Data Science in CPG or FMCG Industry || Retail Analytics

hi in this video I'm going to talk about how detangle tax can be use in the CPG or FMC's industry but the FMC's industry produces a large amount of data but often we do not use the data for decision-making and those that use the data for decision-making are far ahead of their competitors well in offenses industry we have small retailers as well as we have big retailers and oftentimes we the big retailers are able to have nice the data that they have because they have money to invest in IT infrastructure and so on but the smaller retailers are unable to do so but they can do it they can obviously harness the power of data because in the f-insas industry we actually see a lot of data even in small retailers so that's always important because without having a lot of those update and it is cannot be use and FMC's industry irrespective of the size of the retailer you still have a lot of data to use Tim Leon as I've said CPG industry produces a large volume of data so we can actually use the data for a number of purposes for example prediction things visualizing data for decision-making automating ting automating decision-making and so on the problem is that the gut best decision-making doesn't quite work in the CPG in the defense industry because the volume of data is just used and high dimension decision-making is extremely difficult for human beings so only the machines can make better decisions so in FM's is in the industry it is always important to make sure that you use data Robin got this so what are the use cases that we see where data and it is can be used in the CPC or consisted of it can will be used in the demand forecasting can will be used both in the supply side or as well as in the demand side we can use time series modelling whether other linear regression type modeling to focus demand and that can be automated so with the pressure button you can see what is going to be the demand for a given product tomorrow or a month from or two months from now so we can have both short-term forecasting and long-term forecasting that can will be done using the data that we have they store the data and then using the model you always forecasted and then you automated is your IT system so that the people who are in the sales department they can sink will press a button and get the focused value and then use that decision-making planning and all kinds of things it can also be used in product pricing well if you remember the classical microeconomics demand so using the demand and supply curve we need to decide what should be the ideal price and there are different things to look at it we need to see how price is changing with demand I would demand it changing with price because they both relate to each other and that can be done using the data that you have because in different point in time you have different prices of the same product and you see what's the demand and you can actually see what price you actually optimize your revenue and that's the price one should go for so using data I can always analyze the market competition competition with your own products so you happen only distilled it I will tell you how can optimize on that store assortment so this is more about the invention analytics so we allocate inventory more efficiently by using data and that's a big challenge you can use operation t-shirts to do that you can also use machine learning algorithms to do that what happening currently is that most of the ERP systems whether it is Oracle a.s.a.p and so on do not quite have the sophisticated of forecasting optimization algorithms they are building a lot of structure around the current system but it's still not very efficient as much efficient as you see the ml or AI system in other industries but it's developing alright the smaller companies retailers do not have access to you know sophisticated ERP system such as Oracle or and the local AFV system do not support forecasting or optimization they do not have the ML routines or the operation research routines or the packages hence it's a bit of a challenge but if you have developers web data scientists you can always develop your own forecasting algorithms and it can also be used for personalized recommendations is very popular in the online retail world as you can see it seems heavily in the e-commerce industry but it also be using offline while often this is underestimated which not quite to realize even in the offline mode you can use data to recommend products what might does it and many offline companies do use the data for recommendation but the smaller retailers do not get in the retail industry so it's very easy to build subsystems and you can always use the advanced machine learning and data science algorithms and now this you know with the availability of many open source library and also you know the cloud and all kinds of algorithm already implemented in the cloud services whether it's Amazon whether its Microsoft you know it's easy to develop and implement personalized recommendation engine then we can also use data analytics for cross-selling and upselling those who are familiar with the retail industry this is a familiar words both in the online as well as offline world more in the online world less in the offline world but it can will be implemented in the end role as well well actually use analytics to see if you can cross-sell or upsell products and you can customize different kinds of discount offers based on the customer data that you have on the customer behavior that you understand from the data and can be done to get a visualization – you know lots of basic detangles is like for this linear regression where it is clustering that kind of and those who are not familiar with cross-selling upselling cross-selling is basically you know selling a complementary product or a different product which is related to a product but the customer is buying so it's product a and product to me so product is need to the car please so somebody grows product a you also sell product B to them and that's absolutely on the other hand is simply selling different words and of the same type of product okay both are very profitable for the tellers retail it is can also be used for different types of management decisions whether it's product development whether it is product launch in revenue optimization and so product development is important thing nowadays in the e-commerce company so it's very well use in online yule but not often use offline water based in customer behavior launching is very important we use the a/b testing right for product launch in to see what actually works better for you be testing can will be used product development as well so many e-commerce companies use them and all kinds of it errors actually can use a be tested in order to come up with better decision managers co-management can use advanced analytics as well better decisions revenue optimization is very important or how we can actually optimize revenue using the data so lots of ready-made or built-in algorithms are available for for you to use them to optimize in revenue so that can also be used what are the current challenges in the industry in the C key industrial do not actually know much about data analytics so there's a lack of awareness so notice the senior managers when the mid-level managers or beginning managers do not quite know they're not know much about the impact it makes in the revenue and the profit okay so that's one thing the current IT systems the ERP systems that companies have been using for decades now they do not support advanced analytics machine learning or artificial intelligence and the cost of implementing these algorithms also quite huge and that's one big reason why companies do not go for it or only the big companies go for big retailers go for ai and Emily in between the ERP system but nor the smaller ones and the problem with the data governance right so unlike in finance unlike in any other industry the data governance is is very bad in the retail industry and the data that they have is not a good quality to be usable for return with it so that's one area where little companies are also paying the best so what the future looks like for CPG industry rotate analytics well contribute data will be used or it will be the center of decision-making for slowing future and retailers that we use data for decision-making will have a serious advantage over for the competitors that's what I believe automated decision-making is an important area where analytics will play a big role because it's for big companies it's very difficult that management will decide everything like it has to be more automatic it has to be more scientific so decision-making will be very automated and through IT systems especially for the small ticket clients you do not lose mass by making the small instead so automatic decision making will be given preference over so there will be no waiting time for analysts or sales people to date addition it will be purely automatic so nobody then micro second edition will be made tech will be the center of innovation in CPC that's already happening but it's going to happen even more

One Comment

  1. Norwall Music said:

    Love the science videos! 👍🔬📐

    July 28, 2019

Leave a Reply

Your email address will not be published. Required fields are marked *