Seven ingredients of successful data-driven innovations


22 december 2017


The past year, I have analysed the innovation strategies of companies like Google, Oracle, Nike, Vodafone, IBM and Phillips. In this blog I will address two topics: my understanding of why innovations succeed and fail and my learnings from successful data-driven innovations.

Since 2012, over 50% of the Fortune 500 companies have made innovation an integral part of their strategy. Due to success of tech scale-ups that leverage new technologies such as Data Science, Artificial Intelligence and Blockchain to disrupt incumbents, companies acknowledge that they have to innovate. However, 90% of innovation labs fail and according to a recent Capgemini report only 5% of R&D staff feels highly motivated to innovate. This results in budget cutting, increasing costs and not being close to innovation. Companies like Adecco, Microsoft Silicon Valley, Coca-Cola and Disney have even closed their innovation labs.

Innovations fail. Why?

The end user doesn’t embrace the innovation
For a company in the transport and logistics sector, the idea was to optimize the delivery time. By implementing real-time data analytics and optimal route calculation, the delivery time decreased with 10%. However, the end user (carrier) didn’t use optimal route calculation. Why not? The carrier is used to traditional processes. He knows certain roads and uses a standard route to reach his destination. For him it doesn’t feel as an improvement. He had to take new routes and drive on roads that he had never seen before, which felt uncomfortable.

The wrong momentum
The revenue from Oculus Rift hardware is forecasted to amount to 4.95 billion US dollars worldwide in 2019. However, Oculus wasn’t the first to produce a virtual reality headset. In 1995, Nintendo produced the Virtual Boy. This product was described as ‘the single most uncomfortable gaming device on earth’. But was it? Or do we only accept an innovation when it suits our behaviour? Likewise, Google Glass was launched in 2013 and laid to rest after two years. Although it’s officially back as an enterprise-focused product, Glass was never accepted by society due to privacy issues. People simply don’t feel comfortable when they can be recorded without being asked for permission.

People, team structure and responsibilities
Companies are intent on setting up internal innovation teams that connect people from the business and the technical side of the operations. But who has ownership? People from the business might fill in a double position and their innovation-driven KPI’s might interfere with their traditional KPI’s. This leads to less committed and motivated team members, and thus poor performance.

“Innovation is mostly about finding the right momentum to launch new products or services by finding the right balance between people, technology and society.”

Scalability and speed
Tesla told Model 3 reservation holders via Business Insider that it’s pushing back delivery dates of this highly-anticipated vehicle. Tesla has set itself apart by producing the most innovative automobile of the last decade. Their biggest challenge is scalability. If they cannot build factories that produce the amount of cars that their customers expect, this may incite competition. I wonder how long their customers are willing to wait.

Innovation is aboutIn my opinion, innovation is mostly about finding the right momentum to launch new products or services by finding the right balance between people, technology and society.

Here are my top seven take-aways.

  1. Choose the right Product Owner

The most important team member in the innovation funnel is the Product Owner (PO): a Principal or experienced technical leader. He or she guides the (scrum) team, which could contain different technical roles such as Data Scientist, Engineer, UX Designer). The main responsibility of the PO is to set out the vision and force the development team to stick to it. They clearly and explicitly align end-user expectations with the powers of the scrum team. This way, a PO helps to ensure that the team can focus on delivering optimal value.

  1. Stimulate performance by creating ownership and responsibilities

Besides the PO, you need a bridge to your business: the Business Owner (BO). The BO makes sure that the PO is facilitated in everything that he needs, which is mostly input and end-user feedback. The PO is responsible for the deliverables and the BO ensures that the stakeholders have a representative who is available for quick questions. This way, you maintain a high pace of delivery.

  1. Bring together the right people

You want to gather people who have an intrinsic motivation to be a part of this innovation team; people with a great passion for the business and data-driven technology. People need to understand the reason why your company wants to innovate and how this will succeed. Altogether, the team must understand the current way of working, your business and the technical challenges they will face.

  1. Find the right momentum and choose the optimal focus

The focus or momentum of ideas should be optimal: neither too futuristic nor too close to the current business. When discussing ideas ask yourself the next questions. How would it influence our way of working? How big is the shift of daily processes that the end-user will experience? Will end-users accept or embrace these changes? Does the technology empower our vision?

  1. Provide a roadmap for the innovation funnel

Organise a workshop or an inspiration day to capture a handful ideas which contain a specific Minimum Viable Product. Find a tangible small use-case that provides a measurable value instead of a beautiful idea that doesn’t show its value right away. After choosing two or three ideas, determine as quickly as possible if this idea could be a winner. Before scaling up, build a prototype in a maximum four weeks and let end-users test it.


  1. Hunt for feedback

If you want the best end product, you have to anticipate real user scenarios. As you need to foresee possible reasons for failure, make sure you thoroughly test your prototype. Collect feedback from all stakeholders about how this innovation influences their processes. If and when the end users actually embrace the innovation and the privilege to use it, then scale up. Don’t stick to a nice idea and ruthlessly cancel it if needed. Cancelling an idea as early as possible gives you more time to test other potentially successful prototypes.

  1. Ensure adoption of change

Not all employees will immediately adopt an innovation despite its obvious benefits. You need to understand their culture and how the innovation influences their way of working. To make sure that people accept an innovation, host seminars about the changes and show them how the innovation improves their way of working. Visualise the goal that your company wants to achieve, and how this innovation is a part of it. The aforementioned carrier will eventually use real-time route calculation, but that needs explanation.

About the author

Derk Disselhoff is a Consultant at Quantillion, a strategy consulting firm in the field of Data Science. This company entered the market in February 2017, and has since then grown to a team of 35+ consultants who develop and deliver data-driven solutions to Top 500 companies and government institutions in the Netherlands. Quantillion’s cases include advanced data driven recruitment technology, realtime crowd management, data-driven M&A technology, predicting optimal treatments in healthcare, predictive server optimization and data-driven real estate advisory.


Geef een reactie

Het e-mailadres wordt niet gepubliceerd. Vereiste velden zijn gemarkeerd met *

  • *