The Easiest Work to Automate ≠ The Most Valuable Work to Actually Change

Kelly Coutinho

September 25, 2026 • 5 min read

For much of the last three decades, enterprise technology has been organized around functional domains, with planning, procurement, manufacturing, and logistics developing their own applications, workflows, and data. It delivered control, standardization, and scale, but left much of the work of connecting decisions across systems to people, spreadsheets, emails, and meetings.

Now, intelligence can increasingly sit inside applications, but it can also move between them: retrieving context from one system, reasoning across information from another, and triggering an action somewhere else. It makes connection more important than ever – and our recent vibe coding workshops with the Zero100 community make this opportunity obvious.

We worked with Citizen teams to define problem statements and rapidly prototype MVPs to share with their tech teams. Roughly half of the opportunities identified weren’t requests for new applications. They were attempts to reconnect workflows that had become disconnected across the existing technology estate.

Our analysis of Zero100 community data shows where manual intervention persists despite existing solutions, highlighting where disconnection is strongest. Below are those where the most manual hours are being tracked.

The takeaway brings us back to our first enterprise technology principle: enterprise technology is only valuable when it changes how work gets done. Increasingly, that means looking beyond the individual task to how intelligence connects decisions across the business.

The Application Is No Longer the Boundary of the Work

Much of enterprise AI today understandably starts at the level of the task (think: summarize the document or retrieve the information). Among our community who have mapped digital use cases and maturity across The Loop, 75% are focused here. While there is value here, a task is rarely where an operating outcome ends. Task-level automation can deliver real productivity and cost benefits, but the opportunity here is not limited to efficiency. When technology changes a decision chain, it can also affect availability, service, inventory, speed to market, and the ability to capture demand. The same technology can therefore operate on both sides of the P&L, changing the cost of doing the work and the value created by doing it differently.

This is what Zero100 PowerThreads are about.

Take forecasting as an example. Automating commentary around the forecast saves time. But the larger economic impact comes when that intelligence changes what the business does next: what it buys, where it positions inventory, how it allocates capacity, or when it replenishes. The opportunities become more consequential as intelligence moves from improving an individual activity to shaping a connected set of decisions.

This changes the technology problem. Improving a task may require a better model or AI embedded in an existing application. Changing a decision chain requires more: context that moves across systems, business rules that can be interpreted, actions that can be coordinated, and judgment applied along the way. Therefore, the easiest work to automate may not be the most valuable work to change. Some of the bigger opportunities lie in changing how one decision informs the next.

Where Is the Work Different?

There is another distinction becoming more important as AI capabilities become more widely available.

Enterprises contain a huge amount of work that does not need to be unique. Processing a standard transaction differently isn’t necessarily a source of competitive advantage, and not every bespoke process is evidence of work being different. Sometimes it simply reflects years of accumulated technology and organizational complexity. Standard work should increasingly benefit from standard capabilities. But there are also places where how an enterprise works matters enormously.

Not every change to work needs to create competitive advantage. Automating standard work can create significant value in its own right. But as AI makes it easier to change how work gets done, another distinction becomes increasingly important: where does doing the work differently actually create advantage?

David Friedberg, CEO of bioengineering company Ohalo, gave a useful example on Sa lesforce’s recent earnings call. Ohalo has chosen not to rebuild standard capabilities such as CRM, ERP, or communications but is building where differentiation sits: plant breeding, prospecting, and selecting the right product for a farmer using proprietary data and context. It is concentrating its own technology and operating logic where doing something differently creates value.

As technology participates in more of the work, knowing where doing something differently creates value is a prerequisite to actually realizing value beyond efficiency and productivity.

The Changing Relationship Between Horizontal and Vertical Technology

As intelligence increasingly moves across application and functional boundaries, the relationship between horizontal and vertical technology also changes.

Vertical applications continue to provide deep functional capability, systems of record, controls and specialized workflows. In many environments, that also includes governance, regulatory requirements, permissions, auditability, and domain-specific rules.

At the same time, increasingly powerful horizontal capabilities operate across those boundaries. Models, retrieval, reasoning, context, and agents can span applications, and we’re seeing organizations evaluate broader platforms for their ability to connect intelligence and work across the existing technology estate.

That raises the bar for both. Horizontal technology needs enough enterprise context and control to operate reliably in real workflows; vertical technology needs sufficient depth in the domain, its constraints, and the decisions that make it distinctive.

Between them sits something increasingly important: the enterprise’s own context, judgment, and operating logic, alongside the governance and permissions that determine how technology is applied to its work. The gap here isn’t missing technology, but how technology connects to the work.

What Should I Do Now?

To get more value from standard work, we suggest looking across your existing technology estate for capabilities that are already available but underused. Where greater standardization or adoption could improve productivity, execution, or consistency, focus on embedding those capabilities into the work before adding more technology.
To identify where changing the work could create greater value, start with the decisions and workflows that materially affect business outcomes, particularly those where how your organization operates can create competitive advantage. Then assess whether the current technology supports those decisions and prioritize changes accordingly.