Introduction to Dataset:
The dataset that I have choose to work on is related to startup ecosystem. The word startup has become so popular in he last 5 years that everyone knows about them which is very obvious because of the digitization and the technological advancements and people adapting to the modern way of life. Unlike the olden days hardships, humans are leading a very comfortable life. Everything is at their fingertip and if there is anything they feel short of then there comes a startup to solve that problem and help us to live a very comfortable life.
There are many companies that have become a part of our day to day life. It is very much likely that some have become very synonymous that we are using the startup name more than the actual word just like how google replaced the term search on the net.
My dataset talks about a fraction of those startups that are prominent in India but also has its market presence outside of India.
I have found the dataset on kaggle.
What my dataset contains?
My dataset contains 4350 rows and 8 columns.
Date: Time period in the date format (DD/MM/YYYY)
StartupName: Name of the Startup
IndustryVertical: Which industry the startup is working in
SubVertical: Which sub-industry the industry is categorized
CityLocation: where the startup is headquartered
Investors: Who invested in the startup
InvestmentType: What type of investement it was
AmountUSD: How much amount the startup raised in USD
What I want to do with the dataset?
I am looking to understand certain patterns from my analysis and answer the below questions:
- Which startup received the most funding?
- Which Industry Vertical received th most funding?
- Does location really matter to raise funding?
- Which type of funding is most prominent among startups?
input <- read.csv("C:/Users/gunde/Downloads/Indian_Startup_Funding.csv",stringsAsFactors = TRUE)
startup <- data.frame(input)
glimpse(startup)
Rows: 3,044
Columns: 8
$ Date <fct> 09/01/2020, 13/01/2020, 09/01/2020, 02/01/2~
$ StartupName <fct> BYJUS, Shuttl, Mamaearth, WealthBucket, Fas~
$ IndustryVertical <fct> "E-Tech", "Transportation", "E-commerce", "~
$ SubVertical <fct> "E-learning", "App based shuttle service", ~
$ CityLocation <fct> "Bangalore", "Gurgaon", "Bangalore", "New D~
$ Investors <fct> "Tiger Global Management", "Susquehanna Gro~
$ InvestmentType <fct> Private Equity Round, Series C, Series B, P~
$ AmountUSD <fct> "20,00,00,000", "80,48,394", "1,83,58,860",~ Date StartupName IndustryVertical SubVertical
0 0 0 0
CityLocation Investors InvestmentType AmountUSD
0 0 0 0 glimpse(startup)
Rows: 3,044
Columns: 8
$ Date <fct> 09/01/2020, 13/01/2020, 09/01/2020, 02/01/2~
$ StartupName <fct> BYJUS, Shuttl, Mamaearth, WealthBucket, Fas~
$ IndustryVertical <fct> "E-Tech", "Transportation", "E-commerce", "~
$ SubVertical <fct> "E-learning", "App based shuttle service", ~
$ CityLocation <fct> "Bangalore", "Gurgaon", "Bangalore", "New D~
$ Investors <fct> "Tiger Global Management", "Susquehanna Gro~
$ InvestmentType <fct> Private Equity Round, Series C, Series B, P~
$ AmountUSD <fct> "20,00,00,000", "80,48,394", "1,83,58,860",~startup$AmountUSD <- as.numeric(gsub(",","",startup$AmountUSD))
Date StartupName IndustryVertical SubVertical
0 0 0 0
CityLocation Investors InvestmentType AmountUSD
0 0 0 0 ## Convert Date variable to Date format
startup$Date <- as.Date(startup$Date,format("%d/%m/%Y"))
## Confirm
str(startup)
'data.frame': 3044 obs. of 8 variables:
$ Date : Date, format: "2020-01-09" ...
$ StartupName : Factor w/ 2453 levels "#Fame","121Policy",..: 276 1911 1308 2307 656 1579 2441 568 295 495 ...
$ IndustryVertical: Factor w/ 878 levels "360-degree view creating platform",..: 200 827 190 275 254 440 358 822 191 7 ...
$ SubVertical : Factor w/ 1943 levels "\"Women\\\\'s Fashion Clothing Online Platform\"",..: 461 75 1647 1279 515 1487 1216 28 112 1681 ...
$ CityLocation : Factor w/ 100 levels "Agra","Ahemadabad",..: 5 29 5 62 55 19 29 74 29 5 ...
$ Investors : Factor w/ 2405 levels " Sandeep Aggarwal, Teruhide Sato",..: 2114 2061 1857 2316 1997 468 218 1827 1523 1365 ...
