Generate and read analytics reports from the App Store Connect API

Sponsored
RevenueCat logo
Develop with RocketSim, Ship with Helm.

Helm Pro yearly subscribers now get a 30% discount on RocketSim thanks to contingent pricing on the App Store.

Apple has recently introduced over 50 new analytics reports with hundreds of new data points and metrics to help developers understand how their apps are performing.

These reports include data such as App Store Engagement, App Store Commerce, App Usage, Frameworks Usage and Performance.

While this new data offers a lot of insights and can be very valuable, it’s available exclusively through the App Store Connect API and the way of retrieving the data is not very intuitive.

In this article, I’ll show you how to access these new metrics using Antoine Van Der Lee’s App Store Connect Swift SDK.

Installing the App Store Connect API Swift SDK

As soon as the new metrics were announced in the latest version of the App Store Connect API, I decided to open a pull request to regenerate the Swift interfaces in the App Store Connect Swift SDK to include the new endpoints using the latest App Store Connect’s 3.4 Open API specification.

These changes were merged and released in version 3.2.0 of the SDK, which is the version we’ll be using in this article.

To install the App Store Connect Swift SDK, you can just add it as a dependency to your package manifest file:

Package.swift
// swift-tools-version: 5.10

import PackageDescription

let package = Package(
    name: "ASCAnalytics",
    platforms: [.macOS(.v13)],
    dependencies: [
        .package(
            url: "https://github.com/AvdLee/appstoreconnect-swift-sdk.git", 
            exact: "3.2.0"
        )
    ],
    targets: [
        .executableTarget(
            name: "ASCAnalytics", 
            dependencies: [
                .product(
                    name: "AppStoreConnect-Swift-SDK", 
                    package: "appstoreconnect-swift-sdk"
                )
            ]
        ),
    ]
)

Configuring the App Store Connect API Swift SDK

Now that the SDK is installed, we just need to configure it with an App Store Connect API key:

ASCAnalytics.swift
import AppStoreConnect_Swift_SDK
import Foundation

let configuration = try! APIConfiguration(
    issuerID: "🙈",
    privateKeyID: "🙈",
    privateKey: "🙈"
)

let provider = APIProvider(configuration: configuration)

If you’re not sure how to create a key for the App Store Connect API, I would encourage you to read the ‘Creating an App Store Connect API key’ section of my ‘Fastlane and App Store Connect API keys’ article.

Generating an analytics report

The first thing you need to do to access new metrics is to generate a report request for a specific app. This can be done by getting the ID of the app you want to generate the report for and then making a POST request to the /v1/analyticsReportRequests endpoint.

ASCAnalytics.swift
// MARK: - Get the ID of the app
// https://api.appstoreconnect.apple.com/v1/apps?sort=bundleId&fields%5Bapps%5D=name
let request = APIEndpoint
    .v1
    .apps
    .get(parameters: .init(sort: [.bundleID], fieldsApps: [.name]))

let appsResponse = try await provider.request(request)
let qreateAppId = appsResponse.data.first { $0.attributes?.name == "QReate - QR code generator" }.map { $0.id }

guard let qreateAppId else { exit(1) }

// MARK: - Create a new Report Request
let relationships = AnalyticsReportRequestCreateRequest.Data.Relationships(
    app: .init(data: .init(type: .apps, id: qreateAppId))
)
let attributes = AnalyticsReportRequestCreateRequest.Data.Attributes(accessType: .ongoing)
let data = AnalyticsReportRequestCreateRequest.Data(
    type: .analyticsReportRequests,
    attributes: attributes,
    relationships: relationships
)
let createRequest = AnalyticsReportRequestCreateRequest(data: data)
// https://api.appstoreconnect.apple.com/v1/analyticsReportRequests
let requestReport = APIEndpoint.v1.analyticsReportRequests
    .post(createRequest)

_ = try await provider.request(requestReport)

Despite what it may seem, the payload to the POST request is quite simple. We just need to specify the id of the app we want to generate the report for as a relationship (in this case my app QReate) and the access type of the report.

The access type parameter can be one of two values:

  • .ongoing: The most common type of report request, which generates daily data for reports of all frequencies.
  • .oneTimeSnapshot: A one-time report request to obtain historical data.

