Attribution in Localytics helps you find where your users originated. All app users come from somewhere—whether that’s a paid Facebook ad, a download link via email, or even organic traffic in the app store. To help you tell the full story of your users’ lifecycle, we have a feature called Attribution.
Attribution helps you figure out the download source for your app, so you can better target new users in the future.
You should think of Localytics as just a piece of your whole attribution puzzle. Our attribution features allow you to connect data from your current attribution partner (or partners) to give you a better picture of your customer lifecycle.
It’s important to note that Localytics should not replace your attribution partner, but instead join in your data suite.
Sometimes Attribution and Acquisition are used interchangeably in the marketing world. But in our case, Localytics can help you understand users you’ve already acquired—and then we can backtrack through your acquisition channels to get a clearer picture of how those users interact with your app. At this time, Localytics’ attribution feature is not built to help you acquire new users.
You should use an outside source for your acquisition product—and then Localytics can connect the dots to your current users via attribution.
Knowing what your users do in your app is only one piece of the puzzle. You’re likely tracking what your end users click outside your app, and Localytics can take care of what your users do within your app.
Attribution allows you to connect these two siloed pieces of data—helping you to invest in conversions instead of clicks or installs.
App stores ultimately decide what information is available to your app, in regards to attribution. Google and Apple have different rules for this, so it’s important to make sure you’re following them each correctly.
Google allows you to use regular UTM data to pass information to your app from the originating URL. Apple does not allow for UTM parameters, so we recommend checking out this article for more information.
Localytics helps you connect users to their attribution source through three mobile attribution methods. Those methods are:
Ad network integrations receive data from a third-party source.
Ad networks can communicate information about which URL devices clicked in order to download your app. Localytics then matches this information to report originating ad network and campaign. We’ll communicate that information back to your ad network, notifying them of the install. Read more here.
Like ad network integrations, add-tracking vendor integrations receive their data from a third party source (in this case, the ad network), as opposed to Localytics.
Ad-tracking vendors don’t display ads themselves but pull together data from a variety of ad networks. From there, they’re able to determine the originating ad network and source through custom work.
They’re then able to relay that information to Localytics when the end user opens the app for the first time. For more on ad-tracking vendor integrations, check out this article.
The last type of attribution source is custom redirect URLs. These can come from Localytics! We’ll generate a redirect URL so that we can directly track devices that click the link. If we’re able to observe a matching device in our data, we’ll report the originating ad network and campaign. For more on redirect URLs (for non-integrated sources), check out this article.
There are a few words that are used in the Attribution space that are helpful to know before you dive in. The main two are postbacks and device fingerprinting.
Postbacks are a programmatic way for one system to communicate with another system. In the attribution context, Localytics will issue a postback to the integrated ad network once we learn that a new device has originated from that network.
By sending this information, the integrated ad network can keep track of how many users have actually opened the app.
In some cases, Localytics also supports event postbacks. This means that if you acquire users from app-install campaigns on select integrated ad networks, you can designate specific events tracked by Localytics that we postback to the originating ad network.
When a user you acquired from an integrated ad network completes one of the in-app events you designate, Localytics uses an API call to send the information back to the network.
Localytics supports posting back the Event name; it does not support posting back Attribute values or associated revenue values. Not all integrated ad networks allow for Event postbacks. Check out this article for a list of those currently available.
You should only select important conversion events to postback. Do not postback all in-app events, as doing so will dilute the significance of event postbacks as indicators of end-user value. Consider designating events for postback that indicate monetization (e.g. user completed an in-app purchase), engagement (e.g. end user completed my app's tutorial), community contribution (e.g. user rated an item).
As noted above, redirect URLs work by briefly routing URL-clicking devices to a Localytics destination for an invisibly brief period of time, just long enough for Localytics to collect device information before redirecting to the intended download destination.
When available, Localytics collects standard advertising identifiers which have the benefit of being definitively unique. These identifiers are not always accessible, for example, if you include URLs to drive downloads in email, mobile web, or organic social media campaigns. Notably, iOS has a more restrictive model that often makes traditional redirect URLs impractical due to the unavailability of devices' IDFA, as described here.
Given this, Localytics also supports a special type of attribution called "fingerprinting." Fingerprinting is functionally similar to traditional redirect URLs, except that rather than depending on a device's advertising ID, Localytics will retrieve a number of other device identifiers that are individually non-unique, but can be combined to create a single unique identifier, i.e. a device "fingerprint."
This method has the advantage of allowing you to track installs that otherwise would not be trackable if you were to only depend on advertising IDs. A rarely-observed disadvantage, however, is that under some idiosyncratic conditions outside our control device fingerprints may end up being non-unique and therefore not function as expected - e.g. if a high volume of people in the same area with the same device model download the same app at the same time.
Once you've set up your campaigns, you can view their success in the Attribution Report. Click into any of the campaigns you're running to see more details about them. For more on the Attribution Report check out our article.