$ InvestmentType : Factor w/ 55 levels "Angel","Angel / Seed Funding",..: 26 43 41 21 36 40 26 40 44 30 ...
$ AmountUSD : num 2.00e+08 8.05e+06 1.84e+07 3.00e+06 1.80e+06 ...## Make a new column for year
startup$year <- as.numeric(format(startup$Date,"%Y"))
## Year frequency table
yeartable <- table(startup$year)
yeartable
2015 2016 2017 2018 2019 2020
931 993 687 309 111 7 ## Visualization using Bar chart
library(plotrix)
pie(yeartable,labels = yeartable, edges=10, main = "No. of startup's funder each year")
The pie graph shows the no of startups funded to ratio over the year from 2015 to 2020
Ola Cabs Swiggy BYJUS Paytm Medinfi
8 8 7 7 6
ZuperMeal Zuppler Zuver Zwayam Zzungry
1 1 1 1 1 #Creating dataframe
i3 <- data.frame(category = c("Consumer Internet","Technology","E-Commerce","Healthcare"),
count = c(942,478,287,72))
#Calculating percentages
i3$fraction <- i3$count / sum(i3$count)
i3$ymax <- cumsum(i3$fraction)
i3$ymin <- c(0, head(i3$ymax, n=-1))
i3$labelPosition <- (i3$ymax + i3$ymin) / 2
i3$label <- paste0(i3$category, "\n value: ", i3$count)
# Make the plot
ggplot(i3, aes(ymax=ymax, ymin=ymin, xmax=4, xmin=3, fill=category)) +
geom_rect() +
geom_label( x=3, aes(y=labelPosition, label=label), size=4) +
scale_fill_brewer(palette=4) +
coord_polar(theta="y") +
xlim(c(1, 5)) +
theme_void()
##Industry vertical frequency table
industrytable_tail <- tail(sort(table(startup$IndustryVertical), decreasing=TRUE),25)
industrytable_tail
Video Games
1
Video Intelligence Platform
1
Video Streaming
1
Viral Content web Platform
1
Virtal
1
Virtual Health consultation app
1
Virtual Reality activity based learning platform
1
Virtual Reality Headset creator
1
virtual reality, 3d simulation and stereoscopic products
1
Visual Blogging platform
1
visual search and discovery platform
1
Voice Call incentivization mobile app
1
Warranty Programs Service Administration
1
Waste Management Solutions
1
Wealth Management Platform
1
Web Content Publishing
1
Wedding Planning Platform
1
Wedding Venues & Vendors Marketplace
1
Weight Management Service
1
WiFi first Cloud communication platform
1
wine
1
Women Ethnic Wear Online Marketplace
1
Women Lifestyle Marketplace
1
Womens Fashion Wear Portal
1
Workforce Management Software
1 ## Visualization using Bar chart
set.seed(5642)
sample_data <- data.frame(name = c("Consumer Internet","Technology","E-Commerce","Healthcare","Finance","Logistics","Saas","Education","Food & Beverage","FinTech") ,
value = c(942,478,287,72,63,32,28,25,23,19))
# Create bar plot with labels
plot<-ggplot(sample_data,
aes(name,value)) +
geom_bar(stat = "identity")+ theme_minimal()+
geom_text(aes(label = signif(value)), nudge_y = 3,)
plot+
coord_flip()
This diagram clearly shows us that there are many startups working in the consumer Internet followed by technology, e-commerce, finance, transportation.
If we compare the funding raised it has a very different story all together:
e-commerce start ups are 1/3 in number to consumer internet by they raised almost 4x amounts than consumer internet on average.
The transportation Industry clearly stands out. This clearly explains the supply chain issues.
