Digital Transformation: Resilient Business Models with Pınar Köse-Kulacz

We were in for a treat in the last gathering of our Digital Transformation Learning Circles. Our guest was by Pınar Köse-Kulacz, Head of AI & Data, Arçelik. Our discussion was on resilient business models. Our guiding questions were: How do rapidly emerging artificial intelligence and data tools make business models more resilient and how do they change some of those business models to prepare better for the future?

This series is co-hosted by İbrahim Gökçen, Strategic Advisor at Open Insights, and Dilek Duman, COO of DenizBank. 

Pınar Köse-Kulacz, our guest speaker, has been leading the AI Function of Arçelik for the last one and a half years. Before her current job, she has worked as the IoT program manager working on the consumer IoT and then moved on to digital transformation. She is a computer engineer and has been around since 1998 working at various jobs in the occupation. She is also Dilek’s mentee:-) What a lovely circle we have! 

This learning circle series is co-hosted by İbrahim Gökçen, Strategic Advisor at Open Insights, and Dilek Duman, COO of DenizBank. 

Pınar Köse-Kulacz, a computer engineer by training, has been leading the AI Function of Arçelik for the last one and a half years. Before this role, she worked as the IoT program manager on the consumer IoT. S. She is also Dilek’s mentee:-) What a lovely circle we have! 

We started the discussion with a recap.  Since the 1990’s there has been a huge emphasis on awareness around AI, of course quite positive, but at the same time, there are lots of risks and issues that come with it, because now everybody knows about AI but not so many people know how to make it successful and scale AI initiatives. So what is the state of AI today? What are some of the proof points we have seen so far? Well, we heard that: 

• AI has evolved into machine learning with the introduction of deep learning. With these super capital computing environments and systems, we could record all of this data.
• With all the mobile technologies, the data starters multiplication to a point where everybody started generating data.
• There is some AI that is already functional in our lives and we don't even notice using it. For example all image recognition systems, or the technology behind Google Maps

Here are key takeaways: 

• People start with technology since it’s available to everyone, but not everyone knows how to use this powerful tool properly.
• People try to apply this tool - technology - to any problem they face and that is one of the biggest challenges today.
• People don’t start from a problem to solve but they start from a technology to apply. They start looking for problems.  In all projects, the main thing is not the technology but the problem definition. But you have to solve, for what you are planning for is very critical, which is also valid for AI projects. The definition of the problem is sometimes not so clear. Another critical challenge is the data culture, which is very important because people are not very knowledgeable about how they can use data for their problems. So sometimes an expert model (the people know-how) is more valuable than an AI model.
• Organizations are usually trying to benefit from this expert model, instead of using data models, so it takes some time to change the culture of the people at companies. For the people to accept that this data is valuable and also understand the machine learning algorithms and AI logic behind that is really important. So if you are an AI expert in a company, you have to have more open communication with the people.
• Most AI stories we hear are success stories. But many didn’t work as expected. 
• A key success factor is how you collect and maintain your data: The data size, especially in large companies like Arçelik,  is huge. Companies with such complexity grow not only organically but also inorganically. Inorganic growth means data will perpetuate. So, putting together all that data and making sense out of it is the biggest challenge. If there is no data, if there is no clean and meaningful data with the required level of depth and breadth, you just can't do anything.

We can hear you asking: What are some practices for AI and data to make businesses more resilient to future sharks and future disruptions?

• Asking questions about KPIs is essential to begin this process.
• Start collecting data as soon as you can.
• Focus and classify your data to be able to see the overall performance.

This session gave us a lot to think about!  We thank our guest Pınar Köse-Kulacz for her valuable insight into the topic and our gracious hosts Dilek Duman and İbrahim Gökçen for another amazing event.

To watch the rest of this rich conversation you can visit the TurkishWIN Learning Center here.