About data analytics with Cetin Karakus
**Data analytics is talk of the town these days and playing a vital role in increasing sales and profitability. What are the top strategy and technical issues and concerns related to data analytics that every c suite must know.
It is fair to assume that most executives will share the belief that the data their organization possesses is like an untapped mine ready to be dug to produce untold riches. I think there is no need to convince anyone of the importance of the data and data analytics in the success of a business. The billion-dollar question is how to actually unleash the value in the data in practice?
The naive answer is just to throw some new tools and technologies and expect it to automatically turn a pile of data into readily actionable business insights. This is the perspective vested interests like technology vendors and certain consultancies spread around because it is good for business, their business that is, not yours.
The reality is far more complex and subtle. It is true that technology plays a key role in handling the data, especially when the scale, speed, and variety of the data handled changed from what an enterprise was used to. However, data becomes valuable only when it can be turned into information and that only happens when a certain structure(i.e. form) and semantic(i.e. meaning) is attached to the data and various types of connections (i.e. relation) between seemingly disparate datasets are established.
This trio of form, meaning and relation should be the pillars of any successful data analytics strategy and should guide both the corresponding systems & technology development and, more crucially, people and teams organization.
You will have more technology and infrastructure-focused teams bringing data into a common, canonical form and hence making its utilization downstream more streamlined and efficient.
You will have people who understand the business well working on attaching domain-specific meaning to the data which has already been put into a common form.
And you will have people who start asking questions, setting up hypotheses and testing those hypotheses through controlled experiments using the data which is already structured well and has a certain business-specific meaning. They will find hidden relations buried deep inside the data through such efforts and that could lead to the actionable insights that can create real business value.
**How data analytics is changing the way of core functions like HR, Finance, IT and marketing. What are the gaps being noticed? Where c suite is lacking? What efforts are required to augment the performance of data analytics program
The most obvious change I see is the wider use of KPI-type common metrics and dashboard-type visualizations being used. Such kinds of basic analytics capabilities usually come prepackaged with the new generation of cloud-based systems which seem to be replacing the traditional, on-premises systems which used to run such functions.
As much valuable as such basic analytics for the day to day operations of a business are, they are far from utilizing the inherent value of such the data they are handling. The main gap I see is the disconnect between such systems and more business-focused, "front office" functions such as product development and sales. A sound data analytics strategy relies on a holistic approach to the data and incorporates all the data from different systems and domains and tries to find hidden connections between seemingly disconnected activities. That is when real business insights that can be acted upon emerge.
**If CEO decides to incorporate data scientist role in his portfolio than what new portfolio looks like? What will be the new KPI and KRA getting added in new role?
Data analytics/science is a cross-cutting effort and have to be spread across the organization to be successful due to the holistic approach I was referring to earlier. It is a bit like a risk function in the sense that there has to be an active risk perception across an organization for any risk management function to be effective. Similarly, there has to be a wider perception of helping with data analytics activities across the organization, be in helping with structuring the data (form), or in attaching meaning to it (semantics) or in finding connections between disparate datasets (relation).
Chief data scientist should help CEO to establish such a perception across the organization, perhaps including it as one of the core values of the company, just like any other core values of the company such as safety.
Once such a "data-awareness culture" is established across the organization, its ongoing effectiveness could be monitored and improved with appropriate KPIs such as how much of the data the organization deals with does not conform to the standards established along the fundamental pillars of form, meaning and relation, what is the pace of addressing such non-conformity, what is the conversion rate of finding new relations in the data to the successfully executed business actions, etc.
**What initiative c suite must take in order to get involved in data analytics strategy? Where to draw the lines when it comes to involvement in day to day functioning?
C suite can play a crucial role in establishing a "data-awareness" culture across the organization.
To truly leverage data requires rank-and-file employees in the organization to share the vision that there is an inherent value in actively contributing to the efforts constituting the aforementioned three pillars of data analytics strategy.
C suite should champion such a culture and principles behind it and empower lower ranks to act upon those principles and to take initiatives inline with them.
**Rethinking and redefining the role of CEO's in identifying data talent and building a strong data analytics team and fostering data culture
CEO's most important role should be to establish "data awareness" as one of the core values of the company and ensure that a data analytics strategy like the three pillars one mentioned earlier is adopted and executed across the company.
Once that is done, talent management becomes a lot easier of a problem thanks to both the organic, internal talent growing up inside the company and also the steady flow of external talent being attracted to the superior data culture established in the organization.
It is fair to assume that most executives will share the belief that the data their organization possesses is like an untapped mine ready to be dug to produce untold riches. I think there is no need to convince anyone of the importance of the data and data analytics in the success of a business. The billion-dollar question is how to actually unleash the value in the data in practice?
