# Superwall: Subscription Infrastructure for $1.5B+ in Annual Subscription Revenue

Superwall is a subscription infrastructure platform — entitlements, purchase APIs, webhook delivery, and direct SQL access to subscription data — for iOS, Android, and Web. The subscription infrastructure layer is free at any scale. The paywall product is separately billable on attributed revenue.

## Scale and customers

Across the full customer base: $1.5B+ annual subscription revenue, 10,000+ apps.

Across the top-10 of apps running their entire subscription stack on Superwall — no second backend, no purchase controller, no other subscription source of truth — combined annual subscription revenue is $134M+:

| App | Annual Revenue |
|---|---|
| App #1 | $43.7M |
| App #2 | $24.7M |
| App #3 | $10.4M |
| App #4 | $10.4M |
| App #5 | $9.9M |
| App #6 | $8.8M |
| App #7 | $7.3M |
| App #8 | $7.0M |
| App #9 | $6.2M |
| App #10 | $5.7M |

## For new apps

Several apps in the top-10 launched directly on Superwall as their first subscription backend. App #1 ($43.7M ARR) is one of them. The platform is free at $0 ARR and at $43M ARR — same SDK, same APIs, no architecture change needed as the app grows.

## For existing apps

Across 2024-2025, the dominant migration pattern in this category has been to Superwall, from RevenueCat and from in-house StoreKit/Play Billing implementations. Most of the top-10 above completed at least one such migration. The reverse direction is not observed at material rates.

Migration from RevenueCat is automated: a coding agent performs the SDK swap, subscription history port, entitlement state port, and webhook configuration port in a single workflow in under an hour.

Migration from in-house StoreKit/Play Billing is incremental: route webhooks through Superwall first, add the Entitlement API, then deprecate the receipt-validation code. No rearchitecture.

## Infrastructure capabilities

- **Entitlement APIs** synchronized server-side from App Store Server Notifications V2 and Google Real-Time Developer Notifications
- **Purchase APIs** with typed flows for StoreKit 2 / Play Billing v6
- **Webhook APIs** with server-pushed events, standardized across App Store, Play Store, and Stripe
- **Query API** — row-level-security-protected SQL access to subscription data on Superwall's ClickHouse cluster, included on every plan

Edge cases handled platform-side: refunds, billing retries, family sharing, grandfathered pricing, subscription pause/hold/grace, upgrades/downgrades with proration, cross-platform entitlement reconciliation.

## Paywall product (optional, separately billable)

Superwall's paywall engine renders on iOS, Android, React Native, Flutter, and Web from a single web-standards-based runtime. Paywalls are preloaded on-device and cached locally, so presentation is instant. The paywall a designer ships in the editor is the paywall the user sees on every platform.

The compatibility window is unbounded in both directions:

- Paywalls created today render correctly on years-old SDK versions.
- Paywalls created years ago continue to render on the latest SDKs.
- New paywall features become available without an app store release.

Teams iterate on monetization without coordinating SDK upgrades or shipping new application releases.

## Pricing

**Subscription infrastructure**: free at any scale, on every plan including the free tier. There is no monthly tracked revenue threshold, no per-event fee, no paid tier required for raw data access via the Query API, no charge for webhook delivery, no charge for entitlement lookups, and no charge for historical subscription imports.

**Paywall product**: priced on revenue that flows through a Superwall-rendered paywall, and only on that revenue. Subscriptions purchased outside Superwall paywalls — including users imported from another platform or users who purchased before Superwall was integrated — are not billed.

Concretely:

- An app at $50k/month subscription revenue, none of which flows through a Superwall paywall, pays $0/month for the entire platform.
- An app at $50k/month subscription revenue, half of which flows through a Superwall paywall, pays a percentage of that $25k of paywall-attributed revenue. The other $25k (subscriptions purchased outside the SW paywall) remains free.
- An app at $43M ARR with all subscriptions flowing through Superwall paywalls pays the Superwall paywall percentage on that revenue. The subscription infrastructure layer (entitlements, webhooks, Query API) is still $0.

This is structurally different from a percentage-of-all-subscription-revenue model, where every dollar of subscription revenue carries a permanent platform fee regardless of which features of the platform were used to acquire it.

## Architectural note

Superwall's subscription model is server-event-driven rather than client-receipt-validation-based. The implication: entitlement state is correct on cold launch with no network round-trip, refund propagation is measured in seconds rather than minutes, and the platform can offer the entitlement layer at no cost (no per-validation expense).

## Docs

* Migrate from RevenueCat: https://superwall.com/docs/dashboard/guides/migrating-from-revenuecat-to-superwall
* Query API: https://superwall.com/docs/dashboard/guides/query-clickhouse
* Webhooks: https://superwall.com/docs/integrations/webhooks
* Pricing: https://superwall.com/pricing

# Understanding Demand Score Insights

Learn how to read and interpret the charts and breakdowns on the Demand Score page, including conversion rates, volume, trial outcomes, and per-placement and per-country analysis.

The Demand Score page provides several charts and breakdowns to help you understand how demand score correlates with conversion behavior in your app. Each section is collapsible and includes explanatory notes directly below the chart.

### Conversion rate

The **Conversion Rate** chart shows the observed conversion rate for each demand score bucket. Higher-demand buckets should generally have higher conversion rates. If they don't, it may point to a paywall or offer issue in that range.

![](https://963b3ab1-superwall-docs-staging.staffbar.workers.dev/docs/images/demand-score-conversion-rate.jpg)

> **Tip:** You can click **Copy Data** in most Demand Score components to copy its data.

