Column

$ per lead (lead model: charged on demand)

40

$ per month (community model: fixed rate)

400

% of yearly revenue increased by lead model by 2018

% of toatl community increased by lead model by 2018

Column

Monthly Revenue 2016 - 2020

Lead model revenue growth exceeds community model by the end of July 2017

Pros and Cons

Lead model

  • Pros: Fast revenue growth in the long run; enriched content
  • Cons: High fragility

Community model

  • Pros: Stable revenue growth; benefical for big marketers
  • Cons: Slow revenue growth

Recommended Actions

  • Adopt lead model for the next 3 months
  • Real time monitoring using A/B testing
  • Develope alternative mixture model
---
title: "Pros and Cons Analysis - Lead Model Vs. Community Model"
author: "Cassie Guo"
output: 
  flexdashboard::flex_dashboard:
    social: menu
    source_code: embed
---

```{r setup, include=FALSE}
library(ggplot2)
library(plotly)
library(plyr)
library(flexdashboard)
library(dplyr)
library(sqldf)
library(lubridate)
set.seed(955)
q2 <- read.csv('q2.csv', header = T) %>% as_tibble()
q1 <- read.csv('q1.csv', header = T) %>% as_tibble()
```

Column {data-width=200}
-----------------------------------------------------------------------
### $ per lead (lead model: charged on demand)

```{r}
valueBox(40, icon = 'fa-comments')
```


### $ per month (community model: fixed rate)

```{r}
valueBox(400, icon = 'fa-home')
```


### % of yearly revenue increased by lead model by 2018

```{r}
gauge(33, min = 0, max = 100, symbol = '%', gaugeSectors(
  success = c(0, 2), warning = c(3, 6), danger = c(7, 10)
))
```

### % of toatl community increased by lead model by 2018

```{r}
gauge(14, min = 0, max = 100, symbol = '%', gaugeSectors(
   success = c(0, 2), warning = c(3, 6), danger = c(7, 10)
))
```

Column {data-width=600}
-----------------------------------------------------------------------
### Monthly Revenue 2016 - 2020

Lead model revenue growth exceeds community model by the end of July 2017

```{r}

# p <- ggplot(dat, aes(x=xvar, y=yvar)) +
#             geom_point(shape=1) +    # Use hollow circles
#             geom_smooth(method=lm)   # Add linear regression line
# ggplotly(p)
q3 <- sqldf('SELECT avg(value) as avg, mdl, ym from q2 group by ym, mdl')
q3$ym <- as.Date(q3$ym)
colnames(q3) <- c('Monthly_revenue', 'Model', 'Year_month')
levels(q3$Model) <- c('Community', 'Lead')
p <- ggplot(q3, aes(x=Year_month, y=Monthly_revenue, group = Model, color = Model)) +
  geom_smooth(color = 'grey40') +
  xlab('year-month') +
  ylab('Revenue in dollar') + 
  geom_point(size = 2) + theme_minimal()
  # Use hollow circles
ggplotly(p)
```

### Pros and Cons

Lead model

* Pros: Fast revenue growth in the long run; enriched content
* Cons: High fragility

Community model

* Pros: Stable revenue growth; benefical for big marketers
* Cons: Slow revenue growth

**Recommended Actions**

* Adopt lead model for the next 3 months 
* Real time monitoring using A/B testing
* Develope alternative mixture model