It's 2 AM. A monitoring dashboard lights up with alerts. Somewhere in a global bank's data center, a certificate has expired - again. The on-call engineer pulls up the incident, spends 20 minutes searching through past tickets to figure out what their colleague did last time, cobbles together a fix, and closes the ticket. By morning, nobody remembers it happened.
Three weeks later, the same thing happens on a different server. A different engineer. Another 20-minute scramble. Another closed ticket. Another forgotten resolution.
Now multiply that by hundreds of similar patterns across thousands of incidents every quarter. The knowledge to prevent these fire drills exists - scattered across closure notes, trapped in the heads of your most experienced engineers, invisible to the L1 team handling the 3 AM call. Every organization has this problem. And very few have a systematic way to solve it.
That's the story behind LEAP.
The LEAP Philosophy: Your Incidents Already Know the Answer
AIOps LEAP - now formally the Learning-Enhanced Automation Platform - was created from a deceptively simple insight: your historical incident data already contains the blueprint for your automation strategy. You don't need consultants mapping processes on whiteboards. You don't need engineers guessing which runbooks to write first. You need a system that can read the signals your operations are already generating and turn them into action.
Here's how that story unfolds, step by step.
Chapter 1: Clustering - Finding the Patterns Hiding in Plain Sight
Picture a room full of 10,000 closed incident tickets from the last six months. To a human analyst, it's an overwhelming wall of data. To LEAP's machine learning engine, it's a puzzle waiting to be solved.
LEAP applies ML-based clustering to group similar incidents into automation opportunity clusters. And for many of our customers, this single step is where the lightbulb moment happens. Suddenly, you're not staring at 10,000 individual problems. You're looking at 40 distinct patterns, and you can see that one cluster alone accounts for 340 incidents, each taking an average of 45 minutes to resolve. That's over 250 hours of engineering time spent doing the same thing, slightly differently, over and over again.
Customers tell us this visibility alone changes how they think about their operations. Before LEAP, the automation strategy was driven by gut feel and squeaky wheels. After clustering, it's driven by evidence.
Chapter 2: Prioritization - Not All Opportunities Are Created Equal
Knowing you have 40 automation opportunities is powerful. But which one do you tackle first? The one your VP keeps asking about or the one that's easiest to build?
LEAP removes the guesswork. Its AI-driven prioritization framework evaluates each cluster based on estimated savings, operational impact, incident frequency, and alignment with your business objectives. The high-frequency, high-impact patterns rise to the top. The tempting-but-low-value distractions fall to the bottom.
Think of it as the difference between a team that automates the easy stuff and feels productive, and a team that automates the right stuff and delivers measurable ROI. LEAP ensures you're the latter.
Chapter 3: Resolution Mining - The AI Detective
This is where the story gets interesting.
Imagine an analyst who could read every single resolution note from those 340 incidents in your top cluster, cross-reference them with your knowledge base articles, identify the common threads, filter out the noise, and produce a clean, step-by-step resolution procedure. Now imagine they could do that in minutes, not weeks.
That's LEAP's resolution mining. The platform's GenAI capabilities dive deep into six months of your team's collective troubleshooting wisdom, the workarounds that worked, the steps that didn't, the tribal knowledge that only existed in one engineer's head, and distils it into a standardized, repeatable runbook.
The 2 AM engineer from our opening story? They'd no longer need to search through past tickets. The answer would already be there - curated, validated, and ready to execute.
Chapter 4: Outcomes & Artifacts - Closing the Loop
Discovery without action is just a report. LEAP was designed to close the gap.
The final stage transforms mined resolutions into operational artifacts: playbooks that plug directly into your ServiceNow workflows, knowledge base articles that capture organizational know-how for your L1/L2 teams, and value dashboards that track real cost savings and MTTR reductions. It's the difference between saying "we should automate this" and actually having it running in production, measured, and improving.
This pipeline, "cluster, prioritize, mine, operationalize," is what makes LEAP fundamentally different. Instead of starting with a blank page and asking "what should we automate?", you start with evidence and ask "which of these proven opportunities do we tackle first?"
What's New in January 2026
If LEAP's story so far has been about turning incident chaos into automation clarity, the January 2026 release is about making that journey faster, smarter, and more precise at every step. We're not just showing you what to automate anymore - we're helping you understand how, generating the solutions, and integrating them seamlessly into your daily workflows.