Cleanstartup <- startup[complete.cases(startup), ]
head(Cleanstartup)
Date StartupName IndustryVertical
1 2020-01-09 BYJUS E-Tech
2 2020-01-13 Shuttl Transportation
3 2020-01-09 Mamaearth E-commerce
4 2020-01-02 WealthBucket FinTech
5 2020-01-02 Fashor Fashion and Apparel
6 2020-01-13 Pando Logistics
SubVertical CityLocation
1 E-learning Bangalore
2 App based shuttle service Gurgaon
3 Retailer of baby and toddler products Bangalore
4 Online Investment New Delhi
5 Embroiled Clothes For Women Mumbai
6 Open-market, freight management platform Chennai
Investors InvestmentType AmountUSD year
1 Tiger Global Management Private Equity Round 200000000 2020
2 Susquehanna Growth Equity Series C 8048394 2020
3 Sequoia Capital India Series B 18358860 2020
4 Vinod Khatumal Pre-series A 3000000 2020
5 Sprout Venture Partners Seed Round 1800000 2020
6 Chiratae Ventures Series A 9000000 2020dim(Cleanstartup)
[1] 3038 9summary(Cleanstartup)
Date StartupName IndustryVertical
Min. :2015-01-02 Ola Cabs: 8 Consumer Internet: 941
1st Qu.:2015-11-03 Swiggy : 8 Technology : 478
Median :2016-07-15 BYJUS : 7 E-Commerce : 287
Mean :2016-09-22 Paytm : 7 Healthcare : 72
3rd Qu.:2017-06-12 Medinfi : 6 Finance : 63
Max. :2020-01-13 Meesho : 6 Logistics : 32
(Other) :2996 (Other) :1165
SubVertical CityLocation
nan : 931 Bangalore:863
Online Lending Platform : 11 Mumbai :609
Online Pharmacy : 10 New Delhi:424
Food Delivery Platform : 8 Gurgaon :344
Education : 5 Hyderabad:164
Online Education Platform: 5 Chennai :142
(Other) :2068 (Other) :492
Investors InvestmentType
Undisclosed Investors: 104 Private Equity :1355
Ratan Tata : 25 Seed Funding :1350
Indian Angel Network : 24 Seed/ Angel Funding : 60
Shell Foundation : 21 Seed / Angel Funding: 47
Kalaari Capital : 16 Seed\\\\nFunding : 30
Sequoia Capital : 15 Debt Funding : 25
(Other) :2833 (Other) : 171
AmountUSD year
Min. :1.00e+04 Min. :2015
1st Qu.:1.00e+05 1st Qu.:2015
Median :5.00e+05 Median :2016
Mean :1.26e+07 Mean :2016
3rd Qu.:4.00e+06 3rd Qu.:2017
Max. :3.90e+09 Max. :2020
Cleanstartup1 <- Cleanstartup
str(Cleanstartup1)
'data.frame': 3038 obs. of 9 variables:
$ Date : Date, format: "2020-01-09" ...
$ StartupName : Factor w/ 2453 levels "#Fame","121Policy",..: 276 1911 1308 2307 656 1579 2441 568 295 495 ...
$ IndustryVertical: Factor w/ 878 levels "360-degree view creating platform",..: 200 827 190 275 254 440 358 822 191 7 ...
$ SubVertical : Factor w/ 1943 levels "\"Women\\\\'s Fashion Clothing Online Platform\"",..: 461 75 1647 1279 515 1487 1216 28 112 1681 ...
$ CityLocation : Factor w/ 100 levels "Agra","Ahemadabad",..: 5 29 5 62 55 19 29 74 29 5 ...
$ Investors : Factor w/ 2405 levels " Sandeep Aggarwal, Teruhide Sato",..: 2114 2061 1857 2316 1997 468 218 1827 1523 1365 ...
$ InvestmentType : Factor w/ 55 levels "Angel","Angel / Seed Funding",..: 26 43 41 21 36 40 26 40 44 30 ...
$ AmountUSD : num 2.00e+08 8.05e+06 1.84e+07 3.00e+06 1.80e+06 ...
