When AI Moves Analytics Into the Workflow

Hi there,

The next phase of SaaS won’t be defined by who has the most dashboards.

It will be defined by who can put intelligence closest to the moment of work. Users don’t want to leave the product, export data, wait on a report, or open another BI tool just to understand what’s happening. They want answers where the work is already happening.

That is why analytics is changing from a reporting layer into a product layer. Static reports, dashboards, self-service tools, and AI agents will all still matter. But the companies that win will know how to connect those layers into one faster, smarter, more useful experience.

In this issue, we’re looking at how analytics is evolving, why embedded intelligence is becoming harder to ignore, and what it takes for SaaS companies to become truly AI-native.

One of the most consistent patterns in technology is not replacement—it’s accumulation.

Over time, innovation tends to build in layers rather than wipe the slate clean. A simple example can be found in computing devices. We began with mainframes, moved to desktop PCs, then to laptops, smartphones, tablets, and now wearable devices like smartwatches. Each new form factor didn’t eliminate the previous one—it added a new layer of capability. Today, all of these coexist, each serving a distinct purpose depending on context, need, and convenience.

This same pattern is playing out in the world of analytics.

Analytics—especially reporting—has evolved through several clear stages:

Each of these stages represents a meaningful leap forward. But importantly, none of them have disappeared.

Rather than a linear progression where one paradigm replaces another, analytics is becoming a multi-layered ecosystem:

These layers coexist because they solve different problems. The value isn’t in choosing one—it’s in orchestrating all of them effectively

There’s an increasingly popular narrative that AI will fully replace human involvement in analytics—that reports will become entirely autonomous and primarily consumed by machines rather than people.

That view misses an important nuance.

AI will absolutely transform analytics. It will automate workflows, surface insights faster, and even initiate actions. But it will not eliminate the need for human judgment, context, and trust. Instead, it introduces a new layer—one that augments, rather than replaces, the existing stack.

The future of analytics isn’t about betting on a single paradigm. It’s about recognizing and embracing the layered nature of the ecosystem.

Organizations that win will not be those that abandon prior approaches in favor of the latest trend. They will be the ones that:

Technology evolves by addition, not subtraction. Analytics is no exception.

A recent analysis from AlixPartners (covered by Business Insider) makes one thing clear: AI isn’t disrupting SaaS evenly. It’s exposing where real value lives.

And for many products, that’s a problem.

Standalone analytics, dashboards, and surface-level insights are among the most vulnerable categories. Why? Because they live outside the workflow. And anything outside the workflow is easy to replicate… or replace.

Here’s the shift:

The real moat in SaaS isn’t features. It’s embedded intelligence built on proprietary data and workflows.

AI can generate dashboards. It can summarize trends. It can even build lightweight analytics layers on demand. But it struggles to replicate deeply embedded systems that are:

That’s where defensibility is moving.

For SaaS leaders, this changes the roadmap. Analytics can’t be a feature you add, it has to be infrastructure you build (or integrate) into the product itself.

Because if your users have to leave the workflow to get insights, someone else will bring those insights to them—faster, cheaper, and increasingly, automatically.

🔑 Key Takeaways

AI is quickly becoming the front door to software. Users aren’t just navigating dashboards anymore. They’re asking questions, getting instant answers, and triggering workflows through conversational interfaces.

That shift creates a real risk. If AI can access your APIs and data directly, your UI and customer relationship can get bypassed.

AI-native companies don’t fight that shift. They design for it.

The first step is understanding what you already have. Most SaaS platforms are sitting on the foundations AI needs: structured data, APIs, multi-tenant architecture, user permissions, and centralized workflows. The question isn’t whether AI can integrate with your product. It’s whether you control how that integration happens.

From there, the roadmap gets clearer.

Start by embedding AI into the product experience, not bolting it on as a chatbot. Then create governed access layers, like MCP, so external AI agents can interact with your platform on your terms, with your rules for security, permissions, and context.

Next, move beyond answers. The real value is not just helping users understand what happened. It is helping them decide what to do next and execute that action inside the workflow.

That is where monetization changes too. AI-native SaaS will not only charge for seats or access. It will focus on usage, execution, automation, and outcomes.

The winners won’t just expose data. They’ll become the intelligence layer powering insights, decisions, and actions wherever users choose to work.

AI will not replace SaaS.

AI-native SaaS will replace non-AI SaaS.

Key Takeaways

The bigger point is simple: AI does not make analytics less important. It makes the placement of analytics more important.

When intelligence lives outside the workflow, it becomes easier to ignore, copy, or replace. When it lives inside the product, tied to real user behavior and real decision points, it becomes part of how customers get work done.

That is where SaaS is heading. Not toward fewer analytics experiences, but toward better-layered ones. Reports, dashboards, self-service tools, and agents all have a role. The advantage goes to the products that know when each one should show up.

See you in the next edition,

Arman Eshraghi
The Embedded Intelligence Brief
Connect on LinkedIn