6 Install Attribution Options for Small Apps

TL;DR: The six install attribution options for small apps are Apple App Store Connect campaign links, the Google Play Install Referrer built yourself, Google Analytics for Firebase, WarpLink, Tenjin, and AppsFlyer. Free store reports are enough if campaign totals answer your question. A link-based service fits installs that come from links you own, and a full mobile measurement partner (MMP) fits paid ad networks. The six options do different jobs, so the table below shows which job each one does before it shows a price.

The Six Options at a Glance

Install attribution means connecting a new install back to something that caused it: a campaign, a link, or an ad. The six options below cover three different jobs. Store reports count installs per campaign. A link-based match connects an install to the link that drove it. Paid-network attribution credits an ad click or view. No single option does all three equally well, so compare the job first.

The list order runs from the most DIY to the most enterprise, and it is not a ranking. Everything here was checked on each vendor's own public page in October 2026. Vendor prices change without notice, so verify on the linked page before you commit.

Option	Best for	Matches an install to a tapped link	Paid ad network attribution	Free allowance (unit)	Paid entry price (unit)
Apple App Store Connect campaign links	Campaign totals for iOS App Store traffic	No, counts per campaign token only	No	No separate allowance listed	None listed
Google Play Install Referrer	Android installs from links you build and host yourself	Yes on Android, if you build the redirect, storage, and reports	No	No fee listed, you build the rest	None listed
Google Analytics for Firebase	In-app event and campaign reporting	Partly, needs campaign tags or a campaign event	Auto-tagging (not on iOS) and Apple Search Ads	No charge	None listed
WarpLink	Links you own: shares, referrals, email, QR codes, creators	Yes, exact on Android, scored on a first iOS install	No	$0, 10K clicks a month	$19 a month, 100K clicks
Tenjin	Paid campaigns on a small budget	Counts installs after ad clicks and views	Yes	2,000 conversions a month, then $0.04 each	$200 a month, up to 10,000 conversions
AppsFlyer	Teams about to buy paid installs at scale	Through OneLink deep links, which need a separate paid subscription	Yes, basic on Growth, cross-channel on Enterprise	12K conversions, first year only	$0.07 per conversion after that

The units are different on purpose. WarpLink counts clicks, Tenjin and AppsFlyer count conversions, and the store reports have no meter at all. Do not compare a click allowance with a conversion allowance as if they were the same thing. The price units section shows the arithmetic.

1. Apple App Store Connect Campaign Links

Best for: teams that only need to know how many App Store downloads each campaign produced.

App Store Connect Analytics lets you create a campaign link for each campaign. The campaign name is the campaign token, which can be up to 30 characters, and the link carries it in a ct= parameter next to a pt= provider token that identifies your developer account. Apple reports impressions, product page views, downloads, usage, sales, and subscriptions per campaign token, split by territory, device, or page type.

The limits matter for a small app. Campaigns appear in Analytics only after 24 hours, a metric shows only when it reaches a minimum of 5 in the date range, and a campaign shows up once it has first-time downloads from at least five individual users. Apple says rows covering very small groups can be withheld or combined to prevent identifying individuals. That is the right design for privacy and the wrong one if you need to know which single link a single install came from. It covers iOS only, and Apple's page lists no price for it.

Source: Apple, Campaign links.

2. Google Play Install Referrer, Built Yourself

Best for: Android-first teams with engineering time who want exact Android matching without a vendor.

The Play Install Referrer API returns the referrer URL of the installed package, the timestamp in seconds of the referrer click, the timestamp in seconds of when the install began, the app version at first install, and whether the user interacted with your app's instant experience in the past 7 days. Kotlin and Java apps use the Play Install Referrer Client Library. Google says the information stays available for 90 days and does not change unless the app is reinstalled, and that your app should call the API only once, on the first run after install. You need a Google Play Console account to use it.

