My name is Harpreet Singh. I'm going to talk about how do we leverage the sales data analytics to optimize the pharmaceutical drug launches. So I'm going to talk about some of the typical challenges we have, when we are talking about the drug launches, how data analytics, work as a solution. I'll also cover some of the aspects of the predictive modeling AI machine learning, in the scenarios. also we'll look at, some of the, cases where the real time analysis help with the insights. Also, we will, we'll cover some of the KPIs and then we talk about some of the advanced data architecture in the industry and then some of the best practices, for these analytics deployment. So I think to start with, I think the very first thing about when you think about any drug launch in the pharmaceutical industry. It takes a lot of time and efforts and resources if you think about an average one, Development of one product costs around 2. 6 billion dollars It provides you an exclusivity of few years So then I think the market timing and entry is critical also as a first product It gives you a competitive advantage in the in that therapeutic area Also, I think the the initial months are very critical because that's where 60 percent of the revenue is generated in like just first six months of the launch so For the companies it is very important then they optimize these launch strategies To make sure that they are in their highest revenue potential right and helping a lot of patients at the same time so now Let's talk about, we understand the challenges, but how do the data analytics, from the sales side help with that? So I think the sales data analytics, in the launch phase, for the product provide real time data insights, which help you to optimize in every step of the cycle, right? So for example, if you think about a forecasting. different predictive models can be used, to forecast of the product revenue and which are typically in a very good accuracy range. also, there's a lot of element of the real time data processing. There's a lot of rich data on the pharma side. So processing of those data also provide you a quick insights to react on the market events. Also based on this, We are able to penetrate a market potential or market share at a higher rate. So all these different elements of the analytics and data addresses these challenges and then we'll talk some of those in details in the next few slides. I think the very first thing is about when you think about the advanced analytics, predictive modeling, AI, machine learning, we do have a lot of data, right? So if you think about the data from the prescriber, from the doctors, patient level data, medical claims, hospitals buying product. So there's a lot of rich data. That's where a lot of, AI algorithm can be used to provide those insights and insights. Some of the capabilities of the predictive models can be related to how do you target the physician. So if you want to look at which doctors you need to target for the pre launch and the post launch phase, a lot of AI driven segmentation we created, then you can look at who are the high potential prescriber and then you can optimize the engagement for that. Also from the pricing point of view, there are different models available where you can use. multiple inputs and have Provide what should be the right price for the product and providing the competitive advantage at the same time I think when you think about these models, they have evolved a lot in last few years right now We are talking about 80 to 90 percent accuracy in these models. So we are getting a really good accuracy from these predictive models when you think about the insights Also, a lot of these models can also be used for the territory deployment and that provide an efficiency of at least 30 percent efficiency with the, with deploying those territories for the launches. I think the other important thing, when we think about a lot of data, you think about there's like millions of data points, also because that launch is so crucial. There's limited time frame when we have to react a company has to react on the market events So real time data processing definitely helps a lot. And then if you think about the real time processing Reduce the market response time from 15 days to like few days and like maybe three days, right? So these Definitely improved accuracy in forecasting. you can you are able to predict things more accurately also at the same time You Any kind of adjustment you need to make, in, in, in that launch strategy, you can do it like as less as like in 72 hours. so these are very critical things and timely things during the first six months of the launch. And then, before the launch and after the launch, there are a lot of KPIs which play a critical role, in understanding if the product is on a right path for the success. these KPIs are derived by different data analytics, pieces around it. So when you think about, I think we have some of the examples here, think about the conversion rate, right? So when you launch a product, how many doctors have actually prescribing this product? So definitely, I think the, if you can achieve 15, 20 percent conversion in first three months, that kind of correlate with 85 percent of the probability of you would achieve, your first year target. Also, when you think about the patients, the persistency on adherence to the product is also important, right? So how long a patient is staying on therapy for 3 months, 6 months, right? So if you have a good persistency, good adherency, you have a very high chances of product is getting a successful launch. So these early conversion rate, provides a momentum for the launch, for the commercial success and higher persistency definitely build that brand loyalty and market leadership at the end. so these benchmarks, these KPIs, you can track them on a weekly basis. You can track them on the monthly basis based on the objective of their brand launches. So now I think, There are different pieces, right? I think we talk about the AI, predictive models and then the real time, generation. also we talk about this plethora of data across doctors, hospitals, patients. The other important thing with that much of data, you also need, a very high processing speed, right? So we are talking about like petabytes of data. So now a lot of pharma companies are spending A lot of money on their infrastructure around the data warehouse right on the aws cloud based system. So to process this kind of data, right? So there's definitely a lot of cloud based warehouse that can handle like millions of transaction per second Once you set up these infrastructure on top of that We are implementing like python or r kind of a language which Provides you these models and these high accuracy Also hundreds of dashboards are built on top of that. So you have your data structure then your python and r and then you have a lot of dash dashboard tableau dashboard. You think about Power bi dashboard that also helps you to see how What's happening in the market and then these dashboard create like I would say hundreds of report for different customers based on their need like sales marketing finance etc I think with all of these things in place. and helping a successful launch. I think there are some best practices right across the industry, which we have seen. I think the first thing we just touched about, the data quality has to be there. It's garbage in, garbage out, right? If your data quality is not good, you will not have that highest, higher accuracy, right? And there, there are definitely a dedicated team in, in, in the pharma companies, which are working on the lot of validation process, right? To make sure that the millions of records what we are having has a higher accuracy and it's a good quality data they have a lot of different Framework on data governance, which provides you like continuous audit of this humongous data Also the other thing when you think about the best practices, there are a lot of different models Which requires resources and An effort to implement that so one of the Approaches maybe apply a pilot product, right? So for example, you want to have a prediction model which predict how many new patient will come into the market So instead of doing like a full fledged national level implementation, which would require millions of dollar You can start with a small territory, right? You start with one territory and test that model and Get that feedback from the sales and marketing and do you see a value out of it? So This phase approach definitely help you so you start with territory and then based on that You can apply that fine tune the model and apply throughout the nation so these Scalable implementation, definitely give you a lot of confidence and at the same time these models are built based on feedback, during the pilot phase The kpis we talk about so they're like at least 50 kpis pre launch and post launch With a higher accuracy that also provide a critical insights right throughout that and then if you have to do a course correction You can do it do that during that launch phase Also, I think that the model updates and then is are there any biases and the consistent performance tracking is also a critical piece I think with that maybe I think just A conclude all of these thing is that the integration of the sales state analytics and in pharmaceutical industry has proven to be transformative right and providing a substantial improvement in terms of the market penetration the resource optimization a successful launch so so definitely is solving a very complex challenge, right and then We talk about the development cost and then how compressing the launch window and then the traditional methods like previous methods were, I think it's helping we, it's helping us to overcome those challenges. And then advanced predictive modeling, AI, machine learning, these are, even accelerating these insights for the company and making sure that the launches are successful. So results, and I think the approaches are clear. The pharmaceutical companies are implementing these data driven strategies on an average of 30 23 percent improvement on launch trajectory. And then they see a higher market penetration and significantly reduced response time. with that, that's all from my side. Thank you.