$ year : num 2020 2020 2020 2020 2020 ...Cleanstartup1$CityLocation <- as.character(Cleanstartup1$CityLocation)
Cleanstartup1$CityLocation[Cleanstartup1$CityLocation != "Bangalore" & Cleanstartup1$CityLocation != "Mumbai" & Cleanstartup1$CityLocation != "New Delhi" & Cleanstartup1$CityLocation != "Gurgaon" & Cleanstartup1$CityLocation != "Pune" & Cleanstartup1$CityLocation != "Hyderabad"] <- "Others"
head(Cleanstartup1)
Date StartupName IndustryVertical
1 2020-01-09 BYJUS E-Tech
2 2020-01-13 Shuttl Transportation
3 2020-01-09 Mamaearth E-commerce
4 2020-01-02 WealthBucket FinTech
5 2020-01-02 Fashor Fashion and Apparel
6 2020-01-13 Pando Logistics
SubVertical CityLocation
1 E-learning Bangalore
2 App based shuttle service Gurgaon
3 Retailer of baby and toddler products Bangalore
4 Online Investment New Delhi
5 Embroiled Clothes For Women Mumbai
6 Open-market, freight management platform Others
Investors InvestmentType AmountUSD year
1 Tiger Global Management Private Equity Round 200000000 2020
2 Susquehanna Growth Equity Series C 8048394 2020
3 Sequoia Capital India Series B 18358860 2020
4 Vinod Khatumal Pre-series A 3000000 2020
5 Sprout Venture Partners Seed Round 1800000 2020
6 Chiratae Ventures Series A 9000000 2020table(Cleanstartup1$CityLocation)
Bangalore Gurgaon Hyderabad Mumbai New Delhi Others Pune
863 344 164 609 424 528 106 startup2 <- Cleanstartup1
startup2$IndustryVertical <- as.character(startup2$IndustryVertical)
#clean code
startup2$IndustryVertical[startup2$IndustryVertical != "Consumer Internet" & startup2$IndustryVertical != "Technology" & startup2$IndustryVertical != "ECommerce" & startup2$IndustryVertical != "Logistics" & startup2$IndustryVertical != "Education" & startup2$IndustryVertical != "Healthcare"] <- "OtherSectors"
table(startup2$IndustryVertical)
Consumer Internet Education Healthcare
941 25 72
Logistics OtherSectors Technology
32 1490 478 TopIT <- table(startup2$InvestmentType)
TopIT.df <- as.data.frame(TopIT)
top10IT <- top_n(TopIT.df,10)
top10IT
Var1 Freq
1 Debt Funding 25
2 Private Equity 1355
3 Seed / Angel Funding 47
4 Seed Funding 1350
5 Seed/ Angel Funding 60
6 Seed/Angel Funding 23
7 Seed\\\\nFunding 30
8 Series A 25
9 Series B 21
10 Series C 14sum(startup2$AmountUSD)
[1] 38277567120boxplot(startup2$AmountUSD, horizontal = TRUE, xlab = "Amount in USD", main = "Startup Investment plot")
temp = startup2[complete.cases(startup2),] %>% filter(CityLocation == startup2$CityLocation[3])
#Looking for outliers in the data and removing them
outliers = boxplot(temp$AmountUSD ~ temp$IndustryVertical, plot=FALSE)$out
temp.out = temp[-which(temp$AmountUSD %in% outliers),]
# Plotting a boxplot
p2 = ggplot(temp.out, aes(x = reorder(IndustryVertical, AmountUSD), y = log(AmountUSD))) +
geom_boxplot(outlier.shape = NA, show.legend = FALSE) + coord_flip() +
stat_summary(fun = mean, col = "honeydew4", geom = 'point') +
labs(x = "IndustryVerticle", y = "Amount in USD(Millions)", title = "Plotting outliers per Top5 IndustryVerticle on average", subtitle = "average given by dot")
p2
boxplot(startup2$AmountUSD ~ startup2$InvestmentType, horizontal = FALSE, xlab= "Amount of Investment in USD", ylab = "Year", main = "Year Wise Investment Analysis")
boxplot(startup2$SubVertical ~ startup2$CityLocation,horizontal = TRUE, ylab= "Amount of Investment in USD", xlab = "Industry Sector", main = "Industrial Sector Wise Investment Analysis", boxwex = 0.5)
mytable <- xtabs(startup2$year ~ startup2$IndustryVertical)
boxplot.default(mytable, ylab = "AMount of Investment", xlab = "Industry Sector")
boxplot(startup$IndustryVertical ~ startup$CityLocation,horizontal = TRUE, ylab= "Amount of Investment in USD", xlab = "City", main = "Startup citiwise location Wise Investment Analysis", boxwex = 3)
boxplot(startup$AmountUSD ~ startup$InvestmentType,horizontal = FALSE, ylab= "Amount of Investment in USD", xlab = "Investment Type", main = "Investment type vs Investment Analysis")
plot(mytable, ylab = "AMount of Investment", xlab = "Type of Investment")
b = barplot(head(sort(table(startup$CityLocation), decreasing=T),20),col=rainbow(10,0.5), las=2, ylim=c(0,750), xlab="City Name", ylab="No Of StartUps")
text(b,head(sort(table(startup$CityLocation), decreasing=T),20),head(sort(table(startup$CityLocation), decreasing=T),20),srt=90, pos=4)