And to reflect this evolution, AIOps LEAP has a new name: Learning-Enhanced Automation Platform (LEAP). Because we're no longer just about playbooks. We're about knowledge bases, flow actions, AI agents, and comprehensive automation orchestration across your AIOps and SecOps domains.
1. Sub-Grouping with GenAI: When One Cluster Isn't Specific Enough
Here's a scenario our customers know well. LEAP surfaces a cluster of 200+ incidents labelled "network connectivity issues." That's valuable, but "network connectivity" could mean VPN failures, DNS resolution problems, or load balancer timeouts. Each needs a different playbook.
LEAP now applies AI-powered hierarchical perspective clustering to break large groups (150+ records) into targeted sub-groups. That broad "network connectivity" cluster splits into three focused sub-groups, each specific enough to drive a precise automation, not a generic one. The customers who already loved clustering will love this even more: it's the same insight, sharpened to a finer point.
2. Multi-Source Resolution Mining: Beyond Your Four Walls
Until now, LEAP mined resolutions from your internal incident data. Powerful, but limited to what your team had already documented. Starting this release, resolution mining reaches outward - leveraging advanced web search and agentic artifact creation to pull best practices from authoritative external sources.
Your playbooks now benefit not just from what your team has done before, but from what the broader industry knows works. It's the difference between a runbook written from memory and one informed by collective expertise.
3. Knowledge Base Artifact Creation: From Tribal to Institutional Knowledge
Remember our 2 AM engineer? Even if they don't trigger an automated playbook, they should still be able to find the answer quickly. LEAP can now automatically generate formal Knowledge Base articles directly from automation outcomes - complete with embedded command snippets and executable code blocks.
The tribal knowledge that once lived only in your senior engineer's head? It's now a searchable, shareable organizational asset.
4. Persistent Re-Generation: Playbooks That Get Smarter Over Time
Automation isn't a one-and-done exercise. As new data sources come online - AI Search integrations, and additional KB connections - your resolutions should evolve too. The "Generate Resolution" action now persists throughout the group lifecycle, letting administrators regenerate recommendations as the platform's intelligence grows, with built-in safeguards to manage LLM costs.
5. LEAP Agent: Conversational Automation Architecture
A new agentic workflow brings guided, conversational AI directly into the automation architect's workspace. From any Automation Opportunity record page, the "Explore" button launches an interactive Now Assist Panel experience for creating and managing playbooks and knowledge base articles. It's automation architecture through dialogue, not forms.
6. Native SOW Integration: Meeting Teams Where They Work
LEAP-generated playbooks now appear directly in the Service Operations Workspace Playbooks module. No separate navigation, no context switching. Operations teams discover and launch automations exactly where they already spend their day. Seamless.
7. Transparency & Operational Visibility
Good automation requires trust, and trust requires transparency. This release adds several enhancements that bring clarity to what LEAP is doing and why:
i. The Automation Opportunities page now displays the date range of analysed data - so teams know exactly how current their recommendations are.
ii. Enhanced diagnostic messaging during plugin activation guides administrators through setup with clear, actionable error guidance.
iii. A new dedicated LEAP Properties UI makes critical configuration settings - the ones that drive savings calculations and reporting accuracy - easily accessible and manageable.
iv. Action Insights surfaces optimization opportunities and knowledge gaps, helping automation architects build smarter with every iteration.
The Bigger Picture
Let's go back to that 2 AM engineer one last time.
In the old world, they search through tickets, improvise a fix, close the incident, and move on. The knowledge stays locked in that single ticket.
In the LEAP world, that incident was already part of a cluster that surfaced months ago. The cluster was prioritized based on business impact. A resolution was mined from hundreds of similar past fixes and enriched with external best practices. A playbook was generated and published to SOW. A knowledge base article was created for the cases that need human judgment.
When the alert fires at 2 AM now, the engineer doesn't search. They follow a proven, AI-curated playbook, or better yet, the playbook runs itself.
That's the story of LEAP. And with the January 2026 release, we're writing the next chapter.
Ready to start your LEAP journey? Whether you're a current LEAP user or exploring automation opportunities for the first time, this release gives you the precision, intelligence, and integration to turn operational data into organizational capability.
We'd love to hear how you're using LEAP to transform your operations. What automation challenges are you tackling? What patterns are you seeing in your environment? Share your stories in the comments - we are listening, and your insights help shape where LEAP goes next.
