WebSDK

The Complete Guide to Adobe WebSDK Migration

Legacy Adobe libraries like AppMeasurement and AT.js are reaching end of life. WebSDK (Alloy.js) is the unified replacement that sends data through Adobe Edge Network. This playbook walks you through every stage of migration, from planning to post-cutover optimization.

Why Migrate to WebSDK

Adobe has been clear: AppMeasurement, AT.js, and DIL are legacy libraries that will not receive new features. All future Adobe Experience Cloud capabilities, including real-time personalization, server-side event forwarding, and advanced identity resolution, require WebSDK.

Key Drivers for Migration

Beyond end-of-life timelines, WebSDK solves real problems today. Safari ITP deletes client-side cookies after 7 days, breaking visitor stitching for legacy implementations. WebSDK with first-party server-side collection extends cookie lifespan to 2 years, recovering data that was previously invisible.

  • Single library: WebSDK replaces AppMeasurement, AT.js, DIL, and Visitor API with one tag. Fewer HTTP requests, smaller page weight.
  • Edge Network routing: Data is sent once to Adobe Edge Network and distributed to Analytics, Target, Audience Manager, and any other configured destination.
  • Event Forwarding: After migration, you can forward data server-side to Google, Meta, TikTok, and other platforms without additional client-side tags.
  • Future-proofing: RT-CDP, Customer Journey Analytics, and Journey Optimizer all depend on XDM-formatted data flowing through Edge Network.

Planning Your Migration

A successful migration starts with a thorough audit of your current implementation. You need to know exactly what data is being collected today before you can map it to the new XDM schema.

Step 1: Audit Current Implementation

  • Document every Analytics variable: List all eVars, props, events, and custom dimensions currently in use. Cross-reference with your Solution Design Reference (SDR).
  • Catalog Target activities: Record all active A/B tests, experience targeting, and Automated Personalization activities.
  • Map data element dependencies: Identify which Launch data elements feed which rules and extensions.

Step 2: Define Your XDM Schema

XDM (Experience Data Model) is the standard data format for all Adobe Experience Platform services. Your schema design determines how data flows across the entire ecosystem. Start with Adobe-provided field groups and extend only where necessary.

Common Pitfall

Teams often try to replicate their existing variable structure 1:1 in XDM. This misses the point. XDM is an opportunity to clean up years of analytics debt. Consolidate redundant variables, standardize naming conventions, and remove tracking that no longer serves business objectives.

Step 3: Create a Migration Backlog

Break the migration into discrete work items. Each Launch rule should become a ticket that includes the current implementation, the target XDM mapping, and acceptance criteria for data validation.

Parallel Tracking Strategy

The safest migration approach is to run both implementations simultaneously. Old tags continue sending data to Adobe Analytics while new WebSDK tags send to a parallel report suite. This lets you validate data parity before cutting over.

How Parallel Tracking Works

Create a secondary report suite in Adobe Analytics. Configure your WebSDK datastream to route Analytics data to this secondary suite. For 2-4 weeks, both implementations fire on every page. Compare key metrics daily: page views, events, eVar values, and segments.

  • Tolerance thresholds: Define acceptable variance before you start. A 1-2% difference in page views is normal due to timing differences between libraries.
  • Focus on business metrics: Do not chase 100% parity on every variable. Prioritize revenue events, conversion funnels, and campaign tracking.
  • Document discrepancies: Some differences are expected. WebSDK handles bot filtering and identity differently. Document these so stakeholders understand the changes.

XDM Schema Design

Your XDM schema is the foundation of the entire Adobe Experience Platform ecosystem. Getting it right saves months of rework later.

Field Groups

Adobe provides standard field groups for common use cases: web analytics, commerce, consent, and identity. Start with these and add custom field groups only for data that does not fit the standard model.

  • AEP Web SDK ExperienceEvent: Covers page views, link clicks, and web-specific context like browser and device data.
  • Commerce Details: Product views, cart additions, purchases, and revenue. Maps directly to Adobe Analytics product string.
  • Identity Map: Allows multiple identity namespaces (ECID, email hash, CRM ID) per event. Critical for cross-device stitching.
  • Consent and Preferences: Captures user consent state per purpose, enabling consent-aware data collection.