The naive answer is just to throw some new tools and technologies and expect it to automatically turn a pile of data into readily actionable business insights. This is the perspective vested interests like technology vendors and certain consultancies spread around because it is good for business, their business that is, not yours.
The reality is far more complex and subtle. It is true that technology plays a key role in handling the data, especially when the scale, speed, and variety of the data handled changed from what an enterprise was used to. However, data becomes valuable only when it can be turned into information and that only happens when a certain structure(i.e. form) and semantic(i.e. meaning) is attached to the data and various types of connections (i.e. relation) between seemingly disparate datasets are established.
This trio of form, meaning and relation should be the pillars of any successful data analytics strategy and should guide both the corresponding systems & technology development and, more crucially, people and teams organization.
You will have more technology and infrastructure-focused teams bringing data into a common, canonical form and hence making its utilization downstream more streamlined and efficient.
You will have people who understand the business well working on attaching domain-specific meaning to the data which has already been put into a common form.
And you will have people who start asking questions, setting up hypotheses and testing those hypotheses through controlled experiments using the data which is already structured well and has a certain business-specific meaning. They will find hidden relations buried deep inside the data through such efforts and that could lead to the actionable insights that can create real business value.
**How data analytics is changing the way of core functions like HR, Finance, IT and marketing. What are the gaps being noticed? Where c suite is lacking? What efforts are required to augment the performance of data analytics program
The most obvious change I see is the wider use of KPI-type common metrics and dashboard-type visualizations being used. Such kinds of basic analytics capabilities usually come prepackaged with the new generation of cloud-based systems which seem to be replacing the traditional, on-premises systems which used to run such functions.
As much valuable as such basic analytics for the day to day operations of a business are, they are far from utilizing the inherent value of such the data they are handling. The main gap I see is the disconnect between such systems and more business-focused, "front office" functions such as product development and sales. A sound data analytics strategy relies on a holistic approach to the data and incorporates all the data from different systems and domains and tries to find hidden connections between seemingly disconnected activities. That is when real business insights that can be acted upon emerge.
**If CEO decides to incorporate data scientist role in his portfolio than what new portfolio looks like? What will be the new KPI and KRA getting added in new role?
Data analytics/science is a cross-cutting effort and have to be spread across the organization to be successful due to the holistic approach I was referring to earlier. It is a bit like a risk function in the sense that there has to be an active risk perception across an organization for any risk management function to be effective. Similarly, there has to be a wider perception of helping with data analytics activities across the organization, be in helping with structuring the data (form), or in attaching meaning to it (semantics) or in finding connections between disparate datasets (relation).
Chief data scientist should help CEO to establish such a perception across the organization, perhaps including it as one of the core values of the company, just like any other core values of the company such as safety.
Once such a "data-awareness culture" is established across the organization, its ongoing effectiveness could be monitored and improved with appropriate KPIs such as how much of the data the organization deals with does not conform to the standards established along the fundamental pillars of form, meaning and relation, what is the pace of addressing such non-conformity, what is the conversion rate of finding new relations in the data to the successfully executed business actions, etc.
**What initiative c suite must take in order to get involved in data analytics strategy? Where to draw the lines when it comes to involvement in day to day functioning?
C suite can play a crucial role in establishing a "data-awareness" culture across the organization.
To truly leverage data requires rank-and-file employees in the organization to share the vision that there is an inherent value in actively contributing to the efforts constituting the aforementioned three pillars of data analytics strategy.
C suite should champion such a culture and principles behind it and empower lower ranks to act upon those principles and to take initiatives inline with them.
**Rethinking and redefining the role of CEO's in identifying data talent and building a strong data analytics team and fostering data culture
CEO's most important role should be to establish "data awareness" as one of the core values of the company and ensure that a data analytics strategy like the three pillars one mentioned earlier is adopted and executed across the company.
Once that is done, talent management becomes a lot easier of a problem thanks to both the organic, internal talent growing up inside the company and also the steady flow of external talent being attracted to the superior data culture established in the organization.
Cetin Karakus
Global Head, Quant Technology and Analytics Core Strategies, British Petroleum
Cetin Karakus has almost two decades of experience in designing and building large scale software systems.
Over the last decade, he has worked on design and development of complex derivatives pricing and risk management systems in leading global investment banks and commodity trading houses. Prior to that, he has worked on various large scale systems ranging from VOIP stacks to ERP systems.
In his current role, he had the opportunity to build an investment bank grade quantitative derivatives pricing and risk infrastructure from scratch. Most recently, he has shifted his focus on building a proprietary state-of-the-art BigData analytics platform based existing open source tools and technologies.
Cetin has a degree in Electrical & Electronics Engineering and enjoys thinking and reading on various fields of humanities in his free time.