GETTING STARTED

ScaleDynamics Impulse Engine: Overview

The ScaleDynamics Impulse Engine is a real-time hardware intelligence layer for the modern web.

It solves Hardware Blindness—the critical industry flaw where web applications, ad servers, and Edge AI models treat $1,500 flagship smartphones and 6-year-old budget devices exactly the same. By deterministically profiling the physical capabilities of a user's device at load time, the engine allows you to dynamically route workloads, optimize JavaScript execution, and apply accurate pricing to premium ad inventory.

Impulse is based on the database of our community benchmark Speedpower.run (opens in a new tab), which contains millions of samples from multiple devices (mobile, desktop, laptop, TV...). On the device, Impulse runs a profiling, collect some browser data and request our ScaleDynamics servers to get the metrics. Our servers are using Speedpower.run benchmark database to compute the metrics for the device

All profiling is 100% privacy-safe, utilizing zero cookies, zero fingerprinting, and collecting zero Personally Identifiable Information (PII). Have a look to Security & Compliance for details.

The Deployment Strategy: A Two-Phase Rollout

We built the Impulse Engine to be adopted with zero risk to your current infrastructure. Integration happens in two distinct phases to ensure absolute safety and measurable ROI before any code is actively executed in your production environment.

Phase 1: The 24-Hour Passive Audit

You begin by deploying sd-agent.js. This is a purely observational telemetry agent that runs asynchronously in the background. It does not modify your DOM, block your main thread, or expose any variables to your Ad Server. After 24 hours of traffic collection, ScaleDynamics provides a comprehensive Hardware Audit, revealing your exact monetization blind spots and INP latency risks.

Phase 2: Production Integration

Once you have validated the opportunity using the audit data, you replace the passive agent with the core Impulse Engine (sd-metrics.js). This script evaluates the device and instantly exposes actionable hardware metrics to your browser environment, making them available to your Ad Server, Tag Manager, or frontend framework before the page fully renders.

The Dual-Scoring Architecture

To serve both commercial and technical teams without creating friction, the Impulse Engine translates raw hardware telemetry into two distinct sets of metrics:

  • AdTech Variables: sd_inventory_tier: fixed on a standard 1-to-4 scale (where 1 is Ultra-Premium). sd_device_tier: fixed on a standard 1-to-4 scale (where 1 is High-End). These variables are designed for SSPs, DSPs, and Ad Servers to ensure stable audience segmentation and pricing rules over time. Have a look to AdTech metrics reference for more details.

  • Engineering Metrics: bdp, mdc, air: Open-ended, continuous scales where higher is better, providing developers with highly granular, future-proof benchmarks for JavaScript concurrency and AI routing. Have a look to Engineering metrics reference for more details.

Next Step: Choose Your Installation Path

Because the deployment process for Phase 1 and Phase 2 shares the same ecosystem, you simply need to select the installation method that matches your tech stack:

Developer & API References

Looking for advanced configurations, async promises, or custom event tracking? Skip the guides and dive straight into the technical specs:

Google Tag Manager (GTM)