Giovanni Cervellati
Giovanni Cervellati (Research Manager, European Software Group)

In a world that is moving fast towards simplification — in everyday life and in the workplace — the analytics industry is ready to move away from requiring its workforce to expend most of its energy on writing lengthy, complicated code strings.

The analytics software market has long offered easy-access solutions for enterprises. With data scientists increasingly focusing on real-time insights or predictive analytics, we cannot afford for them to concentrate only on the technical side.

What You Don’t Expect: No-Needed Programming on the Rise

Nowadays, people are keener and keener to have more insight with less effort. This is true for both everyday life and the workplace, and it’s happening faster than we might think.

Think about how much time and effort you would have spent 10, 20 or 30 years ago carrying out a task — it would almost certainly have taken much longer than it would now, and the time it takes has reduced decade by decade.

You’d think the same is happening in analytics. It’s obvious, right? Unfortunately, no.

The analytics world is somehow following a counter-intuitive trend. While platforms have long been providing easy-access solutions to the labour market’s analytics needs, use of open source platforms has surged, requiring data scientists to have coding skills.

The huge rise in frameworks leveraging Python, R and other language applications has reached the point where people identify the data scientist job as a “high-code-employee” rather than a person with deep knowledge of statistics and machine learning algorithms or processes.

Companies and universities have rapidly adapted to this, and are teaching programming languages as part of their analytics courses. As a result, the need for code-skilled professionals has also increased, widening the gap between highly skilled analysts and data scientists and business-literate employees — in some ways, exactly the opposite of what is going on in the outside world.

But why has this happened? The reason for switching from user-friendly software to a code-based application has often been the free access to it (most are open source) and its guarantee of continuity without being kept in check by vendors. The surge in open source software in analytics and ML has led to this unique situation where high-code applications have been taking advantage of classical, low-code applications over the past few years.

The game-changer in the whole analytics and data market has been the move from legacy on-premises solutions to the cloud. Initially, cloud providers further strengthened the code-loop because they were focused on professional developers and have been bundling open source analytics tools and frameworks widely for some time. They like machine learning workloads because they use a lot of compute cycles and data storage, which they can charge for and because they get analytical capabilities for free. The recent implementation of many no-code platforms in the cloud may slow down this trend or at least provide an alternative.

The Path to Data-Driven Enterprises Must Be Simple

In today’s volatile, fast-changing business environment, it’s critical that companies have the best possible analytical capabilities. Whether it be more focused on descriptive, visual analytics or predictive analytics through machine learning, enterprises need fast, easy analytics to speed up their journey to becoming a data-driven enterprise.

The trend towards improving data literacy and data democratisation doesn’t fit with this trend to high coding tools. There is a need for employees who can use and understand the business’ data and take advantage of insights.

To become an intelligent enterprise, you should focus on creating a data culture. This culture should permeate the entire company, with people at the top. Employees should speak the same language, and data scientists are no exception. Leveraging the investments in technology can free the company from cultural silos, leading to better communication with employees more comfortable with data. Viewing data scientists as “nerds” doesn’t make things easy — implementing a widespread data culture will shorten the distance between people in the enterprise, and user-friendliness and natural-language software must be a big part of this. Data scientists themselves can change their approach from “what do you want from the data?” to “you may need it”.

A data-driven enterprise will look to any effort to make data and its use more understandable, more widespread and easier to use — breaking down the walls that keep technical jobs separate from other professionals.

The “Smart” Data Scientist in a Low-Code Enterprise

Relying on code-based applications for analytics and machine learning is making it harder to become a data-driven enterprise.

I have been a data scientist for almost 15 years, and I know that focusing on line-of-code errors makes the job harder, more frustrating and sometimes impactless. Coding is not natural and it’s not easy to use, and it makes the analytics work both more time consuming and harder to understand. You can just lose sight of the main goal.

The era of smart working is already here, and the pandemic has accelerated the move towards it. Home offices, digital transformation and agility are now everyday concepts in today’s companies.

In this scenario, a smart data scientist is not a high-code specialist, but someone who can seamlessly find insights, trends and patterns in the data, and use their knowledge to interpret it in the best possible way.

Technical vendors are pushing their easy-to-use, innovative new solutions. With the new software, you can solve most machine learning issues and tasks even with limited technical knowledge: experts’ added value can almost exclusively come from their ability to understand the big picture of analytics and the “good-looking” results you can get from the data. AI is also moving in this direction, giving access to fast and easy information through natural languages or simple acts.

By investing in these new solutions, an enterprise can open the way to seamless, broader data knowledge and leverage data scientists’ experiences to achieve the best results.

Now is the time to rediscover the power of simplicity: in an over-simplified world, analytics should no longer be “just for the few”.


To learn more about our upcoming research, please contact Giovanni Cervellati, or head over to and drop your details in the form on the top right.

Spread the love