Why Companies need Data Scientists?

It is common sense to assume that businesses make hiring decisions to further their own agenda of making more profit for their investors, improving their brand perception and the lives of the people involved in their operations and off course of their customers. So, these are also the reasons why they would want to recruit a data scientist. Data scientists working for an enterprise are thus in the business of using their data analysis skills to further this agenda. Frequently, the stage the business is in will also determine which one of these goals matters the most. A well-known business in the midst of a reputation crisis would want to improve its brand perception more than other businesses. They might hire for reasons other than the two most common reasons, working towards improving their bottom line or helping them beat their rivals in the market. A start up will however focus more on hiring individuals who will work hard to help it get profitable as soon as possible. In such cases businesses are not unlike individuals, with different needs and goals, in need of the right people to give them direction. Another factor that determines the priorities of a business is the industry or sector it operates in. Internet companies operate in a highly competitive space with customer acquisition as the biggest hurdle to profitability, second biggest being monetizing their user base. A pharma company needs to constantly innovate and operate in a highly regulated space with high product development and opportunity costs.

Having talked about the motivations and priorities of any business let’s get more specific. What can a company hope to achieve by recruiting a data scientist or a team of data scientists. What do they expect them to achieve? First and foremost, it depends on the responsibilities that have been chalked out for different roles. If a person is recruited as a data architect or data engineer his job will be to ensure that people in the company have timely access to the right data without putting the company data security to risk. As such he needs to decide what architectures to employ, how to scale the information systems as the business grows and how to ensure the security of the data architecture. Typically, a data engineer is not tasked with deriving value from data. On the other hand, a data scientist or a business analyst is hired with the stated intention of deriving value from the data. This brings us to another linked question: — what can be achieved with data, or how can a business use data to further the agenda which we discussed at the start of this blog post. The answer to this question leads us to the most common (and not so common analytics use-cases) in the business world. Analytics use cases can be divided into two main stacks: -

Analytics as an accessory to another function: In these cases, analytics has been used as part of the function for some time, however the process of data collection, processing and analysis may not be sufficiently mature, leaving scope for further improvement. Also, the data is usually the data that has been collected as part of business operations or sourced from third party providers in the functional space, for example marketing function of a company using its google analytics data in conjunction with data from a marketing consultancy. The goal here is to improve the performance of the functional unit by: -

1. Helping make better decisions: — businesses need to make many decisions on a daily bases. These decisions encompass all the functional areas and activities required to run a successful business. Within a functional unit, many decisions need to made that if taken well contribute to it’s performance. For example, for marketing, budgeting decisions, for sales, incentive decisions, for operations, supply chain management etc.
2. Helping formulate better strategy: a business’s long-term strategy in any domain is determined by the forces that determine its profitability i.e. it’s competition, new entrants, suppliers, buyers, substitutes etc. Also, the profitability depends on macroeconomic indicators which determine the trend for the industry as a whole. Data scientists can leverage several frameworks to help formulate a optimal strategy for a business to respond to changing market forces.
3. Helping design better products: — Data scientists can also contribute to a company’s profitability by improving the products and services it sells. For example they can help make more failure resilient, easier to use, more attractive, more popular and more useful products that help a company’s growth. In such cases, analytics is used to constantly improve the design and features of new products.
4. Tracking and improving performance: — a business needs to launch various initiatives to help achieve it’s vision. The performance of these initiatives is tracked using metrics that have been designed to track what will contribute to the success or failure of such drives. Once such indicators are in place, performance can be measured against them and problems, if any, can be addressed. In further iterations, success can be replicated and mistakes avoided.

Analytics as an independent function (or analytics as a competitive advantage): In these cases analytics is not tied to a specific function and its objectives derive from the organization’s own mission.

1. Solving Problems: Businesses encounter several issues during their day to day operations. Several of these issues are critical to their success and profitability. For example, customer acquisition for an internet company, stock clearance for a fashion retailer, drug testing for a pharma company or customer retention for a bank. If left unaddressed, these issues can severely impact their profitability. Data scientists help diagnose the trouble spots and ways to address these.
2. Spurring Innovation : Data science is increasingly used in innovation by almost every industry. Increasing sophistication of algorithms and better resources at the disposal of data scientists, for example more compute and storage capacity will only help increase the already impressive number of use cases data science has found.
3. Unlocking value from data: Data scientists can also take a bottom up approach and instead of starting with a task or a goal, start from the data and find uses for it. This frequently contributes to both innovation and unexpected insight into pestering issues and creative ways to solve them.

As analytics becomes more central to organizations, we will see many new ways to store, organize and analyse data and new avenues of growth for the organizations.