Algorithms - nothing works without them anymore

Functionality and background

Digitalisation & Technology, 04.03.2024

They are the secret, invisible helpers in the background without whom nothing would work in the digital world: algorithms. Almost everything we see on the internet has been created by algorithms. And we are often aware of this, for example when searching the web or on social media platforms. But we rarely know how they work.

Algorithms - nothing works without them anymore

The many algorithms that accompany us on our forays through the internet always remain in the background. We don't have to activate them and we don't receive a notification that they have just completed a task for us. As soon as we switch on the computer, an algorithm runs that meticulously controls the start-up process.

Streaming services such as Spotify or Netflix use algorithms to suggest new songs and series that we have never heard of or seen before. And yet they often fit amazingly well with what we have previously consumed on these platforms. This demonstrates one of the great strengths of algorithms: recognising patterns in large data sets.

The algorithms create personalised taste patterns from the music we have listened to and the series we have watched, which they then compare with the entire database. What matches the recognised pattern is then recommended to the user. We often only realise how well this works when it no longer works as usual. If you try out a new service, for example, you first have to train the algorithm so that it suggests convincing titles.

Even in our physical world, algorithms are constant problem solvers and diligent task processors. Modern assistance systems in cars ensure that we automatically stay on track, enjoy a pleasant climate and the navigation system guides us safely to our destination even in the event of an unforeseen road closure.

There are algorithms for all this and much more.

A simple definition

An algorithm is comparable to a cake recipe. Or building instructions for a cupboard. In both cases, instructions are processed step by step to complete a specific task. In the digital world, workflows are usually more complex and invisible to us. Nevertheless, algorithms are nothing more than clearly defined instructions for solving a problem or completing a task.

How do algorithms work?

We now know that without algorithms, we would not be able to do many things the way we are used to. These hard-working helpers carry out simple to highly complex tasks without complaint. Regardless of their complexity, algorithms have a number of fundamental characteristics that make them work:

Unambiguity: Every step of the algorithm must be clear and unambiguous. Ambiguity would lead to confusion about what to do next.

Executability: Each step must be practically feasible and must not encounter insurmountable obstacles.

Finiteness: An algorithm consists of a finite number of steps. This means that it leads to a clear result or fulfils a defined task.

Input: An algorithm requires an input to solve the task. This can be data, information or variables that are relevant for the algorithm.

Output: The algorithm generates output as a result. This can be, for example, the solution to a problem or a specific result.

Efficiency: A good algorithm should work efficiently, i.e. solve the task with a minimum of time and resources.

Repeatability: If the same algorithm is executed several times with the same inputs, it should deliver the same result each time.

What does algorithm actually mean?

Algo comes from the Spanish and means "something". If this something follows a certain rhythm, we speak of an algo rhythm. Sounds logical? But it's completely wrong.

In fact, it is probably an incorrect translation of the name of a Persian mathematician into Latin. Abu Jaʿfar Muhammad ibn Musa al-Chwārizmī became "Algorizmi" in a roundabout way, from which the term algorithm developed over time. The mathematician worked with the Indo-Arabic number system and written arithmetic in the 9th century AD. However, this had nothing to do with today's algorithms.

Are there evil algorithms?

Even though algorithms solve many problems for us and do a lot of work for us, there are always controversial discussions. For example, when they make decisions or deliver results that we don't like. It often has to do with the input if we don't like the result.

Like any technology, algorithms are initially neutral, i.e. neither good nor bad. However, they are programmed by people who use them to pursue different goals. These do not always have to be evil intentions, as numerous practical examples show. There have already been numerous attempts to use algorithms to make unbiased decisions in sensitive areas of society.

A well-known example from the USA: an algorithm was used to calculate the likelihood of prisoners reoffending and came to the conclusion that dark skin colour was the decisive criterion for reoffending. This discriminatory result was subsequently refuted and can be attributed to an insufficient database as input.

To better avoid such cases in future, algorithm researchers have developed a set of rules. The Algo.rules are aimed at anyone working on the design, development and programming of algorithmic systems. The rules are designed to prevent discrimination and enable the safe use of digitalised automatisms.

Text: Falk Hedemann

Your opinion
If you would like to share your opinion on this topic with us, please send us a message to

Related articles

Digitalisation & Technology 20.01.2021

“Bots create acceptance for digitisation”

The artificial intelligence (AI) team at ERGO has a small, highly efficient sister: Every two weeks, the colleagues in the Robotic Process Automation (RPA) unit launch new robots that quickly increase process efficiency in claims or customer management. But what is particularly important to Mark Klein, the Chief Digital Officer, is that the robots have contributed more to the acceptance of digitisation among employees than any other change measure. Four reasons for this.

Digitalisation & Technology 06.11.2023

“From zero to 450 RPA bots”

Mark Klein, Chief Digital Officer of ERGO Group AG, was a guest on the “State of Process Automation” podcast. He discussed with host Christoph Pacher how bots at ERGO relieve employees and improve customer satisfaction.

Digitalisation & Technology 03.07.2024

Digital Culture: Unlocking the potential of innovation

Who drives innovation? How can people be inspired by new technologies (such as AI, VR or the metaverse)? And what mindset is needed for a successful digital culture? ERGO CDO Mark Klein and Nina Michahelles from Google talked about these topics in the first //next Talk.