Next Generation Learning: Driving User Engagement

Okta helps organizations securely connect people to the technologies they use every day. As the leading independent identity platform, Okta serves customers of every size and industry — from digital natives to global enterprises — who rely on it to protect access, streamline workflows, and keep users productive.

To match the pace of product innovation, Okta’s Customer Acceleration leaders, Sharon Castillo and Dan Winted, have a clear mandate: help customers build real, working skills with Okta, not just awareness of features. That means moving beyond traditional product training toward experiences that feel embedded in the product, grounded in real scenarios, and ready for a world shaped by AI. Parta.io is one of the platforms helping Okta make that shift.

Challenge

Traditional product training serves a purpose, but it often falls short of helping customers develop the real-world skills they need. Over time, Okta’s team saw familiar constraints:

• Most content was passive — recorded webinars, linear videos, and step-by-step guides.

• Deep-dive “labs” were powerful, but heavy to build and maintain, especially as the product evolved.

• Learning often happened in a destination (an academy or LMS), separate from where users actually did their work.

And underneath it all, authoring tools made it hard to design beyond the basics. The result: generic training that didn’t always translate into confident, adaptable admins and practitioners.

Strategy

Okta set out to evolve this approach by augmenting traditional methods with a next-generation design focused on:

• Applied learning instead of one-way instruction

• Interactive experiences that make users think, decide, and see consequences

• In-workflow delivery, meeting users in the product and documentation they already use.

Parta.io became a core part of this strategy — a toolset that lets the team design flexible, interactive “widgets” and simulations that can be used across channels, not just in a single course.

Rewriting the playbook

From passive to interactive content

Traditional approach

The classic playbook prioritized awareness and exposure:

• Linear video walkthroughs of the UI

• Step-by-step tutorials showing which button to click next

• Long courses that quickly went out of date when screens changed.

These assets were useful for initial orientation, but they didn’t build adaptable skills. Learners might remember which menu to open, but not why they were doing it — or how to adjust when their context changed.

Next-gen approach

With Parta.io, Okta is moving from “watch me” to “try it and understand why.” Interactive experiences are designed around “aha” moments, where users make a decision, receive targeted feedback, and understand the reasoning behind a recommended action.

For example:

• Instead of simply showing how to configure a setting, an interaction presents a realistic scenario, asks the learner to choose a configuration, and then explains the impact of that choice.

Users are prompted to think about concepts and trade-offs, not just UI clicks. Live feedback clarifies the relationship between their decision and the outcome for security, user experience, or operational efficiency.

The goal is to make content that survives UI changes because it’s anchored in mental models, not menu paths.

Meeting users where they are

Traditional approach

Learning content has historically been centralized in an academy or LMS — a separate destination that requires users to stop what they’re doing and go elsewhere to learn. Brand and content standards were often enforced through static documentation and checklists, making it hard to keep experiences consistent across Product documentation, Help center articles, Community spaces, In-product guidance.

Next-gen approach

Okta is reimagining learning as in-workflow, multi-surface, and reusable. Using Parta.io, the team designs single-source interactive components — what they call widgets — that can be deployed where it makes the most sense (e.g., Okta Docs). A learner might encounter the same decision-support interaction while reading docs, taking a course, or troubleshooting an issue — creating continuity across channels.

Simulations over labs: a hands-on approach

Traditional approach

Complex virtual labs have long been a cornerstone of hands-on product training. They allow learners to click around in a realistic environment, but they come with trade-offs:

• Heavy to set up and maintain

• Expensive to run at scale

• Brittle when the product is changing quickly

• Sometimes more focused on “can you follow the recipe?” than on understanding when and why to use a feature.

Next-gen approach

Okta is augmenting this strategy with lightweight, high-impact simulations built in Parta.io.

These simulations are:

• Scenario-driven – learners are dropped into a realistic context (for example, defining a custom admin role, or choosing how to respond to a suspicious sign-in pattern).