That is the data. The work is yours: a redirect that sends visitors to the Play Store with a referrer value on the link, a click log, the first-launch call, the join between the two, and a report. Play Console also has acquisition reports for tracked channels, which count users who visited your store listing from a UTM-tagged link. It is Android only, and it gives you nothing on iOS, where nothing from the tapped link survives the App Store. Google's page lists no fee. The cost is engineering time.

Sources: Google, Install Referrer API, Install Referrer library, Play Console acquisition reports.

3. Google Analytics for Firebase

Best for: teams that already run Google Analytics in the app and want campaign sources next to their event data.

Google describes Google Analytics as an app measurement solution, available at no charge, that gives insight on app usage and user engagement. For installs, its attribution page lists four methods: auto-tagging, which is not supported on iOS; universal links and Android App Links with UTM tags that you add; Play Store referrals, where the source shows as google-play or (direct); and Apple Search Ads. For a first install you can also send a campaign_details event shortly after the install to record the source yourself.

This is measurement, not link infrastructure. The methods rely on tags you add to links, on the Play referrer, or on Apple Search Ads, so you still need something to create the links, route each platform, and carry the link into the app. It pairs well with a link-based service, since the two answer different questions. Google lists no paid tier on those pages.

Sources: Google Analytics for Firebase, How app key events are attributed.

4. WarpLink

Best for: teams whose installs come from links they own: shares, referrals, email, QR codes, and creators.

WarpLink is link infrastructure that combines deep linking, install attribution, and real-time analytics in one SDK. When someone taps a link, we record the click. On first launch the SDK asks which click led to this install, and the server walks four matching methods in order, stopping at the first success: the Play Install Referrer on Android, the identifier for vendor (IDFV) on iOS, a fingerprint of the request address, language, and timezone, and raw device signals hashed on the server. The first two are deterministic and score 1.0. The last two are probabilistic and score between 0.20 and 0.85, decaying over a match window that defaults to 6 hours and caps at 24. The full cascade is in the attribution docs and the install attribution page.

Pricing is by tracked clicks, not installs: Free is $0 for 10K clicks a month, Starter is $19 for 100K, Growth is $49 for 500K, Scale is $99 for 2M, and Unlimited is $149 with no click ceiling. Every plan includes the whole service, there is no sales call, and the details are on the pricing page. SDKs are available for iOS, Android, and React Native, all at version 1.1.0, and attribution needs an SDK key.

What WarpLink is not: it is not a paid-media MMP. It does not do ad network postbacks, the SKAdNetwork postback pipeline, cost and return on ad spend reporting, or fraud detection. It also has an honest iOS limit. The IDFV only answers a device WarpLink has already seen, so a genuine first install on iOS is a scored fingerprint match, not an exact one, and a network change between tap and launch can break it. Android installs that reach the Play Store through a WarpLink link get the exact referrer match. No advertising identifier is used on either platform.

Sources: WarpLink attribution docs, install attribution, pricing, SDK docs.

5. Tenjin

Best for: small teams that have started buying installs and want paid-campaign measurement with a free entry point.

Tenjin's pricing page lists an All-Inclusive Free plan at $0 a month with 2,000 free conversions a month and $0.04 per additional conversion. Paid plans are S at $200 a month for up to 10,000 conversions, M at $400 for up to 25,000, and L at $700 for up to 50,000, with per-conversion rates of $0.02, $0.016, and $0.014. Enterprise is a custom quote. Every plan lists access to all Tenjin products, SKAdNetwork support, fraud prevention, and unlimited dashboard users, on a month-to-month commitment.

Tenjin defines a conversion as an install after a click on an ad, so conversions reflect installs from paid marketing, and organic installs are not charged. Its mobile attribution page says it uses a last click and view-through model, supports SKAdNetwork, and updates partner integrations weekly. The pages we checked do not describe deep linking or links you own, so we do not rate Tenjin for that job. It answers a different question than a link-based service does: which ad network or campaign paid for the install.

Sources: Tenjin pricing, Tenjin mobile attribution.

6. AppsFlyer

Best for: teams that spend on ad networks and need the full measurement stack.

