High Note case StudyEquation PCit f Dit Bit Lit ?WijYjt –

High Note case Study

Equation: P(Cit) = f (Dit, Bit, Lit, ?WijYjt) – this equation is used to determine the probability of the conversion for a customer its simple called network equation

Cit = Probability of conversion

Dit = Demographic Features

Bit = Behavioral Features

Lit = Location Features

WijYjt = Social Network Features

According to our case study, I declared below variables are used accordingly

Cit =” Number of subscribers conversion to premium service”

Dit =” Age” “Gender”

Bit =” Number of songs” “Loved tracks”

Whereas I refer to the individual person number and t refer to the time period

Defining the objective is the primary goal of predictive modeling by lining up the function of the model with business goals & business strategy.

My focus to convert Free users to paid users. At first, I need to analyze the data in stages.

In figure 1: Which age bracket we need to target, according to given data it must be between 18-25 age group.

Figure 2: Maximum count of songs hear in a given period of time is 10 songs.

In Figure 3: Percentage of users enrolled for at least one loved track is nearly 38%. The next 2 to 5 loved tracks are nearly 20% (both free and premium subscribers).

In Figure 4 we need to analyze the relationship between two variables X and Y whereas X refers to no of songs and Y refers to Age. Here Y is the dependent variable and X is the Independent Variable. According to the given data, the Age 20 category has enrolled more than 800 songs in their playlist. By applying linear regression test the connection between two variables can be abstract visually in a scatter plot graph by using R.

Get quality help now

Dr. Karlyna PhD

Verified

Proficient in: Communication

4.7 (235)

“ Amazing writer! I am really satisfied with her work. An excellent price as well. ”

+84 relevant experts are online

Hire writer

Each of the dots in this scatterplot might represent a connection between a user, age, and number of songs that they pay attention to. By using trendline, we can elaborate the connection between X and Y. With the help of Linear regression test it draws the line through the two variables, the scatterplot graph of the two variables, that reduce the error or the distance between the aggregation of all y values for any x values. So, all of these distances would be minimized by the line drawn by a regression. It is very important to collect accurate data for reducing the margin of errors. By using regression maybe males who are willing to convert premium subscribers rather than females according to fig 6. Why would I waste my ad budget by showing sale ads to the Female category? Analysis comparing Adopters and Nonadopters according to fig 5 and 7 stating that value 1 on X variable has registered higher no of songs and playlists rather than value 0.

Due to the data collected on subscriber demographics, behavior/preferences by knowing past listening behavior, time of day and location. Through continue investment in data collection through machine learning algorithms. These also help to decode customers purchasing behavior and interaction to the website/app (High note). By using segmentation to the target market in different ways to be successful.

Demographic information may authorize precise targeting, which may make for more successful campaigns. Secondly demographic profile of selected customers may not be characteristic of the total people of prospect adopters. Hence, we study local average treatment effects among those customer segments in our sample that are likely to adopt playlist.

When a consumer is on our site by using cookies, we can use their browsing history and past behavioral patterns to display the content. We assess whether nonadopters display the same listening behavior as adopters earlier to adoption. Multiple behavioral and demographic characteristics, it is more likely to have basic differences in the composition of these groups that could clearly explain differences in their behavior. I recommend using quasi-experimental methods. I would recommend providing an additional feature for premium service. By encouraging premium users to recommend/add friends and families. By introducing a new referral program to motivate users to create more playlists. My findings are by reducing the free offering time period to new users. To make sure customers are fully aware of premium membership. Referral marketing helps me to get the value of acquisitions and conversion by rewarding existing customers and generate new customers. Social app marketing is best for targeting people and also relevant to the high note industry it performs well in user quality, conversion, and volume. I prefer Regression analysis was carried out on the characteristics and behaviors of existing high note users showed that conversions to premium service were associated with the following variables. No subscriber friends: if a user has more friends as a subscriber, the user is more likely to be a premium user. Number of playlists: if a user more playlist added into his account, we can count the user chances to be a premium user. The basis on our analysis and proper testing we need to figure out which customers need to be targeted and also implement marketing campaigns on different channels. Likewise, email and social media are the best ways to target new customers. The main thing freemium service to attract new users. If I can generate more traffic towards high note by using different tools in the market and the chances to upgrade to premium will be higher in rate.

I recommend 4 strategies to increase the chances of success in business. The main theme using a freemium model to minimize the customer acquisition cost. Freemium models will create a nurturing process where users can check into our new product without connecting to the marketing or sales team. Predictive analytics assist marketing campaigns to be added customer focus by tailoring marketing activities to target only certain segments. By designing a predictive model to regulate the future actions and responses of customers that are likely to be transform.