Data As A Language: Impact For Learning And Development

What If Data Was A Language?

L&D uses new influence to elevate people and their skills for business impact.
– from LinkedIn’s 2023 Workplace Learning Report

Based on the 2023 report, business impact is the number one target for Learning and Development (L&D). In fact, the top priority of the industry is reportedly “mapping learning to business goals” this year. This is good news! Finally, we’re measuring what matters.

Not so fast!

While aligning learning programs to business strategies is L&D’s No. 1 goal this year, the metrics don’t line up. The top five ways L&D pros are measuring success are vanity metrics, based on satisfaction with programs. Business metrics fall to the middle or bottom of the list.
– from the same report

Let me summarize this: our goals are not aligned with how we measure our success. Hmm… could one of the reasons be because we need to improve data literacy? (The report does list data literacy as one of the action items.)

If you’re not measuring the right things today, jumping to a complex ROI study of your programs that proves causation may seem daunting! It’s like learning a new language to write a bestseller novel before you can engage in basic conversations.

Start Small, Then Iterate, Iterate, And Iterate…

This article suggests starting your data literacy a different way. What if you treated data as the language of impact? What if you needed to learn the language well enough (without the pressure of being perfect) to start showing the value, to start telling the story of impact? And then iterate, iterate, and iterate?

What are some of the main concepts of a language? This is not an exhaustive list but it can give you some food for data thought:

  1. Alphabet/characters (data points)
    In a spoken language, these are the building blocks of words and sentences. In the language of data, these are the individual data points or units, such as numbers, letters, or symbols. While they are important building blocks, they do not have a meaning without context.
  2. Words (variables)
    Words in a language are made up of characters, and in data, variables are made up of data points. In the context of data, words are represented by variables like “sales revenue,” “employee performance,” or “customer satisfaction.” Words can be complex, using compound words, just the way you can use simple variables to calculate a new complex one.
  3. Grammar (data classification rules)
    Grammar rules dictate how words and sentences are organized in a language. In data, basic classification rules dictate what you can or cannot do with different types of data. For example, temperature is a quantitative, continuous-interval type of data. You can do additions and subtractions on them (10F+20F = 30F) but you can’t do division on them (40F/20F = 2).
  4. Sentences (data records)
    Sentences in a language are composed of words organized in a specific order to convey meaning. In data, you can interpret them as statements that could be in a narrative format, visual, or both.
  5. Syntax (data formatting)
    Syntax refers to the rules governing the arrangement of words and symbols in a language to create meaningful sentences. In data, syntax corresponds to data formatting rules, ensuring that data points are correctly formatted and structured. For example, date formats, decimal places, and data type conventions are part of data syntax.
  6. Semantics (data meaning)
    Semantics in language deals with the meaning of words and how they convey concepts. In data, semantics refers to the meaning and interpretation of variables and data points. For instance, understanding that a “high customer satisfaction score” means positive feedback is a semantic interpretation of data.
    This is important! Data is not reality or truth by itself. Who collected data, how, when, how often, etc., can already include bias and a skewed view of reality. That is why context and semantics must be used together! For example, if customer satisfaction is measured through 12 questions after each contact, focusing on the questions can improve the score more effectively than generic training on active listening, empathy, etc.
  7. Context (data context)
    Context in language helps determine the meaning of words or sentences in a specific situation. In data, context is crucial for understanding the relevance and significance of data points. For instance, knowing that a sales figure is for a specific time period and product category provides context. Understanding that a fiscal year may start in January and end in December for one company while schools operate on a September to August period may be crucial.
  8. Dialects (data domains)
    Dialects in language represent regional or cultural variations. In data, dialects correspond to different data domains or industries. Each data domain may have its own terminology and conventions. For example, healthcare data has specific terminology and standards distinct from financial data.
  9. Narrative (data storytelling)
    Narratives in language are stories or accounts that convey information or entertainment. In data, narratives are data stories that transform raw data into compelling insights, making it easier for people to understand and act upon. For example, presenting data on customer churn as a narrative that highlights the impact on revenue and suggests retention strategies.
  10. Fluency (data literacy, data fluency)
    Fluency in a language implies the ability to speak and understand it proficiently. In the language of data, fluency equates to the scale of data literacy, which is the ability to read, write, and comprehend data effectively for decision-making. Data literacy may cover the fundamentals, while data fluency may include sophisticated statistical methodologies and complex data modeling.