AppsFlyer's pricing page shows three plans. Zero is free, described as ideal for apps seeking a simple tool for owned media activities, and marked not suitable for apps with paid activities. Growth is pay-as-you-go, aimed at startups that need basic measurement as they start paid campaigns. Enterprise is custom and requires a demo request. Zero and Growth include a Welcome Package of 12K free conversions for the first year, core marketing analytics, and 30 days of select premium add-ons. After the package, Growth conversions are priced at $0.07 each. Organic installs are not counted as conversions and are free. Raw Data API access, the Protect360 fraud suite, audience segmentation, Data Locker, and custom dashboards are paid add-ons.

For links, AppsFlyer's developer documentation describes OneLink as supporting deep linking and deferred deep linking, with Unified deep linking as the current approach. The updated Zero plan does not include deferred deep linking, while the Zero plan keeps basic OneLink links, QR codes, and click reporting. Deferred deep linking needs a separate paid OneLink subscription, and AppsFlyer does not publish its price, so the 12K conversion allowance does not cover deferred OneLink matching. That puts AppsFlyer on both sides of this list: it can match a tapped link and it can measure paid networks. Enterprise attribution adds cross-channel insight, signal-loss tools, and fraud detection. The cost is the unit: you pay per conversion once the free package ends, so the bill grows with paid success. We cover the full trade-offs in AppsFlyer Alternative for Small Teams.

Sources: AppsFlyer pricing, Zero plan update, AppsFlyer deep linking developer docs, OneLink Standalone Deep Linking Suite.

How We Compared

We scored each option against the same questions, using only what the vendor publishes on its own pages. Anything a vendor does not state is marked as not listed, not guessed.

1. What it attributes to. A campaign, a tapped link, or an ad click or view. This is the first filter, because the three answer different questions.
2. Whether it can name the link behind one install. Aggregate campaign counts and per-install link matches are different products.
3. Platform coverage. How each option behaves on iOS and on Android, since the two stores hand the app different information.
4. Paid ad network attribution. Whether it supports postbacks and paid-campaign measurement.
5. The price and the unit. Free allowance and entry price, with the billed unit shown next to every number.
6. What you have to build or run. SDK work, redirect work, or nothing at all.

How the Price Units Differ

The same month of traffic costs different amounts under different units, so do the arithmetic with your own numbers. Here is one illustration using only published rates. Assume 5,000 paid installs in a month.

• Tenjin: the first 2,000 conversions are free and the other 3,000 cost $0.04 each, so $120. The S plan at $200 would cover up to 10,000, which only becomes cheaper than the free plan's overage rate above 7,000 conversions a month.
• AppsFlyer: the 12K Welcome Package covers it in the first year. After that, 5,000 conversions at $0.07 each is $350.
• WarpLink: the meter is clicks, not installs. 5,000 installs is within the Free allowance only if your links drew 10K clicks or fewer. How many clicks it takes per install is a number only your own traffic can tell you.

The store reports and the DIY referrer have no meter, and their cost is the time to read the reports or build the pipeline.

How to Choose

Match the option to where your installs come from.

• Your installs come from links you own (shares, referrals, email, QR codes, creators): pick a link-based service. WarpLink is built for this and has a free tier, and a DIY Play Install Referrer pipeline is the Android-only alternative if you have the engineering time.
• You only need campaign totals for the stores: use the free App Store Connect campaign links and the Play Console acquisition reports, and add nothing else.
• You already live in Google Analytics: add Google Analytics for Firebase for event data, and keep a link-based service for the link-to-install match.
• You are starting paid campaigns on a small budget: Tenjin's free plan or AppsFlyer's Growth plan, because both support ad network measurement.
• You buy installs at scale and need fraud tools and cross-channel reports: AppsFlyer Enterprise or another full MMP.
• You do both owned and paid: run a link-based service for owned links and an MMP for paid, and do not expect either to replace the other.

If you are unsure, start with the cheapest option that answers your actual question. The free tiers above cost nothing to try, and an install you cannot attribute is still an install. For the implementation behind the link-based path, read Install Attribution Without an MMP.