• Decision-focused – instead of recreating every click, the experience zooms in on the critical moments where an admin must decide between options.

• Fast to build and update – no full lab environment required, so simulations can evolve with the product.

A typical interaction might invite users to “Build your custom admin role” by combining resource sets and permission levels, then immediately show the implications of that configuration.

The result: hands-on learning that delivers the essence of real practice, without the overhead of traditional labs.

What’s next?

Okta and Parta.io see this work as the beginning of a longer journey. Several big questions are shaping what comes next.

AI Search

As more users turn to AI tools to find answers — whether through general-purpose models or AI-powered search in a help center — the question becomes: How do we design learning so it’s fully discoverable and understandable by AI, not just by humans?

This means:

• Structuring content so AI can reliably surface the right interaction or explanation

• Capturing intent and context in metadata, not just prose

• Thinking about Answer Engine Optimization (AEO): optimizing for AI-mediated discovery, not just traditional SEO.

The ambition is that, when a user or an AI agent asks a question, it can point to the best interactive learning object, not just a paragraph of text.

Single-sourcing Content

Single-sourcing is already part of Okta’s practice, but the next stage goes further:

• Designing canonical, reusable building blocks (widgets, simulations, explanations) that can be pulled into Academy, Docs, Help, Community, and in-app experiences.

• Ensuring that AI search is fed from the same source of truth, so answers are consistent regardless of where the question is asked.

This reduces duplication, makes updates faster, and keeps the customer’s journey coherent — even as they move between channels and tools.

Consumerization → Personalization expectation

Consumer platforms like Netflix and Spotify have reset expectations. Users now assume that:

• Experiences will be personalized, not generic

• Systems will remember their history, goals, and preferences

• Recommendations will feel relevant and adaptive.

Okta’s question is: How can we bring that same sense of personalization into the customer’s learning journey with our product?

In practice, that could mean:

• Tailoring learning paths to a user’s role, industry, and current configuration

• Adjusting difficulty or depth based on prior behavior

• Surfacing the right widget or simulation at the exact moment a user needs it — whether in a doc, in-product, or through an AI assistant.

Even more modular content

To enable AI search, single-sourcing, and personalization, content has to become even more modular. Think of it as a library of Lego bricks rather than a stack of monolithic courses:

• Small, well-defined interactions that each teach a specific concept or decision

• Clear inputs and outputs so they can plug into different flows

• Composable sequences that can be rearranged for different audiences and contexts.

Parta.io gives Okta a canvas for designing these modules — self-contained, branded, interactive, and ready to be reused wherever they make sense.

Data analytics

At the center of all this is data. Beyond simple completion rates, Okta wants to understand:

• Which decisions learners make inside simulations

• Where they hesitate, backtrack, or consistently choose sub-optimal options

• How exposure to a widget or interaction correlates with product usage, configuration quality, and support outcomes.

By connecting learning analytics with product and support data, the team can:

• Refine interaction design based on real behavior

• Prioritize new content where it will have the biggest impact

• Demonstrate the value of customer education in terms the business recognizes — reduced risk, smoother deployments, faster time to value.

Okta and Parta.io share a simple belief: in a world of constant change, good design is still good design. The tools are evolving. AI is reshaping how users search, learn, and make decisions. Expectations for personalization and in-workflow support are higher than ever. But the core remains the same: clarity about what users need to do, thoughtfully designed experiences that help them practice, and feedback that builds real confidence.

By moving from passive courses to interactive, reusable simulations and widgets — delivered wherever users are — Okta is rewriting its customer education playbook. Parta.io is the platform helping to make that vision practical, scalable, and adaptable.

And the story isn’t finished. As AI search matures, single-sourced content expands, and data connects more dots between learning and outcomes, Okta’s customer education will continue to look less like a destination and more like what it has always aimed to be: an integrated, intelligent layer of support woven through every step of the customer journey.

To dive deeper, check out the full recording of Okta’s Webinar recording.