##Investor frequency table
Investortable <- head(sort(table(startup2$Investors), decreasing=TRUE))
Investortable
Undisclosed Investors Ratan Tata Indian Angel Network
104 25 24
Shell Foundation Kalaari Capital Sequoia Capital
21 16 15 ## Visualization using Bar chart
barplot(Investortable, xlab="Investor Name", ylab = "No. of Startups which received funding", col="lightblue",cex.names = 0.5 )
b = barplot(head(sort(table(startup2$InvestmentType), decreasing=T),20),col=rainbow(10,0.5), las=2, ylim=c(1,1500), xlab="InvestmentType", ylab="Freqency of Investments")
text(b,head(sort(table(startup2$InvestmentType), decreasing=T),20),head(sort(table(startup2$CityLocation), decreasing=T),20),srt=90, pos=4)
#generalize the case to prevent misreading on multiple elements with different case
startup2$IndustryVertical <- as.factor(tolower(startup2$IndustryVertical))
#Total amount funded
fund5 <- as.data.frame(aggregate(AmountUSD ~ IndustryVertical, startup, sum))
fund5 <- fund5[order(fund5$AmountUSD, decreasing = TRUE), ] %>%
drop_na(IndustryVertical) %>%
head(5)
fund5
IndustryVertical AmountUSD
1 E-Commerce 8189518685
2 Consumer Internet 6323472695
3 Transportation 3916632394
4 Technology 2267804310
5 Finance 1981978000ggplot(data = fund5, aes(IndustryVertical, AmountUSD, fill = IndustryVertical)) +
geom_bar(stat = "identity") +
coord_flip() +
scale_y_continuous(labels = unit_format(unit = "M", scale = 0.0000001)) +
labs(x = "Industry Category", y = "Investment Amount (USD)", title = "TOP 5 Start-up Investment in India") +
theme(legend.position = "none")
ggplot(startup2, aes(x=AmountUSD, y = year, color = IndustryVertical )) + geom_point() +
scale_x_log10() +
facet_wrap(~CityLocation)
The graph shows the distribution of start ups growth in various cities over the years from 2015 to 2020
temp=startup %>% select(StartupName,Investors,AmountUSD) %>% ddply(.(StartupName),summarise,sum=sum(AmountUSD)) %>% arrange(desc(sum))
kable(head(temp,25),"html") %>% kable_styling("striped",full_width=T) %>% column_spec(1:2,bold=T,background="white") %>% row_spec(c(1,2,3,5,6,24,25),bold=F,color="lightblue",background="Green")
| StartupName | sum |
|---|---|
| Flipkart | 4059700000 |
| Rapido Bike Taxi | 3900000000 |
| Paytm | 3149050000 |
| Ola | 984500000 |
| Udaan | 870000000 |
| Flipkart.com | 700000000 |
| Snapdeal | 700000000 |
| Ola Cabs | 669725000 |
| True North | 600000000 |
| BYJUS | 525240000 |
| BigBasket | 507000000 |
| GOQii | 450331046 |
| Zomato | 435000000 |
| Olacabs | 400000000 |
| Oyo Rooms | 350000000 |
| Automation Anywhere | 300000000 |
| Grofers | 297000000 |
| OYO Rooms | 285000000 |
| Vogo Automotive | 283000000 |
| Swiggy | 270500000 |
| Edelweiss | 270000000 |
| PolicyBazaar | 267700000 |
| Zilingo | 235900000 |
| Lenskart.com | 231000000 |
| Quikr | 230000000 |
The table shows the top start ups and the amounts raised by them. It clearly says Flipkart raised the most funding.
startup2$StartupName=as.factor(startup2$StartupName)
temp=startup2 %>% select(StartupName,Investors,AmountUSD) %>% ddply(.(StartupName),summarise,sum=sum(AmountUSD)) %>% arrange(desc(sum))
#Top 10
ggplot(head(temp,10),aes(reorder(StartupName,sum),sum,fill=StartupName))+geom_bar(stat="identity")+theme(legend.position="none",axis.text.x = element_text(angle=90,vjust=0.5),plot.title=element_text(color="blue",size=10,face="italic"))+labs(x="",y="Total Investment",title="Investors High Valued Startups")+coord_flip()+scale_y_continuous(labels=scales::comma)
The table shows the top start ups and the amounts raised by them. It clearly says Flipkart raised the most funding.
Conclusion:
This concludes my analysis on the data set and all my questions were answered during the process.
Which startup received the most funding?
Flipkart
Which Industry Vertical received th most funding?
E-Commerce
Does location really matter to raise funding?
Yes, it really influences. All the startup’s from Banglore raised more funds than any other place.
Which type of funding is most prominent among startups?
Private Equity and Seed round are very close and are very prominent in the ecosystem.
Bibilography:
Project: Case Study 1 Start-up Funding Analysis | Kaggle
CRAN - Package tidyverse (r-project.org)
Welcome | R for Data Science (had.co.nz)
