Becoming a Data-Driven AI Company

Image by Geralt on Pixabay

During the last week I was reflecting on the change processes ongoing in several of the companies that I work with. Although I don’t want to generalize too aggressively, it seems to me that most companies are on a similar journey. Starting from a situation where the key value proposition of the company was expressed in mechanical and physical products, over time electronics, software and data have become more and more important. And as the most recent development, it is now artificial intelligence that is the key focus technology area that is receiving significant attention.

Figure: Typical evolution path for traditional companies

The interesting thing is that companies can get started at each point in the evolution shown in the figure above. No company ever gets free from its original starting point (or at least it is really hard and takes a long time). So, a company that started as a mechanics or electronics company and became initially successful with that technology will always have a tendency to continue to include that technology in its offering to users.

However, no matter where the company started, it is always possible, and often absolutely necessary, to include the next technologies in the offerings of the company. This means that a company that started as a mechanical company tends to have added electronics in the 1970s and 1980s, software over the last decades, has been trying to figure out how to make sense of data and is now looking at artificial intelligence (AI) and trying to figure out what to do with it.

So, rather than assuming that a company only lives inside one technology, sustaining companies that are more than a “one trick pony” manage to straddle multiple technologies and achieve success by investing in, combining and exploiting multiple of the aforementioned technologies.

The challenge, however, is that it often requires different priorities, cultures and ways of working to co-exist in the company. For instance, for most companies mechanics and electronics have been commoditizing and consequently the focus is heavily on cost minimization and efficiency. Software often defines the key value proposition and requires significant investment in order to capitalize on the key differentiation that software provides. Finally, data and artificial intelligence is often still in the innovation stage, meaning that the company needs to innovate and experiment with many different approaches and ideas in order to find those that really deliver value to customers. See also this earlier blog post.

A misconception is, though, that is quite pervasive is that technology investments and monetization of customers is aligned. These are two completely orthogonal challenges. It may well be that the company has to heavily invest in mechanics and electronics, but manages to monetize through offering data various means. Although in most cases there is a correlation between the R&D investment and the monetization strategy, this is not a “law of physics” and it may well be that the company requires the investment in mechanics and electronics to achieve a critical mass to make its data set relevant for monetization.

Concluding, reflecting on the companies that I work with, I see a pattern where the identity of companies constantly gets expanded with additional technologies, starting with mechanics and then electronics and software and now, for many companies, data and artificial intelligence. Each technology becomes the key differentiator at some point in time and then commoditizes, causing the focus to shift from innovation and investment to remain differentiating to cost efficiency and economies of scale. There are several other aspects to this, but I leave those for other articles. For now, my advice is to carefully analyse where you are as a company and ensure that your priorities, resource allocation and focus is where it should be to stay competitive over time.

To get more insights earlier, sign up for my newsletter at jan@janbosch.com or follow me on janbosch.com/blog, LinkedIn (linkedin.com/in/janbosch) or Twitter (@JanBosch).

Leave a Comment