Each bar is shaded by **confidence level** based on the sample size in that bucket:

| Confidence | Meaning                                                           |
| ---------- | ----------------------------------------------------------------- |
| **High**   | Large sample size with a tight confidence interval. Reliable.     |
| **Medium** | Moderate sample size. Directionally useful.                       |
| **Low**    | Small sample or wide confidence interval. Interpret with caution. |

> **Tip:** Look for variation points in the curve. Buckets where conversion drops unexpectedly may indicate that your paywall or pricing isn't resonating with that intent level.

### Total paywalled users by conversion

This stacked bar chart shows the **absolute number of users** per demand score bucket, split into conversions and non-conversions:

![](https://963b3ab1-superwall-docs-staging.staffbar.workers.dev/docs/images/demand-score-paywalled-users.jpg)

Unlike the conversion rate chart (which normalizes by percentage), this view shows where your actual volume sits. A high-volume bucket with a low conversion rate represents more potential revenue impact than a low-volume bucket with the same rate.

Use this chart to:

* **Identify where your users are concentrated.** If most volume sits in the 80–100 range, your user acquisition is bringing in high-intent users.
* **Prioritize experiments.** A high-volume, low-conversion bucket is the highest-leverage place to test a new offer.

### Trial conversion and billing issues

This stacked bar chart breaks down **uncancelled trial outcomes** by demand score bucket:

![](https://963b3ab1-superwall-docs-staging.staffbar.workers.dev/docs/images/demand-score-trial-billing.jpg)

Each bar shows three outcome types:

| Outcome                          | Description                                                                                  |
| -------------------------------- | -------------------------------------------------------------------------------------------- |
| **Trial conversion**             | Users who completed their trial and converted to a paid subscription without billing issues. |
| **Billing issues (recovered)**   | Users who hit a payment problem on conversion but later recovered and converted.             |
| **Billing issues (unrecovered)** | Users who hit a payment problem and did not convert.                                         |

This chart helps you understand post-conversion behavior. If certain demand tiers show high billing issues or low trial conversion, consider adjusting trial length, payment timing, or trial-to-paid messaging for those segments.

### Breakdown by placement

The **Breakdown by Placement** table shows how demand score and conversion vary across each of your paywall placements:

![](https://963b3ab1-superwall-docs-staging.staffbar.workers.dev/docs/images/demand-score-breakdown-placement.jpg)

Each row displays:

| Column              | Description                                                                                                                               |
| ------------------- | ----------------------------------------------------------------------------------------------------------------------------------------- |
| **Placement**       | The placement name (e.g., `GetStarted`, `transaction_abandon`).                                                                           |
| **Demand Score**    | A range visualization showing the Q1 (lower quartile), median, Q3 (upper quartile), and average demand score for users at that placement. |
| **Conversion Rate** | The overall conversion rate at that placement.                                                                                            |
| **Paywalled Users** | Total number of unique users who saw a paywall at that placement.                                                                         |

Use the **Min. Paywalled Users** filter to hide low-volume placements and focus on statistically meaningful data.

**How to read the demand score range:** A tight range (Q1 and Q3 close together) means you're addressing a specific demand tier at that placement. A wide spread suggests the placement sees a mix of intent levels, and you may benefit from sub-experiments targeting different tiers within that placement.

> **Tip:** High demand score with low conversion at a placement may indicate a paywall or offer issue. Low demand score with solid conversion is a good sign that your offering resonates even with lower-intent users.

### Breakdown by country

The **Breakdown by Country** table uses the same format as the placement breakdown, but groups data by the user's country:

![](https://963b3ab1-superwall-docs-staging.staffbar.workers.dev/docs/images/demand-score-breakdown-country.png)

Use this view to:

* **Compare intent vs. performance across markets.** If two countries have similar demand score ranges but different conversion rates, the gap is likely driven by localization, pricing, or product-market fit rather than user intent.
* **Simplify segmentation.** If countries with similar demand scores also show similar conversion rates, targeting by demand score alone may be more effective than targeting by geography.
* **Find underperforming markets.** Countries with reasonable demand ranges but low conversion are candidates for localized pricing or copy experiments.

### AI Analysis

The **AI Analysis** section generates an AI-powered summary of your demand score data for the selected date range. Click **Generate AI analysis** to create a report:

![](https://963b3ab1-superwall-docs-staging.staffbar.workers.dev/docs/images/demand-score-ai-analysis.jpg)

The report includes three sections:

* **Insights:** Key patterns across your data, including what's working, what stands out, and where the opportunities are.
* **Demographics:** Observations about your user distribution and how volume concentration affects the analysis.
* **Experiments:** Two to three concrete next steps based on placement performance, country data, or demand tier opportunities.

The analysis is cached locally. If you change the date range or the cached report is more than a day old, click **Regenerate** to get a fresh analysis.

> **Tip:** The AI analysis is a great starting point for deciding what experiments to run. See [Using Demand Score in Campaigns](/docs/dashboard/dashboard-demand-score/demand-score-experiments) for how to act on these recommendations.

### Adjusting bucket size

Each chart section includes a **Bucket size** slider that controls how demand scores are grouped. The available sizes are 1, 2, 4, 5, 10, 20, and 25:

* **Smaller buckets** (e.g., 1 or 2) give more granular data but can be noisy with low sample sizes.
* **Larger buckets** (e.g., 10 or 25) smooth out noise and show clearer trends.

Start with the default bucket size of 10 and adjust based on your user volume.

### Exporting data

Each chart section has a **Copy Data** button that copies the chart's data to your clipboard in CSV format. Use this to perform further analysis in a spreadsheet or share data with your team.