-
Notifications
You must be signed in to change notification settings - Fork 1
FAQs
The information below provides you with links and explanations to help you find your way around the predict.io SDK Integration guides and how to test your SDK.
Once your App (with predict.io) is started, it will continuously monitor user activities (of course that is battery optimised). Whenever predict.io detects any event (started/ended journey etc) it will inform your App using call back methods; i.e Protocols in iOS. Here is more details on how to work with Protocols.
To make sure that you have integrated correctly our SDK before going live please follow the instructions below to test your implementation before (and after) you submit the app to the stores. This is very important.
Always do field tests, and choose your locations wisely:
- From open skies to crowded sky scrapers
- With limited to no GPS reception
- Change heights during the tests
- Choose different times of the day, weather conditions and locations
- Test most scenarios, from driving real slow to fast and traffic jams
- Test around the device’s power up time, especially for a device that was powered off for days and/or with no internet connectivity
- Use different device settings and different devices
predict.io allows you to retrieve real-time updates on the mobility status of a user. You can run the API in iOS apps. No additional hardware is required.
Main features include detecting when a user:
- has arrived at a destination.
- has just started a journey.
You may use this information to trigger events or notifications specific to your use case.
Our mobile SDK gives you battery friendly background location. So that you always know when and where a user arrives or departs. No fiddling with geofences. No beacons or NFC needed. It uses the sensors embedded in any modern smartphone. You can embed it in minutes. Rather than spending months with the Activity APIs yourself. Our SDK powers many industry leading Android and iOS apps in mobility, retail, hospitality, lifestyle and banking.
The native Activity APIs are one-size-fits-all. But that is not enough for granular mobility detection. Hence we made it better. With native APIs you have a ~50% false positive rate for car trips. We cut it down to 10% for a typical urban driver.
Our SDK consumes only ~2% of a total charge by running in the background. The same result would consume 20% of the battery on standard OS APIs. If you face any different behavior please don’t hesitate to contact us.
Installing our SDK takes you 30 minutes. Building a battery optimized, normalised output for iOS and Android will cost you years. Trust us. We’ve been through it.
Just like you, our primary concern is protecting people’s privacy. Our own privacy policies reflect just that. We have built our system so that no personal identifiable information is being collected by us. We don’t have any user credentials, any unique device ID nor a clear text IP adress. The only device identifier used by the system is a proprietary hash generated by us. Which is not used in any other product anywhere in the world. Hence it can not be traced back across other services. We never know who your user is. And we don’t want to. We have gone through several security audits with large corporates. We are confident to be able to adhere to any of your standard, too
When you implement the predict.io SDK, we collect non-PII (anonymous) data of the users using the SDK. This includes events data as well as device data to improve the service e.g. regarding performance and battery consumption (see Terms of Service for more details).
To make sure that you have integrated correctly our SDK before going live please follow the instructions below to test your implementation before (and after) you submit the app to the stores. This is very important.
Always do field tests, and choose your locations wisely:
- From open skies to crowded sky scrapers
- With limited to no GPS reception
- Change heights during the tests
- Choose different times of the day, weather conditions and locations
- Test most scenarios, from driving real slow to fast and traffic jams
- Test around the device’s power up time, especially for a device that was powered off for days and/or with no internet connectivity
- Use different device settings, and different devices
Easy to answer! No, our SDK does not affect your app, it makes it even better.
- Use this battery disclaimer in your application Description: “Continued use of GPS running in the background can dramatically decrease battery life.”
- Make sure the app details/specs has reason why it need the background location, like there must be some feature of the app which will not work without background location service (e.g turn by turn navigation)
- Background Location (always)
- Motion Activity & Fitness
- Internet
- Access Network State
- Access Fine location
- Access Coarse Location
- Activity Recognition
- Receive Boot Completed
Note: Android plugin will only prompt user for “Location” permission.
- If the method startWithCompletionHandler is returning error, double check that the API key is correctly entered and that the delegate is also set.
- After adding the library and header files, the library and/or header search paths might need to be updated depending upon where in the directory structure, the files have been saved.
- Make sure you have added NSLocationAlwaysUsageDescription key to your info.plist file.
- Make sure you have enabled background location updates.
- While testing, Location Services should be enabled in Settings and the app should also have ‘Always’ permission to allow location access all the time in back ground.
- Use same version of Google Play Services in the whole project which predict.io SDK is using. E.g. if predict.io SDK requires 9.2.1 than use google analytics library of 9.2.1 version as well.
- Make sure you add SDK Proguard Rules in your project's Proguard file.
- Make sure you have given your API_KEY in your plugin.xml file.
- Location\GPS should be enabled\ON in smartphone's settings.
- Airplane mode should be disabled\OFF in smartphone's settings, as predict.io SDK is suspended in Airplane mode for optimal battery performance.
- Device should have internet access while SDK's start() method is invoked.
- Power/Battery saving mode should be disabled\off in device settings.
- Testing predict.io SDK is not possible on any emulator.
- If the app killed by OS or closed by pressing the back button, javascript methods would not fire until the app reopened. For more details Cordova Android Lifecycle
Check out our pricing section to select your plan.