IAP Personalization FAQ

Product

How does Assetario personalization work?

Assetario is a SaaS AI/ML platform that enhances offer recommendations for gaming and app developers and publishers. No SDK integration is required, ensuring a smooth and efficient process. Key aspects of our personalization process include:

  • Outcome: Machine learning models predict user purchase behavior, targeting the perfect personalized offers and content for each user, and driving user engagement and revenue growth.

  • Data Sharing: You provide us with historical event data - simple payments and login data, all anonymized, that you certainly track.

  • Resource Involvement: Minimal resource requirements from your team, involving a total of 7 Man-Days (MD) across engineering, monetization, and data analysis roles (number of MDs based on experience from previous clients).

  • Integration Steps: Our team handles data processing, model training, and dashboard preparation. Integration involves exporting data to Assetario, sharing offer definitions, and integrating a single API call in your app.

Why should I choose Assetario over an in-house solution?

In-House Solutions Limitations:

  • Divert resources from core ML and analytics projects.

  • Yield smaller potential revenue uplift.

  • Lead to higher long-term costs of maintenance and development.

Assetario's Approach and Services:

  • Performance-based pricing: Pay only for measurable results.

  • Expertise in gaming and ML fields with a proven track record.

  • Assetario Personalization services increase the effectiveness of your internal data teams as they can focus on other roadmap items.

  • Clients achieve higher revenue uplift at a lower cost compared to using their tools.

  • Large data pool Assetario uses to improve models.

  • Services include ETL, predictions, A/B testing, ML models, analytics, and dashboards.

Assetario's Infrastructure Know-How:

  • Provides all necessary services and tools without the development cost.

  • Ensures scalability, high throughput, and low latency.

  • Features state-of-the-art models and A/B testing capabilities.

  • Guarantees technical correctness in A/B testing and ML models.

  • Maintains a highly scalable and available infrastructure for data processing and ML.

What distinguishes Dynamic from Contextual personalization?

Product

Dynamic Personalization

Contextual Personalization

Purpose

Get the best SKU offer at any point in the user's journey based on the content and price.

Scale prices of all your in-app purchases based on the country and device of each user to maximize user LTV.

ML Model

Trained on user historical and real-time behavioral data are used as input parameters.

Trained on user historical and real-time contextual data are used as input parameters.

Usage

API is called each time the game needs to offer something for a user.

API is called just at the start of the first user session.

Suitability

Suited for games with a vast range of offers.

Ideal for all apps and games mainly monetizing through subscriptions, store offers, and hard currency purchases.

Benefits

Aids in predicting user purchase preferences and eases the live ops team's workload due to outsourcing segmentation to Assetario’s models.

Earlier and higher revenue uplift as all game or app revenue is personalized from the moment of going live.

Data Requirement

Requires at least 6 months of historical data.

Requires at least 3 months of historical data.

Data and Models

Is data shared between clients?

No, we do not share data between clients. Assetario asks to receive anonymized data. We don’t utilize any personally identifiable information. Assetario is GDPR, CCPA, COPPA compliant. Our data policies fully meet and exceed these policies and other major data privacy regulations.

What data is utilized for personalization?

  • For contextual personalization:

    • API calls use contextual data, like user/user device information and location.

      • Training utilizes historical data including spending features, economic data, seasonality data, and other metadata (See details here).

  • For dynamic personalization:

    • API calls use current spending, device information, user location, and currency balances.

    • The training utilizes historical data like spending features, level-ups, game balances, user playstyles, favorite items, economic data, seasonality data, and other metadata (See details here).

Can Assetario adapt to in-game or app changes, user behavior alterations, or UA changes?

Yes, our models are retrained daily within a predefined training window, swiftly adapting to changes while taking into account seasonality and user behavior over a broader time range. Our models are also periodically updated, ensuring that our clients always have the latest version available.

How can I evaluate or track the effectiveness of Assetario's personalization?

Effectiveness is assessed through ongoing A/B testing, which involves comparing ARPU between control and personalized groups across various cohorts, including daily, weekly, monthly, quarterly, and yearly. This data-driven performance monitoring process is important in demonstrating the value of our solutions. Additionally, all clients have access to dashboards for monitoring the performance of Assetario models.

Integration Process

How long is the integration process?

The integration process usually takes about a month, culminating in the pilot A/B test. Most of this time is dedicated to model training and adjustments on our end.

What are the steps involved in integration?

The integration process is simple: share your data with Assetario, and we handle the rest, including data adjustment and aggregation, model training, dashboard preparation, and predictions on which users or users are likely to make purchases.

What's the resource requirement for integrating Assetario on partners end?

Drawing from our clients' experiences, typical integration entails approximately 5 MD for your engineer, 1.5 MD for your monetization specialist, and 0.5 MD for your data analyst. A total of 7 MDs of your team's time.

How challenging is the integration?

The integration is straightforward. You'll need to export data to Assetario, share the offer definition table, and integrate and QA just one API call within your application (See details here).

What amount of data is required?

The data requirement varies based on the mobile game or app's size, but generally, six months of event data is sufficient.

Is there a need to integrate an SDK?

No, integration involves only one API for predictions.No, integration only involves one API that provides predictions (See details here).

Last updated