The Indian digital payments story has truly captured the world’s imagination. From the largest towns to the remotest villages, there is a payments revolution being driven by the penetration of mobile phones, mobile internet and state-of-art payments infrastructure built as Public Goods championed by the central bank and the government. PhonePe started in 2016 and has been a strong beneficiary of the API driven digitisation of payments in India. When we started , we were constantly looking for definitive data sources on digital payments in India without much success. As a way of giving back to the data and developer community, we decided to open the anonymised aggregate data sets that demystify the what, why and how of digital payments in India. Licensed under the CDLA-Permissive-2.0 open data license, the PhonePe Pulse Dataset API is a first of its kind open data initiative in the payments space.
Our goal is to share this data with everyone (license below), so that you can build your own understanding, insights and visualization on how digital payments have evolved over the years in India.
This data has been structured to provide details of following two sections with data cuts on Transactions and Users of PhonePe Pulse - Explore tab.
- Aggregated - Aggregated values of various payment categories as shown under Categories section
- Top - Totals of top States / Districts /Pin Codes
All the data provided in these folders is of JSON format. For more details on the structure/syntax you can refer to the JSON Structure / Syntax section of the documentation.
Head to the data folder to the find below shown structure. Overall, above mentioned sections data can be found at top level folder structure.
Under each of these sections there are folders for Transactions and Users respectively.
Data for Transactions and Users is grouped under country level within India folder which further grouped the data into each year(for country level data) and there is one folder with name state which groups data for all the available states of India respectively.
Similar to country level data, state level data too grouped into each year. All of these year folders(both at country and state level) have a maximum of four files with names starting from 1 to 4. These numbers represent each quarter in the selected year.
Example: 2021 > 1.json represents data for quarter 1 (Jan, Feb and Mar 2021)
For details on syntax of each of these files, refer to JSON Structure / Syntax.
data
|___ aggregated
|___ transactions
|___ country
|___ india
|___ 2018
| 1.json
| 2.json
| 3.json
| 4.json
|___ 2019
| ...
|___ 2019
|___ state
|___ andaman-&-nicobar-islands
|___2018
| 1.json
| 2.json
| 3.json
| 4.json
|___ andhra-pradesh
| ...
| ...
Transaction data broken down by type of payment at country level.
For complete details on syntax find the comments in below code
NOTE: Similar syntax is followed for state level too. Ex: data/aggregated/transaction/country/india/state/delhi/2018/1.json
{
"success": true, //Ignore. For internal use only
"code": "SUCCESS", //Ignore. For internal use only
"data": {
"from": 1514745000000, //Data duration
"to": 1522175400000,
"transactionData": [
{
"name": "Recharge & bill payments", //Type of payment category
"paymentInstruments": [
{
"type": "TOTAL",
"count": 72550406, //Total number of transactions for the above payment category
"amount": 1.4472713558652578E10 //Total value
}
]
},
...,
...,
{
"name": "Others",
"paymentInstruments": [
{
"type": "TOTAL",
"count": 5761576,
"amount": 4.643217301269438E9
}
]
}
]
},
"responseTimestamp": 1630346628866 //Ignore. For internal use only.
}Users data broken down by devices at country level.
For complete details on syntax find the comments in below code
NOTE: Similar syntax is followed for state level too. Ex: data/aggregated/user/country/india/state/delhi/2021/1.json
{
"success": true, //Ignore. For internal use only.
"code": "SUCCESS", //Ignore. For internal use only.
"data": {
"aggregated": {
"registeredUsers": 284985430, //Total number of registered users for the selected quarter.
"appOpens": 8635508502 //Number of app opens by users for the selected quarter
},
"usersByDevice": [ //Users by individual device
{
"brand": "Xiaomi", //Brand name of the device
"count": 71553154, //Number of registered users by this brand.
"percentage": 0.2510765339828075 //Percentage of share of current device type compared to all devices.
},
...,
...,
{
"brand": "Others", //All unrecognized device types grouped here.
"count": 23564639, //Number of registered users by all unrecognized device types.
"percentage": 0.08268717105993804 //Percentage of share of all unrecognized device types compared to overall devices that users are registered with.
}
]
},
"responseTimestamp": 1630346630074 //Ignore. For internal use only.
}Top 10 states / districts / pin codes where the most number of the transactions happened for a selected year-quarter combination.
For complete details on syntax find the comments in below code
NOTE: Similar syntax is followed for state level too. The only exception is, it won't have data for states. Ex: data/top/transaction/country/india/state/delhi/2021/1.json
{
"success": true, //Ignore. For internal use only.
"code": "SUCCESS", //Ignore. For internal use only.
"data": {
"states": [ //List of states where most number of transactions happened along with total value for a selected year-quarter combination.
{
"entityName": "karnataka", // State name
"metric": {
"type": "TOTAL",
"count": 523797492, //Total number of transactions
"amount": 7.549953574123948E11 //Total value of all transactions
}
},
...,
],
"districts": [ //List of districts where most number of transactions happened along with total value for a selected year-quarter combination.
{
"entityName": "bengaluru urban", //District name
"metric": {
"type": "TOTAL",
"count": 348712787, //Total number of transactions
"amount": 4.324013412317671E11 //Total value of all transactions
}
},
...,
],
"pincodes": [ //List of pin codes where most number of transactions happened along with total value for a selected year-quarter combination.
{
"entityName": "560001", //Pin code
"metric": {
"type": "TOTAL",
"count": 111898471, //Total number of transactions
"amount": 1.5427512629157785E11 //Total value of all transactions
}
},
...,
]
},
"responseTimestamp": 1630346629360 //Ignore. For internal use only.
}Top 10 states / districts / pin codes where most number of users registered from, for a selected year-quarter combination.
For complete details on syntax find the comments in below code
NOTE: Similar syntax is followed for state level too. The only exception is, it won't have data for states. Ex: data/top/user/country/india/state/delhi/2021/1.json
{
"success": true, //Ignore. For internal use only.
"code": "SUCCESS", //Ignore. For internal use only.
"data": {
"states": [ //List of states where the most number of users registered from, for a selected year-quarter combination.
{
"name": "maharashtra", //State name
"registeredUsers": 37077537 //Number of registered users
},
...,
],
"districts": [ //List of districts where the most number of users registered from, for a selected year-quarter combination.
{
"name": "bengaluru urban", //State name
"registeredUsers": 9955387 //Number of registered users
},
...,
],
"pincodes": [ //List of pin codes where most number of users registered from, for a selected year-quarter combination.
{
"name": "201301", //Pin code
"registeredUsers": 541127 //Number of registered users
},
...,
]
},
"responseTimestamp": 1630346630074 //Ignore. For internal use only.
}How frequently data gets updated?
A: The frequency is once in three months.How to report a bug or raise a query?
A: Use GitHub issues to raise a query.