Custom Extensions

For business-specific data like internal campaign codes, customer tier, or account type, create a custom mixin. Keep the namespace clean and documented. Every custom field should have a clear business justification.

Identity Configuration

Configure your identity namespace strategy early. Decide which identifiers are primary (typically ECID for anonymous, CRM ID for authenticated). The identity graph behavior, whether to merge or keep profiles separate, has significant downstream impact on segmentation and activation.

Validation and QA

Thorough validation is the difference between a smooth cutover and a data crisis. Use multiple tools and validation layers to catch issues before they reach production.

Adobe Experience Platform Debugger

The browser extension shows every WebSDK network request in real time. Verify that XDM payloads contain the expected field values, that the correct datastream is targeted, and that responses include the expected Target decisions or personalization offers.

Assurance Sessions

  • End-to-end tracing: Assurance connects your browser session to the Platform UI, showing exactly how each event is processed through Edge Network, Analytics processing rules, and Target decisioning.
  • Schema validation: Assurance flags XDM fields that do not match your schema definition, catching type mismatches and missing required fields.
  • Identity resolution: View how Edge Network resolves identity graphs for each event, confirming that visitor stitching works as expected.

Automated Regression Testing

Build a test suite that validates critical data points after each deployment. Tools like Selenium or Playwright can automate page loads and verify that the expected network requests fire with correct payloads. This prevents regressions during the migration period.

Cutover Strategy

When parallel tracking confirms data parity, it is time to plan the cutover. A phased approach reduces risk and gives your team time to respond to any unexpected issues.

Phase 1: Low-Traffic Pages (Week 1)

Start with informational pages that do not affect revenue tracking. Blog posts, about pages, and support documentation are ideal candidates. Monitor for 5-7 days.

Phase 2: Mid-Funnel Pages (Week 2-3)

Move to category pages, search results, and product listing pages. These pages have higher traffic and more complex tracking requirements. Validate event firing and eVar population before proceeding.

Phase 3: Revenue Pages (Week 3-4)

Finally, migrate checkout, cart, and purchase confirmation pages. These carry the highest risk, so ensure your rollback plan is ready before deploying.

Rollback Plan

Keep the legacy implementation in a separate Launch library that can be published within minutes. If WebSDK introduces a data collection issue on critical pages, you can revert to legacy tags immediately while investigating the root cause. Never delete legacy rules until at least 30 days after successful cutover.

Post-Migration Optimization

Migration is not the finish line. Once WebSDK is live across all pages, you unlock capabilities that were impossible with legacy libraries.

Edge Configuration Tuning

  • Response compression: Enable Gzip or Brotli compression on Edge Network responses to reduce payload size for Target offers and personalization content.
  • First-party domain: Configure CNAME-based first-party data collection to extend cookie lifespan and bypass ITP restrictions.
  • Data prep: Use datastream-level data prep to transform, filter, or enrich data before it reaches downstream services.

Event Forwarding Readiness

With data flowing through Edge Network, you can now add server-side Event Forwarding rules. Send conversion events to Google Ads, Meta CAPI, or any HTTP endpoint without adding client-side tags. This reduces page weight and improves data quality since server-side calls are not affected by ad blockers.

Cleanup Legacy Tags

After 30 days of stable WebSDK operation, archive (do not delete) your legacy Launch rules and extensions. Remove AppMeasurement, AT.js, and Visitor API libraries from your Launch property. Monitor page load performance. You should see measurable improvements in Time to Interactive and Total Blocking Time.

Continue Reading

// HIRE US

Ready to migrate to WebSDK?

Our team has guided dozens of enterprise migrations from legacy Adobe libraries to WebSDK. Let us help you plan and execute a seamless transition.

"We've helped dozens of teams to structure their data. Let's see how we can help you."

Kamil Glowacki
Kamil Glowacki
CEO & Founder, DataCoreUnity

By submitting this form, you agree to our Privacy Policy.