Cart Before the (Digital) Horse: Why Some Businesses are Unprepared for the Age of Agents

For insiders, Agentic AI isn’t a surprise. Rooted in the work of Alan Turing and his team at Bletchley Park during WWII, digital agents are a natural (if impressive) evolution of computing. And, like the IBMs of yore, Agentic AI will be a core driver of value for business.

Agentic AI by the numbers, according to research conducted by IDC:

• “By 2025, 50% of organizations will use enterprise agents configured for specific business functions, instead of focusing on individual copilot technologies to achieve faster business value from AI.”  

• “By late 2026, 65% of organizations will leverage AI to bring immediate employee and business value with AI-driven technology assistants, advisors, and agents enabling improved decisions.”

• “By 2027, 80% of critical AI decisions will require human oversight supported by visual explainability dashboards, potentially slowing processes but enhancing accountability.”

In spite of the promise of Agentic systems, many organizations are unprepared. Why? Well, companies lack the ecosystem needed to support AI-powered applications. The pre-cloud systems that many still rely on aren’t suitable for the new tech, the proverbial train ready before the track. Companies want the value Agentic AI will bring, but many are at risk of flying off the rails in their rush to adopt.

Given that infrastructure preparation is needed, this is just the moment when we should be investing in IT. However, with economic headwinds, companies are seeking to reduce operating costs, not expand them.  

“The top driver for IT investment is reducing IT costs,” according to IDC.

While reducing costs in times of scarcity is a must for the bottom line, wanton cost-cutting is not without its repercussions. Companies need to invest tactically and prudently to prepare for agentic workflows, remaining conscious of their balance sheet.

In this blog, we’ll offer some background on agentic technology and trends in its deployment in general. Then, we’ll talk about what’s most important in preparing for agentic AI, even with the constraints of a tight budget.

All About Agentic

First, let’s talk broadly about the basics of Agentic technology in the modern economy: What is Agentic AI, when is all this happening, and why is it important for your business moving forward?

What is Agentic AI? Well, digital agents represent a set of technical advancements in Artificial Intelligence. While the exact definition is evolving, “agentic” generally connotes a couple things. “Agentic” systems have a combination of natural language processing capabilities and machine learning. These predictive models are what allow AI to “understand,” generating intelligible, coherent responses to human inputs.  

In terms of results, what sets it apart is autonomous analysis and task execution. Basically, digital agents think and act on your behalf, without the need for active maintenance and interference.

When is agentic AI coming? Experts’ best guesses are that we’ll see digital agents blossom in the roughly next 2-4 years. More specifically, in 24-36 months, “Assistants and advisors [will gain] high autonomy in specific functional areas. Agentic AI [will achieve] near-complete autonomy for most business-critical processes,” according to IDC.

Why is agentic AI being adopted? The promises are big: Unimaginable efficiency. Context-aware, automated decision-making. Huge cost savings. Mitigated risk.  

Agentic AI, if it pans out as imagined, will mean true automation. Perhaps for the first time, we’ll have systems that bring order in task execution, without adding complexity in upkeep. And, while we don’t purport to own a crystal (or: silicon) ball, there’s good reason to believe these promises will come to pass.

What does Agentic have to do with ERP?

Agentic AI is a burgeoning topic in ERP. With agentic-powered technology, third-party platforms can streamline clunky ERP systems like never before. It’s a great application of the areas where agentic is strongest. Instead of tedious manual review, agentic-powered tech learns as it goes, exponentially improving system function without the inefficiency and expense of manual changes. You get the intelligence of a human, with the speed and scale of AI.

All by themselves, improved workflows and reduced IT costs are great news. After all, ERPs represent the biggest IT infrastructure investments a company makes. Managing them can be monumentally expensive with the wrong (or, as is often the case, with no) technology partners by your side.  

Modernizing, streamlining, and optimizing ERP management does more than cut IT costs, though. Getting your ERP in order with digital agents, gets you ready for agentic deployment, well beyond your IT stack.

How does agentic AI for ERP empower you moving forward? Well, ERPs might be best known for processing payroll, tracking inventory, and issuing financial statements, but they do more than that. They facilitate dataflows, prepare for reporting, and record workflow steps. These functions are all crucial pieces of data for agentic transformation. Getting your ERP updated and optimized, and getting visibility into all the corners of your IT infrastructure, is key to introducing digital agents across departments.

With agentic optimization in ERP, you’ll lay the groundwork for digital agents throughout your organization.

How can I set up my critical enterprise apps for Agentic workflows?

Moving to agentic, while well worth it, isn’t without effort. It’ll take significant investments in your IT stack. You’ll need to consider:

Cloud migration: The essential first step for agentic implementation is cloud migration.  Strategic investments in cloud migration, like moving from Oracle EBS to Fusion, greatly expand the possibilities for AI integration. This is because cloud systems offer resilience, reduction in costs, and the performance and capacity that agentic computation demands.  

Observability: The second factor central to incorporating agentic AI is visibility into system functioning. Without the ability to see what’s happening in your system, you won’t be able to automate workflows. According to IDC, with the rise of Agentic AI, “KPIs will shift, and with [them] traditional measurements.” Furthermore, in our view, AI integration will redefine traditional metrics, focusing more on numerical markers of efficiency and productivity. In order to demonstrate numerical progress toward reaching business goals to stakeholders, you’ll need deep visibility into your system. Observability is key.

Continuous Optimization: Last but not least, you need to achieve ongoing, continuous optimization of your ERP. Your ERP orchestrates or enables a broad swath of workflows, and it’s important that it runs like a well-oiled machine. Agentic is powerful, but like an army marching to battle, it takes significant resources to clothe and feed.  

Get ready for Agentic, with an Agentic-enabled platform  

As a front-runner of Agentic in ERP, Opkey is a champion of AI transformation in SaaS. Digital agents are changing business, and we can help you check agentic readiness off your list, while meeting the budgetary demands of today.

Our ERP Lifecycle Optimization Platform is the unified system to expand the impact of your IT team. Powered by a proprietary, ERP-specific Small Language Model, our platform allows you to manage configuration, operation, and ongoing optimization of your critical enterprise apps from one, no-code dashboard. You can deploy digital agents to migrate to the cloud, customize your apps, test your system, and train your people. All this at the touch of a button.

With help on cloud migration, robust observability, and agentic optimization, you’ll be prepared for agentic implementation throughout your business.  

In short: thanks to Opkey’s agents, you can finally afford the help you need to adopt new tech.