Q&A with Ashwin Gopalakrishnan: Challenges with AI Adoption in Large Organizations

Ashwin Gopalakrishnan recently joined MathCo. as a Senior Partner in the United States. Ashwin is a seasoned professional with a specialization in analytics, strategic marketing and product management. He has over a decade’s experience in leading high-growth initiatives, building analytics teams, and developing an organization’s analytics capabilities. In this Q&A, he shares his insights on common industry-agnostic challenges for making the most of analytics consumption.

Question 1: There is a statement often heard in the industry that “Asking the right question is imperative to solving business problems.” Do you subscribe to this belief?

Ashwin: Yes, I do subscribe to the belief that data won’t speak unless you ask it the right questions. However, this is not the only gap. An emerging trend that has persisted for years now, is that a lot of analysts’ time is spent in generating the right kind of data.

Data management and access capabilities are still evolving and many companies continue to be in the nascent stages of maturity and lack quality data to work with. The gap will persist unless there is a shift from the prescriptive approach of employing data science as an antidote for siloed problems. There needs to be active, intra-organizational dialogue to promote a comprehensive data-science driven culture. Only then can data maturity be comprehensively integrated across processes, as opposed to siloed efforts to bring about organizational change.

Question 2: What challenges do you notice in the context of usage and adoption of AI in large organizations?

Ashwin: It is concerning that AI is still considered as a niche solutioning by many.

To this day, many leaders leverage analytics capabilities solely to corroborate their “gut feelings” on business drivers and trends, indicating that there is a concerning dearth of AI evangelization. And fractional AI adoption is an inescapable repercussion of this mindset.

Additionally, companies may not leverage AI solutions proportionately across all their functions. Among pharma companies as an example, there is ubiquitous AI adoption within R&D teams for swift drug discovery, testing and improved data processing. Similarly, AI is readily employed at scale in pharma manufacturing for process automation, asset/factory floor maintenance and quality control. However, functions like finance, sales & marketing, supply chain, still harbor skepticism when it comes to institutionalizing AI.

Therefore, AI adoption can only be catalyzed when AI ceases to be regarded as a premier solutioning.

Question 3: If AI acceptance is still at an emerging stage, what about the myriad of AI-related trends and technologies that come up each year? How can companies figure which trend to bank on?

In my opinion, the only trend that has stood the test of time, and will continue to ‘generate buzz,’ is evolved analytics maturity. I will not deny that there is sustained awareness about the importance of data and analytics, given that these continue to be featured as critical value creators, enterprise assets and crucial competencies in leading corporate strategies.

However, the reality is that many companies are yet to attain data and analytics maturity. Even companies that have copious amounts of data at their disposal are unable to generate rich insights.

The banking sector aptly exemplifies this. While the sector ranks among the highest when it comes to industry-specific data generation, they have among the lowest rates of analytics and BI adoption – less than 30%! [1]

Without critical operational maturity, something as fundamental as data science would remain a buzzword even a decade from now and operational processes will still be run conventionally, and at sub-optimal levels.

Therefore, companies would do better to not get derailed by the latest buzzwords, as they are far too many and terribly transient, and ultimately only point to the need for analytics maturity.

Question 4: Given your experience in the pharmaceutical industry, does the same hold for the buzzword ‘Pharma 4.0’ as well?

If one looks up the term Pharma 4.0, they will find that it refers to adept, digitally transformed pharma manufacturing processes, i.e., processes that are more streamlined, compliant with industry norms, and quick to respond to unforeseen disruption. The concept emphasizes on digitization and automation of manufacturing and supply chain processes, thereby enabling swift turnarounds, reduced margins of error, enhanced predictive maintenance capabilities, improved product quality, reduced risk, better business understanding, proactive identification of new opportunities and much more.

While the concept of Pharma 4.0 might be well etched out, the maturity and adoption levels of organizations looking to embrace the same, might vary. There is still a lot more untapped potential in utilizing Pharma 4.0 for expanding the use of digital tools and AI powered solutions across organizations. These abilities can only be tapped into by digitally transformed firms. Those with limited digital capabilities are likely to lose out on an enviable competitive edge.

Question 5: Do you think pandemic disruption has also perhaps changed the way executives view pre-existing, conventional processes? How has it altered the ways of working in the pharma and healthcare industry?

At the start of the COVID-19 pandemic, there was an initial shock to the system. Different operational teams within the healthcare industry reacted and adapted differently. For example, while manufacturing and supply chain processes adapted really well, executives focused their efforts on solving demand problems.

Overall, the healthcare industry made substantial strides with regard to communication capabilities, especially between patients/caregivers and physicians. Traditional in-person visits drastically reduced and tele-interactions increased. There was also a growing need to prioritize patients that needed care. Additionally, the general fear of misinformation led to patients with relatively less severe medical conditions abandoning medication or not adhering to the needed treatment.

What was also observed in the healthcare industry, was a spike in digital transformation investments as stakeholders strived to bridge these operational gaps. Catalyzed digital transformation resulted in organically improved analytics capabilities, and as a result, these stakeholders were better prepared for further AI adoption.

Question 6: Finally, shifting focus to your career – you have worked in healthcare marketing for a long time now. What has prompted you to make a career change to analytics consulting & focusing on AI/ML based solutions?

In the last few years, I have witnessed how the healthcare industry has been pushing the envelope by enhancing data and analytics capabilities like a CPG firm or a retailer would, focusing on engaging customers better, tailoring offerings to their needs instead of a one-size-fits-all approach and creating maximum value for users.

My career shift, therefore, could not have been better timed. It bodes really well that MathCo.’s intrinsic way of working is to customize solutions with pertinent AI and ML tools – to help customers discover what is best for them to drive decision making and improve their overall analytical maturity.

By building custom analytics products for businesses across industries and their specific pain points, I think that MathCo. is truly enabling robust, end-to-end analytics journeys - right from the stage of conceptualization until consumption. This next gen problem-solving approach is key to ensuring that the future is filled with innovative, industry-agnostic processes, that revolutionize our ways of working and help customers derive immense value.

I am sure that the experience will facilitate a wide variety of learning – along with tremendous fulfilment when analytics products truly get mobilized, and fuel data into decisions that herald in next-gen analytics capabilities.

Senior Partner, Region Head, TheMathCompany

Ashwin Gopalakrishnan