How much text do you generate every day?
Think about that for a minute, then think about that multiplied across every employee in your organisation. The texts, the emails, the employee surveys, the social media posts…
Skills in text analytics are a big deal moving forward for securing top jobs within HR. It has always been a serious field, but increasingly, there is a requirement to back decision-making with strong ties to data.
Text analysis can make up a significant part of that data. Here’s why it’s important to step up those skills:
Why are text analytics important?
Text or natural language is an omnipresent part of human life. We communicate our deepest thoughts via natural language and these can translate into considerable insights for HR analysts.
Text analytics is the application of algorithms to process text information. In the last few years, we have seen an explosion in “unstructured” data, which includes things like text, images, audio and video. This data is unlike the classic view of data – numbers contained within databases.
This type of analysis is important to HR because it can lead to more detailed insights than numbers alone will tell. NLP (Natural Language Processing) analysis provides information that is not just limited to the business, but gives clues to human intent and disposition. This is a boon for HR because realistically, the thoughts and attitudes of team members play a huge role in overall performance.
What can we learn from text analytics?
The CEB Talent Analytics Quarterly notes three core applications of text analytics that will help HR practitioners transform how they operate. These include:
- “Counting words and phrase frequency. By counting words and phrase frequency, and combining these with statistical analysis, it’s possible to identify trends and themes. This type of analysis can help HR leaders determine what employees care most about.
- Using natural language capabilities to understand themes. As semantic language becomes more sophisticated, it’s possible to analyze text in a way that helps interpret meaning. This approach to text analytics can be used to look for patterns of complaints in open-ended questions such as the need for more vacation time or concerns around policy changes.
- Identifying features. These types of analysis use large data sets to look for common features within. For example, you might take two batches of resumes of individuals that your business hired and individuals that your business did not hire. From there, an analysis could uncover what similarities occurred in each group to further refine your recruiting strategy.”
HR might use these applications across areas like recruitment, survey or feedback analysis, succession planning, appraisal and social media analysis. Text data represents some rich, often untapped opportunities.
NLP can help with enabling greater accuracy in data analysis and quicker improvements of HR business processes. It also helps to reduce human bias in those important decisions as they can be made based on significant data.
What are the challenges of text analytics?
There are some challenges and limitations when it comes to text analytics. For example, running algorithms and automating tasks such as sentiment analysis might sound simple enough, but as we all know, human language can be very nuanced.
An algorithm may not pick up tone, for example, especially if someone is being sarcastic. This means that while NLP is a useful tool, it doesn’t entirely replace the human operator who can break down language complexities.
Context plays a major role in good analysis, and this is something that is difficult to teach a machine. The fact that organisations tend to develop their own languages can also be tricky. For example, they may have their own jargon or colloquialisms that will be lost on an algorithm. As machine learning gets more advanced and humans get better at analysis though, there is room for much more potential with text analytics.
This segues nicely into the next challenge – the human operator! Adoption of text analytics is definitely slowed by a lack of practitioners with the skills to properly access and analyse the data. Cleaning and processing data can be a huge task and it requires a practiced eye. For example, if you were to use text analytics as a recruitment tool, you would need to be able to zero in on very precise sections of a CV.
Finally, there are some challenges with the adoption of text analysis within HR in organisations. Some remain skeptical of its utility to their company, while others are concerned with potential data protection and compliance issues. Sometimes there is a lack of training and support for the analytics function within HR and sometimes, HR practitioners themselves discourage use of algorithms for fear of their job being taken by a computer. (A fear that is relatively unfounded given our discussion on the need for a human operator to interpret context and language nuances).
One thing that is certain though is that use of text analytics for HR is increasing rapidly. HR practitioners would do well to step up their skills in this area – doing so may even be a competitive advantage.
Is there a skills gap in HR?
When digital competencies such as data analysis skills are assessed, HR has been found to lag behind other departments. Harvard Business Review reports: “most have struggled to advance their own digital competencies. This neglect has hindered their ability to leverage data into talent strategies that can help transform their businesses.”
One of the key findings in a survey of nearly 28,000 business leaders across different industries and trajectories was: “On average, HR leaders lag far behind other professionals in their ability to operate in a highly digital environment and use data to guide business decisions.”
These differences have spurred a credibility gap between HR professionals and their more digitally-versed colleagues. This has meant that executives don’t necessarily trust or involve HR with making strong decisions for filling talent needs.
Developing strong data analysis and visualisation skills are an important step to close the gap. On top of these skills, HR practitioners need to be able to present the data well, such as in graphical format. According to the HBR report: “HR needs to get more proficient with sophisticated software programs such as Power BI, Tableau, or R Studio, all of which give visual context to data.”
How can HR practitioners step up their skills?
One of my core beliefs is that the best way to get better is to seek out learning opportunities and to practice. You can work with practice datasets and learn to clean and extract the information that you need. Entering competitions is another great way to practice, learn and improve.
HR practitioners should try out text analytics for themselves using, for example, Julia Silge’s online book.
It also helps to learn about the powerful tools that are available to help perform analysis of HR texts, for example:
https://monkeylearn.com/sentiment-analysis/
https://azure.microsoft.com/en-gb/services/cognitive-services/text-analytics/
Another thing to consider is that it is an advantage to have strong HR skills and knowledge, paired with an aptitude for HR analytics. Many data scientists may say they can perform text analysis in HR, but they probably don’t have the skills or knowledge to mine to the data that is really useful in an HR context. Experts that are versed in specific HR analytics are in high demand because they are more rare – you need to have that core understanding of what makes data useful for HR.
Final thoughts
Text analytics skills are a major skill to develop going forward for any HR analytics professional. Mining text and other unstructured data can help to build a stronger picture for decision-making than traditional, numerical data alone.
To date, HR practitioners have been somewhat behind their colleagues when it comes to digital and analytical skills. HR has much to contribute through these skills, proving that they can make accurate, data-based decisions.
The key for HR practitioners is to become better-versed in data analysis, including text analytics. HR analytics skills will be a competitive advantage for employment in the discipline moving forward as those people help advance the activities of HR in the organisation.