From deeper Fraud Signal context to perpetual KYB and a more powerful AI Assistant, Q1 was a big quarter. See everything we shipped in our latest blog post.

Testing workflow changes before they go live is now easier than ever.

We're excited to introduce Backtesting, your new tool for validating workflow logic against real historical data before making any changes active. Backtesting allows you to replay past events through a test version of your workflow and compare outcomes side by side, all without touching your production environment.

With Backtesting, you can:

• Validate before go-live: Confirm your workflow is behaving as expected by running historical data through it before setting it active.
• Experiment safely: All results are fully informational and excluded from live queues, analytics, and billing. There's no risk in testing.
• Optimize continuously: Measure how rule or threshold changes affect outcomes and alert volumes without waiting for new data to come in.
• Investigate outcome shifts: Click into any individual evaluation to view rule explainability and see exactly which rules fired and why.

To get started, open any workflow, click "Test," and select the Backtest option from the Testing Suite. You can also check out this guide for more information.

We’ve redesigned Alerts within Investigations to help your team move faster, stay focused, and take action with confidence.

• See what matters, faster
  A cleaner, more intuitive layout makes it easy to follow both simple and complex alert journeys. Key steps—like Actions, Routers, and Recon—are now easier to spot so your team can quickly understand what’s happening.
• Simplified timeline for quicker decisions
  The new timeline view reduces visual clutter and clearly aligns statuses, helping you scan alerts and make decisions more efficiently.
• Works seamlessly across your setup
  Improved responsiveness ensures a consistent experience across different screen sizes and resolutions.
• Handle more complex cases with ease
  With support for Multi-Entity Alerts, you can investigate cases involving multiple entities in a single, unified view.

We’ve improved Fraud Signal Indicators to make them more precise and actionable. Indicators have always existed—now they’re generated and surfaced at the individual evaluation level, giving you clearer insight into why each score was produced.

What’s new

• See the top drivers behind each specific Fraud Signal evaluation on Alerts and Investigations
• View an entity’s current score and latest indicators directly in the entity panel
• Explore a full history of evaluations with a timeline view of how risk has changed over time
• Click into indicators for clear explanations of each signal

API update

• Indicators can now be retrieved in the Fraud Signal payload (available to all clients)
• Adds a small amount of latency (~100ms on average)

Why it matters
This update makes Fraud Signal more transparent at the moment decisions are made. Instead of a single, static set of reasons, you now see exactly what drove each evaluation—helping teams move faster, stay consistent, and trust the system without relying on rules.

We’ve introduced improvements to the Documents experience for those using Alloy's SDK to make reviewing and managing documents faster and more efficient.

if you’re using Alloy's SDK with one of the supported plugins below, you can now launch Entity Documents directly from the document evaluation page from a banner at the top of the page.

When you open the documents modal from an evaluation, any supplied data (if available) will now appear alongside the document, allowing you to easily compare document details with the supplied information.

Supported Plugins

This side-by-side comparison and banner is currently supported for the following plugins:

• Veriff
• Persona
• Sumsub
• Onfido
• Prove
• Socure DV
• Incode
• Vouched
• Alloy Multi-Document Collection

Additionally we have added the ability to upload documents directly within the document window—no extra steps required. Once uploaded, your documents will automatically appear at the front of the document navigation, making them easy to find and access right away. This is available for all Applications and Investigations users. 

AI Analysis from Alloy’s AI Assistant is now available in the Alloy platform.

What is the AI Assistant? Alloy’s AI Assistant is a native AI agent that brings agentic automation to tasks driven by a standard operating procedure (SOP), giving teams the ability to delegate any step in application and investigation review to AI. It understands Alloy’s decisioning logic, client-specific configurations, and historical data, enabling it to analyze outcomes, automate actions, and commit results back to the system of record.

AI Analysis is one of these actions. They give your teams instant, clear explanations of what happened, why, and what to do next within Alloy. Instead of hunting across rules, evaluations, data sources, and notes, reviewers can quickly understand the full context of a decision in one place. 

With AI Analysis, you can:

• Understand complex decisions instantly: Get a clear, structured overview of Evaluations, Applications, Alerts and Investigations, including key signals and contributing factors, in seconds.
• Take next steps accurately and with confidence: Understand what needs attention and where to focus review efforts, with analysis summaries grounded in Alloy’s decisioning metadata to reduce ambiguity and misinterpretation.
• Reduce manual review time: Reduce time spent gathering context so reviewers can focus on decision-making, escalation, and exceptions.
• Maintain audit-ready transparency: Every AI Analysis is generated from permissioned data with a full audit trail, supporting compliance and regulatory review.

Contact us at [email protected] or reach out to your Alloy representative to enable or learn more. 

Our new Policy Library feature lets you create, save, and reuse your own snippets, or modular blocks of workflow logic. A new tab offers an account-specific repository for custom decisioning snippets, making it easy to standardize logic, modularize workflows, and build faster as your policies evolve.

With this feature, you can:

• Create and reuse your own policy snippets: Save commonly used logic once and reuse it across workflows within your account (the ability to publish snippets is configurable by role permission).
• Standardize decisioning internally: Ensure teams are building from the same trusted logic, reducing inconsistencies and rework.
• Browse and search with ease: Quickly find snippets by name or use case directly from the Alloy workflow editor.
• Deploy with confidence: Similar to Alloy-created snippets, you can insert snippets into workflows with guided merge logic and preview before setting live.

With Alloy’s ongoing investment in self-service capabilities, the Policy Library offers a single place to access both Alloy-recommended logic and your organization’s defined policies. 

You can find more details in this help article. 

Actionable, agile and easy - over the last year we shipped a wave of new features to help teams stay ahead of fraud risk while growing their business. Learn more about everything we released last year in our latest blog post.