The $400M Domo Deal Forces a Choice

Hi there,

The analytics market is being pushed toward a clearer choice.

Progress Software’s agreement to acquire substantially all of Domo’s AI and data platform business is more than another transaction. It signals that established analytics companies must decide whether to make drastic changes and become AI-ready or maximize cash flow through strong renewals, disciplined operations, and improved margins. Many may choose the latter, accelerating consolidation across the market.

At the same time, AI coding agents are making it easier to generate interfaces, select frameworks, and connect systems quickly. But speed does not remove the need for architecture. Multi-tenancy, governance, permissions, performance, and secure data access still determine whether an embedded analytics product can work in the real world.

In this edition, I look at the strategic fork facing the industry, how embedded analytics is evolving into embedded intelligence, and why the platform beneath the AI matters more than the demo.

Let’s get into it.

Progress Software will acquire substantially all of Domo’s AI and data platform business for $400 million in cash. The deal is structured as an asset purchase, not a merger.

Sometimes, I tend to oversimplify the situation, and the way I have framed the strategic fork facing analytics companies might be one of those cases. I see two viable paths today:

This is the higher-risk, higher-upside strategy.

It isn’t about adding a chatbot to an existing BI product. It means rethinking the whole organization, including the people, the processes, and the products.

My honest advice to most analytics companies would be to avoid that path unless they can innovate fast, learn fast, and act fast. This is a true racing path, and speed is the key to success.

This is the Progress Software model.

It counts on a loyal installed base and high renewal rates, combined with flawless operations management, to lower expenses and improve margins.

This can create excellent shareholder returns even without much revenue growth.

Embedded Analytics is no exception, and the players will face the same choices. In my view, Embedded Analytics will evolve into a broader domain, which I would call Embedded Intelligence.

One common mistake is to mix the three concepts of AI, software, and data. Embedded Intelligence will be the convergence of these distinct yet related technologies. It will be multimodal. It will be more autonomous. And it will function more as an orchestration layer than as a larger software platform with more capabilities.

There is a lot of confidence right now that AI can help teams build almost anything.

In many cases, that confidence is justified.

But I would be careful applying that same thinking to embedded analytics.

The easy part is generating a dashboard.

The hard part is everything around it.

Can different tenants securely access only their own data?

Can power users build and modify dashboards without engineering tickets?

Can everyday users personalize the experience without breaking governance?

Can the platform support pipelines, automation, permissions, versioning, and scale as customer needs evolve?

That is where “just use a chart library” starts to fall apart.

The visual layer is only the visible part of analytics.

The real work sits underneath: data movement, governance, tenancy, security, extensibility, and operational control.

AI can accelerate a lot.

But it does not remove the need for a real platform foundation.

For SaaS companies whose core product is not analytics, the better strategy is not to code everything from scratch.

It is to start with a complete embedded analytics platform that covers the 90% you should not have to rebuild and then use AI to customize the 10% that makes your product unique.

AI can make embedded analytics look simple.

A user types a question, a chart appears, and the product suddenly feels intelligent. That is useful, but it does not tell you whether the platform is ready for real customers.

The harder questions sit beneath the demo.

Can every tenant securely access only its own data? Can the AI work from consistent business definitions? Can the platform support different permissions, workflows, and customer requirements without creating more engineering work?

That foundation becomes even more important as analytics moves beyond answering questions.

The next generation of AI will monitor data, identify unusual activity, recommend what should happen next, and connect those insights to real workflows. But none of that works reliably without strong governance, secure multi-tenancy, flexible APIs, and an architecture that can adapt as AI changes.

A polished feature may win attention today.

The stronger investment is a platform that can keep evolving tomorrow.

Key Takeaways

The analytics companies that succeed will not simply be the ones that add the most AI features. AI alone will not replace analytics or create a lasting competitive advantage.

They will be the companies that choose a clear strategic path and align the organization around it.

Becoming AI-ready requires speed, experimentation, and a willingness to rethink the people, processes, and products behind the business. Maximizing cash flow requires strong renewals, loyal customers, disciplined operations, and improved margins. Both paths can be viable. Most established players may choose the latter, which would lead to further consolidation across the analytics industry.

For SaaS teams, the same principle applies at the product level. AI can accelerate development and improve the user experience, but it cannot replace the architecture, governance, and security needed to make embedded intelligence reliable.

The demo may create interest.

The foundation determines whether the product lasts.

See you in the next edition,

Arman Eshraghi
The Embedded Intelligence Brief
Connect on LinkedIn