Your server is fast.
Does your app feel fast?
See API performance from your users’ side of the connection. Find slow requests, separate network delays from server time and get notified when errors become an incident.
iOS, tvOS & Android · PRO & ENTERPRISE · Hosted in Switzerland
Illustrative data
Find where the wait really happens
A slow API call is not always a slow backend. Break the round trip into phases before you decide what to fix.
GET /products840 ms- DNS
- 18 ms
- Connection
- 42 ms
- Upload
- 20 ms
- Server
- 580 ms
- Download
- 120 ms
- Client unattributed
- 60 ms
Send Server-Timing: total;dur=580 to help refine the server phase for a single-transaction request.
Illustrative timing breakdown
- Separate the causes. DNS resolution, connection setup, upload, server wait and download each tell a different story.
- See more than the average. P90, P95 and P99 reveal slow requests that a healthy-looking average can hide.
- Measure the whole experience. Track request volume, payload sizes and an Appdex score alongside response times.
Spotted a slow request in a bug report, crash or UI test recording? Choose “Analyze request” to open its operation here.
An incident deserves an alert. Noise does not.
Set an error-rate threshold for a backend environment. Appelium looks for convincing evidence before interrupting your team.
- Choose what matters. Monitor client, business, server or overall errors, with a threshold for your environment.
- Confidence before notification. A 95% Wilson score interval helps distinguish a sustained problem from a handful of unlucky requests.
- Fewer repeat interruptions. An active incident stays open while the observed rate exceeds the threshold, with a recovery grace period to reduce flapping.
Evaluated every minute over the latest five minutes of received metrics. Detection depends on traffic volume and data arrival, not a guaranteed one-minute alert.
Server error rate exceeds 5%
An error-rate threshold crossed with statistical confidence, not just a single failed request.
Incident and recovery notifications, with a link back to Appelium.
Illustrative incident
Latency trends help you investigate performance. Network alarms currently trigger on error rates, not response time.
Go from “the API is slow” to a useful answer
Narrow the scope, find the affected requests and connect backend trouble to what happened in the app.
Pinpoint the affected traffic
Break metrics down by operation, host, backend environment, app version, platform or network interface. Compare where errors and delays concentrate.
Keep test traffic in context
Separate test requests from non-test traffic in the workspace network view, or include both when investigating the bigger picture.
See the user-facing impact
Enable diagnostic network logs to inspect requests alongside bug reports, crashes and handled errors. Metrics do not require full payload capture.
Use the networking stack you already have
Enable monitoring in the SDK. URLSession powers collection on iOS and tvOS; add the Appelium EventListener to your OkHttp client on Android.
val client = OkHttpClient.Builder()
.eventListenerFactory(
AppeliumOkhttpEventListener.Factory()
)
.build()A fast request can still be part of a slow flow. Use performance traces to time complete app operations, not just their network calls.
Get started: iOS, tvOS, Android · Network monitoring & Server-Timing guide
Frequently asked questions
How is this different from server-side monitoring?
Appelium measures requests from your mobile app, including DNS, connection setup, sending the request and receiving the response. A fast server can still feel slow to a user on a poor connection. These client-side measurements complement your server monitoring rather than replace it.
What can trigger a network alarm?
Choose a backend environment, an error category (client, business, server or all errors) and an error-rate threshold. Appelium evaluates the latest five minutes of received metrics every minute. A new alarm requires statistical confidence that the error rate exceeds your threshold; low traffic can delay detection. Latency is available for analysis, but does not currently trigger alarms.
Do I need to capture request bodies to monitor performance?
No. Network metrics and full request logs are separate capabilities. On Android, the OkHttp EventListener captures metrics; the optional interceptor adds headers and payloads to diagnostic logs. The SDK also provides filtering and obfuscation hooks for captured network logs.
Which platforms and plans are supported?
Network monitoring supports iOS and tvOS through URLSession, and Android through the Appelium OkHttp integration. Network monitoring and alarming are available on PRO and ENTERPRISE plans.
See the connection your users actually get
Find the slow calls. Understand the failures. Bring the right people in.
You are in a good company - Join over 150 apps!
Over 150 apps in Switzerland choose to use Appelium.
👨💻 Latest content from our Engineering blog:
XPCService
Strongly typed bidirectional communication channel between UIApplication and XCUIApplication with powerful APIs and super fast performance.
More about XPCService
Screen Flow Pattern in UI tests
Robust and scaleable pattern for structuring UI tests. Highly reusable screen verification code with human readable test case code. All flows and interactions statically typed 🙌.
More about the Screen Flow Pattern
SnapshotQuery™
Learn about SnapshotQuery - our latest addition to AppeliumTests which radically improves the performance of your XCUITests while making them more stable and reliable at the same time.
More about SnapshotQuery
Early access and news
Be the fist one to know about all the exciting engineering posts and product announcements.