It was maybe two years ago when I was first in discussions about data becoming a liability to companies. Until that time it had only been seen as an asset. This thinking is becoming mainstream and is really changing the behavior of companies. The masterminds that were devising models to get more data five years ago are now concentrating on how to make services that come without ‘data liability’ or are simply creating entirely new data models.

A Google search now reveals several articles about data as a liability and I have raised the subject with many significant tech and consumer companies in Silicon Valley. One very significant company has even told me how they are systematically deleting information that is not directly linked to their core business. Some other companies have mentioned how they had earlier offered data as a bargaining chip to get good deals with other companies. But they have now seen this no longer working for them and many are avoiding collecting data because of the potential liability attached to it.

We are still in a watershed moment; some businesses and business leaders are still in the old paradigm that they want to get more data and believe it is key to their business. But the most advanced companies are finding new ways to get value and new customer relationship models so that they can minimize their liability yet still get value from customer knowledge and also find a fair data relationship with their customers.

These changes and new models are not always easy to explain to the old paradigm people. They might think that the only way to use the data is to have it in their own hands. Of course, new data regulations first in the EU and later, for example, in California or New York will also accelerate this change and understanding. However, it is not easy to understand different models that utilize data if you don’t have the basic knowledge of data science and an understanding of software business models and how software is written, used and distributed nowadays.

Data traders and brokers are the first ones to really suffer from this change. It is not only that companies have become less willing to buy data generally, but the reputation and image of the data trading business has suffered significantly. There are many good reasons for this and we can say that not all of those companies have been ethical or transparent in their business – sometimes with operations in a ‘grey’ area.

There are at least three models to handle data in a new way:

1. Consumers and customers own their own data and it can be utilized in the customer relationship with customer’s consent;
2. Data is kept in distributed solutions (e.g. distributed ledgers) and it can be public or semi-public, depending on its implementation;
3. Relevant data is collected and requested only when needed and it is not stored and specifically not distributed to 3rd parties.

Many new solutions can also combine components of models #1 and #2. Model #3 can also utilize the model #1 approach to get data from the customer. We are still in the early phases of these models, but it is clear there are significant development resources in place to get these to happen with forward-looking businesses actively looking for new solutions.

Some ICO companies have introduced models where people could own their data and then sell it in return for some tokens. This model’s most relevant point is that people could get value from their own data, but in reality, it is very hard to get this kind of market to actually work. The idea of data exchanges and markets is looking quite dead. During the 25-year history of Internet services, we have seen many market place ideas that have not worked in real life. The personal data exchange model is probably one of those.

It is relevant that people get fair value from their own data, but it most probably comes in other formats not necessarily suited for sale in an open market place. And how do you price your data? Can you sell it for one-time use only? Some have compared it to selling one’s own organs, and I can see the point in the comparison.

The key for new data models is to find new customer relationship models. How people can get value from their data in daily situations. The value can be better experiences, better prices and more relevant services. The company must be able to serve the customer better, if the customer shares data in the transactions. Technology, including AI, offers many new ways to achieve this.
​
Changes take time. Most automobile companies are still making combustion engine cars that are driven by human beings, although everyone knows the future belongs to self-driving electric cars. No serious carmaker can ignore this future and they must also invest in these future cars. It is the same in the data business, many companies must still manage their old model data, but they must prepare for the future of the data business that is much more distributed and customer driven.

One could easily think that IT and digitization are somehow the same thing – or at least support one other. The corporate reality is that it is sometimes the opposite. It is often the legacy IT and the IT department that are the obstacle to new digital models. Is there any way to get traditional IT and digitization to work together or do we need total disruption to change things?

Legacy IT systems have been built to support processes and operating models that were dominant when the original systems or architectures were designed. It generally happened before truly digital companies began to emerge. By digital companies I mean companies that are built on digital data, data-oriented processes and models built on digital customer experience. We can see that companies such as Google, Uber and Amazon are examples of really digital companies.

A former bank executive said to me recently that “he hasn’t invested in bank shares for years and at the bank he felt like he was sitting on a time bomb with core legacy IT systems.” He said that everyone knows they cannot continue like that for long, but it is scary to start to replace systems where most people have their money. New systems might offer better services for lower costs but he is not brave enough to take those steps, because something might go wrong.

New regulation, for example GDPR and PSD2 in Europe, have demonstrated how hard it is to live in the digital era with legacy IT systems. For example, banks should be able to provide data to their customers, but how they do it is not very modern. An executive from another bank told me how they employ someone to manually collect data on an Excel sheet, when someone asks to get his or her data, and then email it to the client.

This is very different from the big public talks about open API banking.

In practice we have also seen that IT departments are typically very skeptical about accepting any new systems, even though top management and business leaders would like them to. Someone could say they are conservative and against change but there are also very practical reasons for this. They have a hard time managing the existing systems and typically it has been hard to get the legacy systems to talk to each other. Each new system has meant expensive system integration projects.

Generally, it is hard for incumbent companies to change and change their operating models. That’s why disruption has happened in many industries and new companies have emerged to kill the old companies. In some cases, the old companies have survived, but most of the new business has gone to the new players (for example media companies, telco carriers and bricks and mortar retailers).

But are there some ways to make the transition. There are no simple solutions, certainly no miracles, but we can suggest some things that can help:

1. The top management must understand technology, digital business models and be brave enough to make mandatory changes early enough,
2. The starting point for new services and processes must be the customer value and customer experience, not technology or internal processes,
3. There are technology solutions, e.g. RPA (robotic process automation) that help to make the transition and enable digital processes on the top of legacy systems, thus getting old and new systems to talk each other and at the same time enable significant information work automatization.
4. Everyone in the organization must be ready and open minded to look for better solutions to serve the customers and offer them value. It means people must go outside their comfort zone and admit their old routines don’t always make sense. It is better to proactive than reactive.
5. Find real facts and not just trust beliefs you like. For example, cloud, automation tools and Machine Learning are not security risks compared to the old in-house system. All systems have risks, and it is important to implement and manage them right. The devil is in the detail.

We will see more significant digitization in most industries. It will kill many big companies and create new significant companies. We will definitely see significant changes in the models to manage and use customer data, build digital processes on that customer data and make all operations be based on that data. Management must have the courage to drive those changes, including a complete transition from legacy IT to totally new systems and models.

That said, there are some softer ways to handle the technological change, but even with those models it is fundamental to keep the focus on customer value, not on internal development.
​
Your focus must not be to develop IT, but your customer value and experience.