Insights: see what your web store earns

Studio → Insights shows, by day or week:

• store visits and checkout starts;
• paid orders, gross, refunds and net of refunds;
• conversion from visit to paid, average revenue per paying player (ARPPU), first-purchase rate and repeat buyers within 30 days.

Every figure can be broken down by source, campaign, pack and identity method (signed token, Player ID or anonymous).

How the numbers are counted

• Visits come from real browsers only. A visit is recorded only after the storefront has loaded in a browser and the player interacted or stayed 8 seconds with the page visible. Page requests from crawlers, scripts and hosting networks are never counted. Very slow connections can be slightly undercounted.
• Revenue comes from orders. Paid orders and refunds come from the same ledger as your order list, so Insights reconciles with Orders to the cent. A refund is counted once, on the day it happened.
• Currencies stay separate. Each currency is reported on its own; nothing is converted.
• Test and live stay separate, and your team's own traffic is excluded by default.
• Memberships are not included yet; this report covers one-off packs.
• Sources come from utm_* parameters on the store link, the in-app placement your SDK launch passes, or the referring site.
• Campaigns are recorded only when they match an offer, promo code or gift code you actually published (see Campaigns).

"Kept vs a 30% app-store fee" compares your net of refunds at the published Appside rate with a 30% app-store commission. It is an estimate before taxes, currency conversion and other costs, not a settlement statement.

Segments and campaign lift

• Your own attributes: Studio → Segments lets you declare up to 32 custom player attributes (string, number, boolean or date) next to the built-in level, spenderTier, country and newPlayer. Send them in the signed player token or the Player ID verify response.
• Segments: a segment is a set of rules that all must match, for example "score greater than 10". Preview shows how many verified players from the last 30 days match. Saved segments are versioned, and offers, promo codes and launch-kit campaigns pin a version.
• Holdout: each campaign can keep a holdout (default 10%). Those players see your base offer, chosen by a stable hash of store, campaign and player, so a player stays in the same group across visits and sign-in methods.
• Lift: Insights → Campaign lift compares paid rate and net revenue per eligible player, treatment vs control, with 95% confidence intervals. Below 30 players or 5 payers per group it says "insufficient data" instead of showing a number. Decide on a fixed end date before you look: checking repeatedly and stopping early makes false wins more likely.

What Appside records on your store (for your privacy notice)

When a player uses your Appside store, Appside records commerce events for your reporting:

• store view, pack view, checkout started, order paid and order refunded;
• with your store, version, pack, amount and currency;
• with how the player signed in, the source and the campaign.

The player is identified by a pseudonymous identifier (a one-way hash), not by name or email. Pseudonymous is not anonymous: it is still linked to a player within your store.

• Retention: detailed events are kept for 400 days, then reduced to daily totals without player identifiers.
• Appside's own analytics: Appside sends Purchase and Refund events (store, pack, amount, currency, campaign, sign-in method, test or live, the pseudonymous identifier) to its product analytics, Mixpanel. They are used to operate and improve Appside.
• Your privacy policy: describe this processing in your own policy for your web store, as you do for your app.