The Race to the Top: Why learning velocity is an indicator of business and career success

When generative AI models like Midjourney’s image generator were unveiled, my immediate reaction was that Adobe’s Creative Cloud suite had lost its defensible moat. But I was wrong. After a period of uncertainty, Adobe has rebounded – even launching a new suite of AI features for Photoshop. What allowed Adobe to stay relevant is their "learning velocity" – the rate at which they can acquire and apply new skills. 

Their high learning velocity did not happen overnight: it resulted from deliberate upskilling and product efforts in the field of AI for several years with releases such as Neural Filters, Sky Replacement, Object Aware Refine Edge, or Refine Hair – predating the widespread excitement surrounding Generative AI.

Many enterprises are trying to crack the learning velocity puzzle to keep innovating and staying ahead of the curve. Some will be left behind or disrupted. But a new type of competitive advantage is emerging – let me tell you about it.

Learning velocity is the new competitive advantage

The risk of disruption for a business increases with the pace of technological advancements. The best defense for businesses is their ability to rapidly leverage the latest innovations – which is why the speed at which employees learn is becoming a business imperative.

Industries where speed and time to market matter most, and where the field is constantly evolving, typically have higher learning velocity. A real-world example of a sector with high learning velocity is consulting. In this industry, companies have to stay at the cutting edge of skills in order to advise their clients. Another example is the tech industry, where fast adoption is critical to capturing markets and competitive moats can be developed rapidly due to economies of scale and low marginal costs.

“A real-world example of a sector with high learning velocity is consulting. In this industry, companies have to stay at the cutting edge of skills in order to advise their clients.”

We’ve seen it with Threads and ChatGPT, which remarkably amassed 100 million users each in just 5 days and 2 months respectively. In research organizations, employees must have cutting-edge knowledge to keep the organization at the forefront of its field, so it’s typical for them to start the day with learning by reading the newest research publications for several hours at a time. For businesses in these sectors, upskilling workers is not just a learning and development objective, but a daily necessity allowing the company to stay relevant.

So, what about businesses that haven’t traditionally embraced rapid change? For sectors and industries with heavy regulations and risk management, such as power and utilities or manufacturing, excellence is often prioritized within processes. But businesses in these areas are now also increasing their learning velocities to adopt new technologies, innovate faster, and mitigate the risk of disruption.

For the first time in history, learning velocity has become measurable

Creating an organization that systematically maintains high learning velocity is a dream for many CEOs, HR managers, and L&D leaders. Imagine knowing how competitive your company is – or could be – in the face of constant change and innovation. Imagine teams throughout a business having their own learning velocity targets, which roll up to a broader learning velocity outcome and expectation for the company at large. This dream is now possible. 

Following Peter Drucker's principle, "What gets measured gets managed," the initial step is to measure learning velocity, then pave the way for enhancements. 

Over the last three years at Workera, we’ve helped the AI community continually measure and update their skills to develop their careers. I’ve also been fortunate to work directly with dozens of enterprises employing large workforces, who are rapidly adopting AI into their businesses. We’ve run millions of skills measurements, across nearly every industry. These measurements are objective, which means they’re not tied to the specific educational content the learner studied but are based on job-related skills. These measurements evolve and adapt to the levels of employees over time to show trends. The chart below gives you an example of a group’s skills improvements over time on a scale of 300 points in the domain of Large Language Models.

What defines "best in class" learning velocity?

It’s one thing to measure learning velocity – but how do we know if we’re doing well, and what defines “best-in-class” learning velocity”? It is still relatively early to tell (and we will soon release more Workera data that better defines this label) but this is where I stand on what “best-in-class” learning velocity is at this moment.

The Workera scoring system is based on 300 points across three proficiency levels: Beginning (0-100), Developing (101-200), and Accomplished (201-300). Each proficiency level has a descriptor explaining what you can expect of an employee in that level.

Level 3 denotes a “best-in-class” organization, which can broadly expect an average score improvement of above 50 points per month per learner per domain. This means that a learner moves up by one full proficiency level in a single domain every couple of months: from Beginning to Developing, or from Developing to Accomplished. To use a concrete example: an employee who doesn’t know much about a domain like large language models (LLMs) in this organization on average becomes conversational in LLMs within two months of studying it. 

In AI domains, Level 3 organizations typically include big tech companies, AI startups, and forward-looking consulting firms. They prioritize deliberate learning activities over organic approaches – for example, with executives communicating clear Skills KPIs and inspiring the teams to reach ambitious targets. Managers are empowered and held accountable to drive skills improvements for their teams. They recruit people with a growth mindset and durable (more foundational) skills that allow them to learn newer, perishable skills more rapidly. These organizations provide the tools to support learning and typically recommend employees to dedicate 2-3 hours to learning every week. 

Level 2 is what we’d call a “standard” organization. These companies can expect around 25 points per month for learners, or approximately twice as slow as a Level 3 organization. Imagine the consequences of being twice as slow as the Level 3 organization, particularly around how quickly the Level 2 organization is able to adopt new technologies and deliver projects. If it took Adobe twice as long to roll out the new Generative AI features to Photoshop, what would the state of play be now? How many people would have adopted alternatives and not look back? 

They would be perpetually in a state of trying to keep pace with faster rivals, and as time progresses, the disparity is only likely to grow.

Level 2 organizations typically treat learning as a benefit rather than a business imperative, but are often thinking about adopting more deliberate approaches in the future. They are generally not satisfied with programmatic e-learning tools. While not a requirement, they typically recommend 2-3 hours per week dedicated to learning. For AI domains, lots of enterprises in industries outside of technology who have a vision and strategy for learning and development with a focus on innovation domains fall in Level 2.

Level 1 refers to a “behind” organization, which can expect around 10 points per month for learners. In these organizations, the problem isn’t just the ability to adopt new technologies and transform but usually the awareness and willingness to do so. Executives are launching change management initiatives as the leap is likely to require a cultural shift.

Note that being Level 3 is a virtuous cycle as it creates the expectation to learn. Similarly, being Level 1 is a vicious cycle, as you are not pushed to learn and those who want to learn and grow often leave the organization. Over time, this creates an exponentially-growing gap between Level 3 and Level 1 organizations. 

What’s more, learning velocity often decreases as proficiency increases. Simply put, employees generally learn faster at the beginning stages than when they are highly skilled in a domain. 



The graph above outlines the three different levels. A business will fall into one level. The columns show the different scenarios related to the different levels, describing what could or should be typical for a business.

Best-in-class companies bet on high learning velocity to yield long-term profits and lessen disruption risks

Investing in learning is always tricky as you don’t see returns immediately. To be able to leverage generative AI within a few months, Adobe had to instill a culture of learning way before the recent advancements in generative AI. It takes visionary leaders who are able to anticipate long term outcomes. Fortunately for them, enterprises in the level 3 group have realized that learning velocity increases long-term financial returns. 

I’m ready to bet that if you were to chart public companies’ stock prices on a proactive period that’s long enough (for example, between 2020 and 2040) you’d find a high and positive correlation between stock price and learning velocity. Of course, time (and learning velocity measurements) will ultimately tell!

Today – given the integration of generative AI into skill development solutions and the advancements in skills measurements – there’s really no excuse for business leaders not to show measurable learning outcomes and safeguard their business’s competitive advantage well into the future.

Be the first to increase your team's learning velocity in cutting-edge generative AI skills.

To find out more about how Workera and Skills AI is fast-tracking learning velocity for some of the world’s most recognizable enterprises, visit workera.ai/skillsAI