Sources

• Apple campaign links: developer.apple.com/help/app-store-connect-analytics/acquisition/campaign-links
• Google Play Install Referrer: developer.android.com/google/play/installreferrer, library guide
• Play Console acquisition reports: support.google.com/googleplay/android-developer/answer/6263332
• Google Analytics for Firebase: firebase.google.com/docs/analytics, support.google.com/analytics/answer/10311900
• WarpLink: warplink.app/pricing, warplink.app/install-attribution, warplink.app/docs/concepts/attribution
• Tenjin: tenjin.com/pricing, tenjin.com/mobile-attribution
• AppsFlyer: appsflyer.com/pricing, dev.appsflyer.com deep linking overview

Frequently Asked Questions

What is the best install attribution option for a small app?

It depends on where your installs come from. For installs from links you own, such as shares, referrals, email, QR codes, and creators, a link-based service like WarpLink matches each install to the link that drove it from $0 a month. For paid ad networks you need a full mobile measurement partner such as AppsFlyer or Tenjin. If campaign totals are enough, the free App Store Connect and Google Play reports may be all you need.

Can I attribute installs without the IDFA?

Yes. WarpLink uses no advertising identifier. It matches on the Play Install Referrer on Android, the identifier for vendor on iOS, and a hashed fingerprint of the request address, language, and timezone, so no App Tracking Transparency prompt is needed. App Store Connect campaign links also report per campaign token rather than per person. The trade-off is that a first install on iOS is a scored match, not an exact one.

Is the Play Install Referrer enough on iOS?

No. The Play Install Referrer is a Google Play API, so it only exists on Android. Nothing from the tapped link survives the App Store trip to the app, so an iOS install is matched by other signals. WarpLink uses the identifier for vendor for a device it has already seen, then a fingerprint with a confidence score of 0.20 to 0.85.

Can store campaign reports identify a creator's exact link?

Only in aggregate. App Store Connect lets you create a campaign token for each campaign, up to 30 characters, so one token per creator works. It then reports impressions, product page views, and downloads per token, but only once a campaign has first-time downloads from at least five users, and Apple may withhold small groups to avoid identifying individuals. It never tells you which person installed.

When does a small app need a mobile measurement partner?

When you buy installs from ad networks and need what they require: network postbacks, SKAdNetwork reporting, cost and return on ad spend data, and fraud detection. WarpLink does not cover those. AppsFlyer says its free Zero plan is not suitable for apps with paid activities, which is the same boundary seen from the other side.

What does unattributed mean after a deferred link click?

It means the install did not match any click. With WarpLink the usual causes are an IP address that changed between the tap and the first launch, IPv4 at one end and IPv6 at the other, a visitor who never reached the Play Store, or a click that expired after the match window, which is 6 hours by default and 24 at most. A miss is recorded as an attribution attempt and returns no match. No install is created, so the dashboard counts only matched installs.

Related Guides

• Install Attribution Without an MMP: how the match cascade works, one rung at a time.
• Install Attribution Without IDFA: what iOS still gives you after App Tracking Transparency.
• How to Track App Installs From a Link: the step-by-step setup on iOS and Android.
• Creator and Influencer Link Attribution: one link per creator, measured to the install.
• AppsFlyer Alternative for Small Teams: the full cost comparison against the largest MMP.
• Best Deep Linking Alternatives for Small Teams: the wider vendor roundup, including deferred deep links.
• Install attribution: the product page for WarpLink's match cascade.
• 5 Deferred Deep Linking SDKs Compared: how five vendors carry a link through the install.
• 5 Mobile Link Analytics Services Compared: click reporting and install matching, service by service.

Getting Started

If link-based attribution fits your installs, create a free account, register your app, create an SDK key, and add the iOS, Android, or React Native SDK. The quickstart covers the setup, and the free tier's 10,000 monthly clicks include deep linking, install attribution, and real-time analytics, so you can test it against everything above before you decide.