Pay-per-event (PPE) pricing model – alpha phase

The pay-per-event (PPE) model is currently in the alpha phase. It provides a flexible way to monetize your Actors on Apify Store. Instead of charging per result, you can set prices based on specific events triggered explicitly by your Actor’s code.

PPE vs. pay per result (PPR)

Unlike PPR, which charges based on the number of results produced, PPE lets you define pricing for individual events. You can charge for specific events directly from your Actor by calling our PPE charging API. The most common events will most likely be Actor start, dataset item, external API calls, etc.

You can see a detailed explanation of the different pricing models in our user-facing docs.

How is profit computed?

Your profit is calculated similarly to PPR:

profit = (0.8 * revenue) – platform costs

where

• Revenue: what you charge for events via our API. 80% of this goes to you, 20% to Apify.
• Platform costs: this is the underlying platform usage of the runs of this Actor. The unit pricing to compute the costs is the same as for PPR.

Only the activity of paid users is considered when calculating your profit. Free plan user activity does not count. Also, revenue from users who fail to pay for Apify services (typically fraudsters), is not included in payouts, as per our T&C.

How to set up PPE for your Actor

• In the alpha stage, just ping Saurav on Discord with the pricing format and the date you want the Actor to be PPE from: Define pricing using the following JSON schema, which specifies event titles, descriptions, and their prices in USD.
• Here is an example of a PPE JSON schema on vdrmota/contact-info-scraper:
  • The keys (e.g. actor-start-gb ) are the event names that you will use in charging via API
  • eventTitle and eventDescription will be visible to the user and are describing your events
  • eventPriceUsd is the price per single event of this type

{
    "actor-start-gb": {
        "eventTitle": "Actor start per 1 GB",
        "eventDescription": "Flat fee for starting an Actor run for each 1 GB of memory.",
        "eventPriceUsd": 0.005
    },
    "pages-scraped": {
        "eventTitle": "Page scraped",
        "eventDescription": "Cost per web page processed for data extraction.",
        "eventPriceUsd": 0.002
    },
    "page-with-residential-proxy": {
        "eventTitle": "Extra: Page opened with residential proxy",
        "eventDescription": "Extra cost per page scraped using a residential proxy.",
        "eventPriceUsd": 0.002
    },
    "page-with-browser": {
        "eventTitle": "Extra: Page opened with browser",
        "eventDescription": "Extra cost per page scraped with a browser.",
        "eventPriceUsd": 0.003
    },
    "page-wait-event": {
        "eventTitle": "Extra: Extended loading per page",
        "eventDescription": "Extra cost per page when choosing a more resource-intensive load event ('load' or 'networkidle').",
        "eventPriceUsd": 0.005
    }
}

Setting up PPE for your Actor

• ChargingManager: You can copy/paste our ChargingManager (you can also test it as an Actor). We use it in our Actors for easy management of charges code. It contains a class ChargingManager and a function pushDataPPEAware that handles PPE charging by managing pricing, tracking usage, and ensuring accurate event charges via API. It calculates remaining budgets, prevents duplicate charges with idempotent keys, handles API errors, and stores metadata in a dataset for auditing. It also handles PPR and allows for a seamless transition between the pricing models. It adapts to the environment and automates dataset management, providing a reliable, transparent, and developer-friendly solution for charging in the PPE model. An SDK version is coming soon.
• If you decide not to use the ChargingManager code, you will need to hit our API explicitly and handle all the edge cases yourself. Please study the code of ChargingManager to see how it’s done.