Explore

Recharge API Integration

What It Is

Recharge is our subscription platform — the source of truth for who subscribed, when, for how long, and at what frequency. The Recharge API gives us access to real-time customer data, subscription behavior, and retention outcomes.

In R, we query this data using the httr and jsonlite packages.

Why It Matters

We train neural networks on historical data. Recharge provides that data — especially labels like:

Which customers subscribed

How long they stayed

When they churned

What they purchased and how often

Without Recharge, we can’t learn from actual subscription behavior.

Step-by-Step Setup

✅ 1. Get Your Recharge API Key

Recharge Admin⁠

⁠

Go to Apps & Integrations → API tokens

Create a Read Access Token

Copy the token (starts with sk_...)

Ask Aaron for the Token

✅ 2. Set Up Helper Functions in R

Run this code in R

library(httr)

library(jsonlite)

base_url <- "https://api.rechargeapps.com"

access_token <- "" # Replace this

# Helper: Generic GET with headers

recharge_get <- function(endpoint, query = list()) {

url <- paste0(base_url, endpoint)

res <- GET(url,

add_headers("X-Recharge-Access-Token" = access_token),

query = query)

cat("Status:", status_code(res), "\n")

if (status_code(res) != 200) {

print(content(res, "text", encoding = "UTF-8"))

stop("❌ API request failed.")

}

content(res, "parsed", simplifyDataFrame = TRUE)

}

✅ 3. Pull Data From Recharge

# Get Customers

customers_df <- recharge_get("/customers", query = list(limit = 250))$customers

# Get Subscriptions (paginated)

get_recharge_data <- function(endpoint, limit = 250) {

page <- 1

all_data <- list()

repeat {

cat("Fetching page:", page, "\n")

res <- GET(

url = paste0(base_url, "/", endpoint),

query = list(page = page, limit = limit),

add_headers("X-Recharge-Access-Token" = access_token)

)

if (status_code(res) != 200) {

print(content(res, "text", encoding = "UTF-8"))

stop("❌ Failed to retrieve data from Recharge.")

}

parsed <- content(res, "parsed", simplifyDataFrame = TRUE)

items <- parsed[[endpoint]]

if (length(items) == 0) break

all_data[[page]] <- items

page <- page + 1

}

return(dplyr::bind_rows(all_data))

}

subs_df <- get_recharge_data("subscriptions")

🧠 Notes for Interns

Date handling: Use lubridate::ymd_hms() to convert created_at and cancelled_at columns to timestamps. To convert date strings (like created_at) into timestamps R can understand, use:

library(lubridate)

# Convert to datetime format

customers_df$created_at <- ymd_hms(customers_df$created_at)

subs_df$created_at <- ymd_hms(subs_df$created_at)

subs_df$cancelled_at <- ymd_hms(subs_df$cancelled_at)

This lets us calculate time between events — like how long someone stayed subscribed.

Avoid API rate limits: Never call inside a for loop without a delay.

Recharge may block you if you send too many requests too fast.

If you ever need to loop through a list of API calls, always add a short pause:

for (page in 1:5) {

Sys.sleep(1) # Pause 1 second between calls

response <- recharge_get("/subscriptions", query = list(page = page))

print(paste("Page", page, "fetched."))

}

You’ll use this to create training labels: subscribed_within_30_days, churned_within_90_days, etc.

We’ll eventually train models using a label like: Did this customer subscribe within 30 days?

Here’s how to create that kind of logic using Recharge’s data:

# Join customer and subscription data by customer_id

library(dplyr)

merged_df <- customers_df %>%

inner_join(subs_df, by = c("id" = "customer_id"))

# Convert dates if not already done

merged_df$created_at.x <- ymd_hms(merged_df$created_at.x) # from customers

merged_df$created_at.y <- ymd_hms(merged_df$created_at.y) # from subscriptions

# Create a new label column

merged_df$subscribed_within_30_days <- ifelse(

difftime(merged_df$created_at.y, merged_df$created_at.x, units = "days") <= 30,

1, 0

)

This label (1 for yes, 0 for no) will later be used as our target variable when training the model.

Load content from share.geckoboard.com?

Loading external content may reveal information to 3rd parties. Learn more

Allow

Load content from share.geckoboard.com?

Loading external content may reveal information to 3rd parties. Learn more

Allow

Want to print your doc?
This is not the way.

Try clicking the ··· in the right corner or using a keyboard shortcut (

CtrlP

) instead.