Getting all available reports

After making the POST request, we can periodically check the report request and retrieve all of its available reports. If we’re only interested in a specific kind of data, we can filter the request to the reports endpoint by category (in this case APP USAGE):

ASCAnalytics.swift
// MARK: - Read all available report requests for an app
// https://api.appstoreconnect.apple.com/v1/apps/6446048195/analyticsReportRequests?filter%5BaccessType%5D=ONE_TIME_SNAPSHOT,ONGOING&fields%5BanalyticsReportRequests%5D=accessType,reports,stoppedDueToInactivity&fields%5BanalyticsReports%5D=category,instances,name&include=reports
let readReportsRequest = APIEndpoint
    .v1
    .apps
    .id(qreateAppId)
    .analyticsReportRequests
    .get(parameters: .init(filterAccessType: [.oneTimeSnapshot, .ongoing], fieldsAnalyticsReportRequests: [.accessType, .reports, .stoppedDueToInactivity], fieldsAnalyticsReports: [.category, .instances, .name], include: [.reports]))

let allReports = try await provider.request(readReportsRequest).data

// MARK: - Get all reports for a report request
guard let reportRequestId = allReports.first?.id else { exit(1) }
// https://api.appstoreconnect.apple.com/v1/analyticsReportRequests/105262f5-0cc0-4c4f-8eed-ff56509ee135/reports?filter%5Bcategory%5D=APP_USAGE
let reportInformation = APIEndpoint
    .v1
    .analyticsReportRequests
    .id(reportRequestId)
    .reports
    .get(parameters: .init(filterCategory: [.appUsage]))

let appUsageReports = try await provider.request(reportInformation)

This will return a list of all available APP USAGE reports for the app, which we can then retrieve by name to see the data we are interested in. Let’s for example get the id for the App Crashes report so that we can use it later to retrieve its data:

ASCAnalytics.swift
guard let crashesReportId = appUsageReports.data
    .filter({ $0.attributes?.name == "App Crashes" })
    .first?.id else {
    exit(1)
}

Getting the report’s segments

Unfortunately, the App Store Connect API does not return the data for the reports directly. It instead splits the data into instances, which are generated for the available granularities (daily, weekly or monthly).

In turn, each instance contains numerous segments, each of which contains a list of URLs to download the data for the reports.

Let’s now get all the segments for a report’s instance:

ASCAnalytics.swift
// MARK: - Get the information for a report
// https://api.appstoreconnect.apple.com/v1/analyticsReports/r2-105262f5-0cc0-4c4f-8eed-ff56509ee135/instances
let instances = APIEndpoint
    .v1
    .analyticsReports
    .id(crashesReportId)
    .instances
    .get()
let instancesResponse = try await provider.request(instances)
guard let instanceId = instancesResponse.data.first?.id else { exit(1) }

// MARK: - Get segments
// https://api.appstoreconnect.apple.com/v1/analyticsReportInstances/3472b36d-b349-41e5-8ff2-25967428947b/segments?fields%5BanalyticsReportSegments%5D=url,checksum,sizeInBytes
let segments = APIEndpoint
    .v1
    .analyticsReportInstances
    .id(instanceId)
    .segments
    .get(fieldsAnalyticsReportSegments: [.url, .checksum, .sizeInBytes])
let segmentsResponse = try await provider.request(segments)

Downloading the segment’s data

Finally, let’s use the url attribute of the segment entity to download the data for the report and write it into a file we can then read:

ASCAnalytics.swift
// MARK: - Download segment file
guard let segmentURL = segmentsResponse.data.first?.attributes?.url else { exit(1) }

let (location, downloadFileResponse) = try await URLSession.shared.download(from: segmentURL)
guard let httpResponse = downloadFileResponse as? HTTPURLResponse,
      httpResponse.statusCode == 200 else {
    exit(1)
}

try FileManager.default
    .moveItem(
        at: location,
        to: URL.desktopDirectory.appending(component: "crashes.zip")
    )

After all the steps above completes, we should have a crashes.zip file on our desktop that, when unzipped, will contain the file with the data for the App Crashes report:

crashes
Date	App Name	App Apple Identifier	App Version	Device	Platform Version	Crashes	Unique Devices
2024-01-15	QReate - QR-code generator	6446048195	1.0.4	Desktop	macOS 13.6	1	1
2024-01-15	QReate - QR-code generator	6446048195	1.0.4	Desktop	macOS 14.2	4	4