remove any data that clearly identifies your learners such as names. Consider replacing name with a random ID to help with indexing responses.
When your data is formatted, you can upload your files. When doing so, include a description of what the dataset includes and any other context. This will help your tool provide recommendations for analysis prompts.
Example: This file contains feedback collected from learners after their 30-day onboarding program.
Now you can prompt with specific questions or requests to analyze trends between responses and find correlations between datasets. For example:
Analyze and summarize any trends between responses for the following questions:
- Q1: How would you rate your onboarding experience? (Very Good, Good, Neutral, Somewhat Poor, Very Poor)
- Q2: How involved was your manager in your onboarding? (Highly involved, Mostly involved, Neutral, Somewhat involved, Not involved).
How does learner performance compare for learners who completed ‘X’ training versus those who did not?
Categorize learners and create personas based off of attitude towards learning and their top training needs.
To learn more about how to construct prompts for AI tools, check out this article.
Limitations of AI
While AI is a great tool, it does not replace insights from your own research. It can lack the ability to correctly understand emotional or situational nuances in survey data, can create personas based off of limited or generalized data and prompts, and relying too heavily on AI for insights just can lead to misinterpretations of your audience data. Also, using incomplete data sets can result in AI “filling in the gaps” with assumptions or public data that may not be relevant to your users.
It always important to spend some time looking at the data yourself and investigate when you get a result that seems incongruent with your own understanding of your learners. Prompt tools to explain how they reached certain conclusions and when in doubt, double check yourself.
Using AI for data analysis promises to help organizations quickly conduct audience analysis and build learner personas, while saving on time and resources. Like any AI application, it is not a total replacement for human interpretation and overreliance can lead to an incomplete representation of your audience. However, viewed as a powerful starting point, AI can help us learn more about the learner data that we’re collecting and open up new questions and perspectives that help us get closer with our learners and develop more impactful training experiences.
Ready to start looking at what your learner data is telling you? Need some help conducting a needs analysis or crafting your training strategy? We’d love to help!
