jwinternheimer

# load libraries
library(dplyr)
library(buffer)
library(data.table)

# set working directory
setwd("/Users/julianwinternheimer/Downloads/subscriptions")

# list files
filenames <- list.files(pattern = "*country.csv")

jwinternheimer

September 2017 is set ot be a month of low growth for Buffer. We are expected to increase MRR by around 15K in the month of September, which is over 50% less than what was gained in the month of August. This growth will equate to a monthly percentage of less than 1%, which is lower than what we would expect.

In this exploration, I will try to identify factors leading to the decrease in net MRR gained. We will focus on Stripe MRR here, as it makes up the majority of our total MRR amount. It is worth noting that Apple has also experienced lower-than-expected growth (it has only gained a few hundred dollars in September as of September 25). We would still not expect growth this low from the Stripe gateway.

```{r include = FALSE, warning = FALSE, message = FALSE}
library(dplyr); library(ggplot2); library(tidyr); library(lubridate); library(buffer)
```

### Gather data
We'll use the [new Stripe MRR breakdown script](https://github.com/bufferapp/buffer-analysis/blob/master/deliverables/mrr-breakdown.md) to analyz

jwinternheimer

library(httr); library(jsonlite); library(buffer); library(DBI)

# Make the request
req <- GET("https://openexchangerates.org/api/latest.json?app_id=xxxxxxxxxxxxxxxxxx")

# Get the status
http_status(req)

# Get the content in the request
content <- content(req)

jwinternheimer

library(twitteR); library(wordcloud); library(tm); library(RColorBrewer); library(praise)

# Twitter Oauth
my_key <- "xxxxxxxxxxxxxxxxxxxxxxxxx"
my_secret <- "xxxxxxxxxxxxxxxxxxxxxxxxx"
access_token <- "xxxxxxxxxxxxxxxxxxxxxxxxx"
access_secret <- "xxxxxxxxxxxxxxxxxxxxxxxxx"

setup_twitter_oauth(my_key,my_secret)

jwinternheimer

library(data.table); library(dplyr); library(tidyr); library(ggplot2); library(xts)

## Import and Tidy
stripe <- read.table('~/Downloads/stripe.csv',sep=',',header=T)
stripe$start_date <- as.Date(as.POSIXlt(stripe$start, format="%Y-%m-%d %H:%M:%S"))
stripe$end_date <- as.Date(as.POSIXlt(stripe$ended_at, format="%Y-%m-%d %H:%M:%S"))

## Aggregate Awesome Start Dates and Count Users
awesome_start_date <- stripe %>%
  filter(plan_id == 'pro-annual' | plan_id == 'pro-monthly' | plan_id == 'awesome') %>%

jwinternheimer

## Import Data into Data Frame
awesome <- read.csv('~/Downloads/awesome.csv',header=T)

## Rename columns of data frame
names(awesome) <- c('user_id','plan_date','join_date','activated','used_extension_first_day',
                         'used_extension_first_week','used_extension_first_month',
                         'prompts_seen')

## Change data to 'date' type 
awesome$join_date <- as.Date(awesome$join_date)

jwinternheimer

library(data.table); library(dplyr); library(tidyr); library(ggplot2);library(NbClust)

## Read Data
data <- read.table("~/Downloads/business_users.csv",sep=",",header=T)
names(data) <- c("user_id","team_members","profiles","months_paying","plan_days","updates","updates_per_day_per_profile")

## Prep Data
data[is.na(data)] <- 0

## Separate Variables We're Interested In

jwinternheimer

library(tm);library(SnowballC); library(wordcloud); library(RColorBrewer);library(RWeka)
library(ape)

## Read Text File of Conversations
get_text <- function(filename) {
  txt <- read.table(filename, header=F)
  names(txt) <- "text"
  
  txt <- as.data.frame(clean.text(txt$text))
  names(txt) <- "text"

jwinternheimer

library(data.table); library(dplyr); library(tidyr)
library(ggplot2); library(scales); library(grid); library(RColorBrewer)


## Import Dataset
updates <- read.csv("~/Downloads/updates.csv",header=T)
names(updates) <- c("update_id","profile_id","num_chars","num_explanation","num_hashtags",
                    "hour_sent","num_followers","follower_tier","link","link_no_content",
                    "photo","text","interactions","retweets","favorites","mentions",
                    "clicks","avg_interaction_length")

jwinternheimer

library(data.table);
library(dplyr);
library(tidyr);


cleanGoogleTable <- function(dat, table=1, skip=0, ncols=NA, nrows=-1, header=TRUE, dropFirstCol=NA){
      if(!is.data.frame(dat)){
              dat <- dat[[table]]
  }
  if(is.na(dropFirstCol)) {