How to implement referral program in mobile Apps f

问题:

We have mobile app that's available in both Google Play Store and Apple AppStore, we wants to implements a referral program to get more users to install and use our App..

Here's the user story:

• Every new user (E.g John) is given to a unique referral link, where he/she can share to FB/TW/Email or SMS.
• When John friend click the link, they are directed to respective AppStore base on their device.
• The moment John's friend install the app, and open the app, our server should get notify, and we shall know that the referral is from John, John will be rewarded accordingly..

We have evaluate a number of Mobile App Install Tracking Tools, most of the tools are best use in Publisher/ Mobile Advertisement..

Appreciate any comments, suggestions

Cheers James

回答1:

We are doing a similar type of referral system in our app which provides certain amount of Reward Points to the referrer (one who actually shared the link to app) on successful install of the app by new user (one who received the link).

I will try to explain our implementation using your scenario: So according to your user story:

• Every new user (E.g John) is given a unique referral link, which he/she can share it to his friends using FB/TW/Email or SMS.

• When John's friend clicks on the link, he is re-directed first to a Servlet which eventually redirects him to respective AppStore based on his platform (Android/iPhone) to download the app.

• Note that we can find out the IP address, user agent and device model of the user who clicked the link using a Servlet.

• We already know that the referral link belongs to John and hence the servlet maps the IP address,user agent and/or device model of John's friend to John's entry in database.

• The moment John's friend installs the app, and opens the app, the application sends the IP address, user agent and/or device model to the server.

• The server checks the entry against all the user to find the referrer and if it finds him, in our case John, John gets rewarded accordingly..

Thats it. We implemented this in the last month itself and it does increased our downloads. Let me know if that's helpful.

回答2:

Android

On android you can create an install broadcast listener, where you can get and save the referrer part of the link to SharedPreferences

https://play.google.com/store/apps/details?id=com.example.app &referrer=example_referral_code_here

For this check
Android - Is it possible to get install referrer programmatically
and
Get referrer after installing app from Android Market

iOS

For now (Jan 2016) It is not possible to get the install referrer code in your app, like with android.

Options i had for ios were:

1. user frameworks/3rd party services like alau.me
2. a workaround with fingerprinting, similar to what Puru Pawar suggests
3. just make the user enter the referral code inside your app

回答3:

If anyone is still looking for solution to this question. I read a lot of blogs and documents for the same problem, and arrived at following conclusion:-

IP Address + User Agent + Device Model is not enough to identify a device uniquely. So, I think it will be better to use dynamic links. You can create dynamic links containing a unique id. And that data will survive url redirection to app store and even after installation your app can get the unique id from dynamic link.Google's firebase can be used for creating dynamic links :-

https://firebase.google.com/docs/dynamic-links/

Or if you want to use a ready to use solution then you can refer branch.io

回答4:

It can be done using Google Campaign Measurement where in the utm_source just give unique number/string for each user which you will receive when user's friend installs the app as Google Play store will broadcast it after the installation. Refer to this question.

回答5:

 Gradle File
 implementation 'com.android.installreferrer:installreferrer:1.0'


 PUT CODE IN ACTIVITY WHERE IT NEED TO RECIEVE REFFERAL CODE

 InstallReferrerClient mReferrerClient;

 mReferrerClient = InstallReferrerClient.newBuilder(this).build();

 mReferrerClient.startConnection(new InstallReferrerStateListener() {
 @Override
 public void onInstallReferrerSetupFinished(int responseCode) {
            switch (responseCode) {
                case InstallReferrerClient.InstallReferrerResponse.OK:
                    // Connection established
                    try {
                        ReferrerDetails response = 
 mReferrerClient.getInstallReferrer();
                        if (!response.getInstallReferrer().contains("utm_source"))
                            edtPRferelCode.setText("" + 
 response.getInstallReferrer());
                    } catch (Exception e) {
                        e.printStackTrace();
                    }
                    mReferrerClient.endConnection();
                    break;
                case 
 InstallReferrerClient.InstallReferrerResponse.FEATURE_NOT_SUPPORTED:
                    // API not available on the current Play Store app
                    break;
                case 
 InstallReferrerClient.InstallReferrerResponse.SERVICE_UNAVAILABLE:
                    // Connection could not be established
                    break;
            }
        }

        @Override
        public void onInstallReferrerServiceDisconnected() {
            // Try to restart the connection on the next request to
            // Google Play by calling the startConnection() method.
        }
    });

    SHARE LINK EXAMPLE
    https://play.google.com/store/apps/details?id=com.yourpackage&referrer=9BE46300

回答6:

An alternative is to let your current users make a unique code, for example, David123 or HenryABC. They share their code (which is easy to remember as it's personal). Within your app you provide a referral box that a new user enters the code to unlock additional features. This way you can track and don't have to worry about a fingerprint. The big issue with a fingerprint method is iOS is not unique enough and mobile networks share IP addresses. So someone with an iPhone 6s in London on EE is likely to have the same fingerprint as 100s of other people.