If You Speak Data, The Language Of Impact, What Do You Actually Say?

The Language Of Description

To understand the concept of data as a language, again, we must first acknowledge that data is not reality, much like a language is not reality. Data is merely a tool, a set of symbols and representations we use to describe our world. Just as words construct sentences and paragraphs to convey meaning, data points come together to form patterns and insights.

Think of it this way: when you look at a chart displaying quarterly sales figures, what you’re seeing is not the actual sales transactions themselves, but rather a linguistic representation of those transactions. It’s like reading a novel that describes an epic battle, rather than witnessing the battle firsthand. When you tell others what happened (visually or verbally) through data, you use descriptive language. It is common for L&D dashboards to share some of the metrics such as completion rate, level 1 satisfaction, or average score.

The challenge of using descriptive language only is that what we say may not be actionable. While we say what happened, we don’t know why it happened (diagnostic language), and we can’t tell if it is going to happen again (predictive language), let alone provide insights for stakeholders on how to prevent this from happening again (prescriptive language).

Insights: The True Currency Of Data

In the world of L&D, as in business at large, it’s not the data itself that drives decisions; it’s the insights that can be gleaned from that data. Just as you need to understand the nuances of a language to appreciate a work of literature fully, you must possess data literacy to extract valuable insights. Much like fluency in a language opens doors to new cultures and experiences, data literacy unlocks the doors to better decision-making and innovation in the workplace through insights. And just like languages, the only way to learn to speak is through speaking.

Consider this scenario: a learning professional is tasked with evaluating the effectiveness of a new training program. Without data literacy, they may believe that the raw change between the pre- and post-assessment is due to the effectiveness of the program. And when performance numbers go up at the same time, they may conclude that that training program caused the change in performance. While this statement may be true (most likely not), data literacy empowers them to understand how to design the measurement and evaluation, how to test if the change is significant, how to tie the results to the performance, and more! Data literacy is the ability to connect the dots between data points, creating a narrative that informs actionable decisions.

For L&D professionals, the goal of data literacy is not just to be correct. It is to make a difference. It’s not just about what you say, it’s also about how you say it. And that is the power of data storytelling.

The Power Of Storytelling

Here’s the twist: data, much like any language, is not enough on its own to change minds or drive action. It needs a story—a compelling, human narrative that resonates with the audience. If data is the language, then storytelling is the art of persuasion. Data-driven storytelling is the secret sauce of effective L&D professionals. It transforms raw data into relatable experiences, making the numbers come alive.

Important: know your audience! Data storytelling can be as simple as a single shocking fact (as a hook), the context of what it means, and recommended actions with all the “look how I got this thing” in the Appendix. Why? Because the value of data insight is relative. A story that a CEO finds valuable is very different from what your SME or your learning design team are interested in. The value of data is relative.

Data’s Relative Value

In the realm of data as a language, it’s crucial to recognize that the value of data (and the insights gained from it) is relative. For instance, a finance executive may prioritize data on cost savings resulting from a training program, while a frontline manager may value data on improved employee performance. The same data can tell multiple stories depending on who is listening. Data literacy not only equips you to tell these stories but also helps you tailor them to resonate with your audience.

Learning professionals in the workplace need to understand that the language of data isn’t one-size-fits-all. It’s about speaking the language that matters most to your audience and translating data into actionable insights that align with their priorities.
– from LinkedIn’s 2023 Workplace Learning Report

Learning Data Literacy: How to Learn a New Language?

There’s always Duolingo for a language. Do we have Datalingo (?) for data??

So, how does one embark on the journey of learning data literacy, treating data as a language of impact? Well, it’s similar to the process of learning a new language, but instead of conjugating verbs, mastering grammar rules, and reading between the lines, you’ll be deciphering the meaning behind charts and understanding statistical concepts.

