Skip to Content
Technical Articles

Role of business analytics in building AI-powered Organizations

There was a 270% increment in the number of enterprises integrating artificial intelligence with their business processes in the four years between 2015 and 2019. However, only 30% of all these companies deploying artificial intelligence (AI) have reported more than 90% success in terms of return of investment. A significant number of enterprises report a 10-49% failure.

The cost of AI implementation is high and the expectations, in terms of returns, are even higher. Failure on this front can cost an enterprise very dearly. In cases of failure the lack of capable personnel often takes the blame but the most cited reason for these failures is an absence of vision.

Artificial intelligence is not a plug-n-play device. According to Gartner, 37% enterprises have deployed AI in some form or the other but something is surely missing. AI needs to be used wisely and the want of that wisdom is what makes the business analysts and their knowledge so vital for building a company which works with AI.

The Journey of AI in a nutshell

AI stands for Artificial Intelligence as you know. A machine with cognitive abilities – that can learn from experience and respond to digital or mechanical stimuli depending on the acquired knowledge. The term was coined back in 1956 at a conference at Dartmouth College.

There was serious research for a couple of decades in Britain. Then the government stopped funding the projects in 1974 for a few years as the results were insubstantial.

In 1997 IBM’s AI system Deep Blue beat chess grandmaster Garry Kasparov. AI has never looked back ever since. In a span of 20 years AI systems have come a long way from winning quizzes and beating Go players to analyzing protein structures.

Today, every business in every field is looking for ways to implement AI in some way or the other. It has achieved the status of some sort of stardust that can magically transform an organization.

What does an AI-Powered organization look like?

It does not necessarily have humanoid robots walking around the office offering tea along with good advice to the few human employees who are left. The focal point of AI implementation is augmentation of the human workforce with extra-human cognitive capabilities.

Let us take Alibaba, the Chinese conglomerate for example; it sells more products than Amazon and E-Bay combined and AI is the pillar that supports their central framework. Alibaba uses Natural Language Processing (NLP) to automatically generate product descriptions. It also uses computer vision to monitor the traffic situation of the smart cities they helped build, to help manage the roads.

Tech giants like Alibaba, Amazon, and Google, count a great deal on the partnership between knowledge workers and intelligent machines.


The building blocks of an AI-Powered enterprise

The implementation of an Intelligent Process Automation system that streamlines and executes various tasks while also significantly improving overall efficiency, takes time, thought, and investments. Only 8% of the companies that deploy AI, manage to achieve a company-wide integration. Others which make siloed usages of AI often fail to unlock its powers.

An organization that aspires to use AI at its present best has to

  • Adapt a data centric workflow rather than a leader centric one.
  • Replace the traditional top down decision-making approach with a real time AI augmented approach.
  • Augment each segment of the business with AI.
  • Remove risk averseness and rigidity in favour of positive experimentation.

The interdependence of business analytics and AI implementation

Business analytics can be simply defined as a methodical study of data to improve business. Various segments of artificial intelligence and business analytics have become interdependent. Let me explain why.

The two way growth that enabled AI

Business analysts try to draw insights from historical and real time data by processing it and sending it through layers of filters and algorithms. The amount of available data has increased beyond human imagination over time and traditional tools have failed to provide the processing power to analyze that much data.

Two things happened in favour of the business analysts quite simultaneously. The amount of accessible data and computational abilities grew together. After a point of time there was enough data, and computers powerful enough to process them.

The long held dreams of artificial neural networks and computer vision could finally be realized as machines could learn from data without manual intervention. And thanks to the ease with which one can find and attend a good business analytics course nowadays, the demand-supply discrepancy can hopefully be redressed pretty soon.

Steps towards creating a self sufficient system

Business analysts can use machine learning algorithms to seek out patterns from large sets of data. This data can be structured – numerical, tables; semi structured – categorized text and images; unstructured – uncategorized, random images, texts, audios, videos. It was possible now to create a lot of value by tapping into the vast amount of data produced daily.

But these are all empty promises without a well thought out plan. And this is where the business analysts come into play.

Step 1

The analyst is presented with a situation or maybe a problem. He or she acquires data to start an investigation.

Step 2

The data passes through analytical models powered by machine learning algorithms. Certain patterns are recognized by the algorithms.

Step 3

The analyst visualizes the findings and draws some insights. The executive body is made aware of these insights.

Step 4

If the insights are found to be actionable they are acted upon.

The fourth step could lead to the integration of IPA systems into the company’s workflow.

An example of how this might work

Let us take a textile factory which produces cotton garments. The executives have seen a slight decline in their profits so they hire a business analyst to review the business processes.

The business analyst gathers up the historical, structured, accounts data. A simple visualization application helps her find out that there is a gradual, significant fall in the sales in a certain region, according to the reports presented by the distributors.

This gives the analyst something to work on. This instigates her to investigate the consumer bases further in order to understand the differences between the regions where profits are stable and the regions where it is declining. This brings up the need to analyze the consumer reviews and social media feeds. This is where she feels the need of AI powered analytics for the first time. She builds a case and the executives approve.

She comes to realize that the consumers in certain regions are experiencing difficulties in communicating with the website. The response time is more than usual. Their queries are often in broken English which the automated chat agent built in the system is failing to comprehend. Hence, the company needs better natural language processing on the website.

Lastly, the reports from the warehouse show an increment in costs due to delayed maintenance of a number of vehicles. This also brings up the necessity to install AI based alert systems which provide information on maintenance needs.

Significance of a holistic approach

Segmented or siloed usage of artificial intelligence rarely gets the job done. A skilled business analyst can ideate a holistic integration of AI, just as seen in the example. If a factory improves customer interaction through AI based response agents, it also needs to bring similar sort of efficiency to the supply chain. If the warehouse management system is upgraded with AI, the logistics need improvements too. Until and unless this holistic approach is adapted an organization can hardly call itself AI-Powered.

What a skilled business analyst can bring on to the table

A business analyst who is equally equipped in terms of technical skills and business acumen can bring positive changes in the business processes.

Recognizing problems

Finding gaps in business processes is one of the most revered skills found in a business analyst. Business analysts can point out possibilities of improvement in an apparently perfect workflow. This applies to any industry and any segment of that industry. This skill is not based on hunches, but a combination of data analyses and experience.

Bridging the gaps between operational and managerial teams

This gap often creates insurmountable obstacles in the path of a company’s growth. A business analyst with his highly effective communication skills and multidisciplinary knowledge can bridge this gap.

Improving client management

Business analysts are supposedly the best people to handle clients. They can bring forth the clearest picture of a certain product to the clients. They are also trained to cover all bases while demonstrating a service or understanding their requirements.

Creating a data oriented mindset around the company

Since the business analyst speaks all the tongues, (the operational and the managerial) he can influence others to adapt a data centric approach towards work. This is absolutely essential for a company that is trying to achieve the degree of efficiency which gives them a strong competitive edge.

Business analytics as a discipline can be a pretty hard nut to crack. But once you have learnt the art of acquiring new skills and adapting to new situations, chances are you will be invincible. A strong understanding of the business domain, along with a skill set that encompasses statistical analysis with R and Python, to visualization with Excel and Tableau, is what makes a great business analyst. And nothing short of that can lead an organization towards success with AI implementation.

Be the first to leave a comment
You must be Logged on to comment or reply to a post.