Start With A Sentence, Not A Novel!

And that leads me back to LinkedIn’s original 2023 Workplace Learning Report. If you treat data as a new language you practice, you see Key Performance Indicators and stakeholders in a different light. Instead of starting out with a complex ROI report, you can explore how insights gained from your data could be valuable for various stakeholders. Gain practice where you’re comfortable, then iterate from there.

For example, instead of a long debate about whether employees would find it more practical to have a downloadable PDF or an animated explainer, you can just run a pilot and use data as A/B testing. Or, let’s say you replace your generic level 1 question with specific confidence, intent to apply, priority, and expected performance support questions and now you can share actionable insights with your stakeholders. Or, maybe business operations wants to know how long it will take to complete a training, so they can schedule accordingly. Without data, you would probably just quote the estimated average “seat time.” But if the aggregated minutes across thousands of agents can cost the business a lot, you may want to do a pilot with a randomized audience so you can share with confidence that 95% of the whole agent population will be within the range of X from the average.

Once you’re confident in your data literacy skills within learning, then you can extend it to the transfer of learning (behavior on the job), and the effect of that transfer, the measurable impact. Just like in any language, one can talk a lot without saying much. It’s not the volume of data you’re after, it’s the perceived value of your insights gained from the data. Why perceived? Because sometimes you need to persuade your stakeholders that your insights would be valuable for them for data-informed decision-making, but in return, they need to work with you on collecting data.

How To Improve Data Literacy Skills?

By reading, writing, understanding, and persuading with data. In other words, by practicing speaking the language.

1. Vocabulary And Grammar

Just as you start learning a new language with its vocabulary and grammar, data literacy begins with the basics. Understand the terminology, concepts, and tools that make up the data landscape. Familiarize yourself with terms like “data points,” “variables,” “correlation,” and “data visualization.”

2. Reading And Listening

Learning a language involves reading books and listening to native speakers. In data literacy, you need to consume data in various formats—charts, graphs, reports, and datasets. Try to decipher what the data is saying, identify patterns, and draw initial insights.

3. Speaking And Writing

Practicing a language involves speaking and writing. In the data world, this means working with data, creating your visualizations, and articulating your findings. It’s like crafting sentences in a new language.

4. Cultural Immersion

Immersing yourself in a language’s culture helps you understand its nuances better. In data literacy, this translates to understanding the context in which data is generated and the culture of your organization. What are the specific challenges and goals that data can address?

5. Continuous Practice

Learning a language is an ongoing process. Similarly, data literacy is a journey of continuous learning. Stay updated with the latest tools and techniques in data analysis and visualization.

What If You Make A Mistake?

They’re going to put you on a corporate stage as a laughing stock and then they’ll make you do the walk of data shame while playing “Walk Like an Egyptian.”

Seriously, who doesn’t make mistakes when learning a new language? Throughout this journey, you may encounter “mistakes” in your analysis. Perhaps you initially focused on the wrong metrics or misinterpreted a data point. Maybe you relied on unreliable data or missed some critical assumptions. These mistakes serve as valuable learning experiences, helping you refine your data literacy skills. You don’t need to know everything to speak. It’s okay not to know something.

It’s okay not to know something. What’s not okay is not to evolve from there.


In the realm of Learning and Development, data is not just a collection of numbers; it’s a language of impact. Data literacy, like learning a new language, empowers L&D professionals to read, write, understand, and speak the language of data well enough to make a difference. It’s the bridge that transforms raw data into actionable insights, allowing organizations to make data-informed decisions that drive positive change.

Make sure you measure the right things! Focusing only on vanity metrics can backfire. Why? Because vanity mirrors may block your long-term vision of the road.

Just as every language has its own nuances and dialects, data’s value is relative and context-dependent. Tailor your data narratives to resonate with different stakeholders, just as a traveler adapts their language skills to connect with people from diverse cultures.

So, embrace data as a language, for within its numbers and patterns lies the power to transform, inspire, and create impact in the world of L&D. Remember, data, like any language, has the potential to tell stories that change the world (okay, let’s start with your workplace), one insight at a time.


Originally published at

Source link