Is Your ERP Ready for AI? The Criteria That Matter in 2026

The executive committee has just approved a budget for AI agents. Three floors down, the same company has been putting off cleaning its supplier data in the ERP for two years. No one has connected the two decisions yet.

McKinsey has just quantified this blind spot, and the name they’ve given it sums up 2026: the great divide between AI agents and the ERP.

“Only about 40% of companies report an impact from AI on their operating profit, and most quantify it at under 5%.” Source: McKinsey, Bridging the great AI agent and ERP divide, January 2026

Why isn’t your ERP ready for AI despite rising budgets?

Because money is migrating toward generative AI while investment in the technological foundation is shrinking: McKinsey finds that nearly half of IT organizations plan to invest in generative AI, with investment levels dropping significantly for core capabilities like infrastructure and architecture. The firm calls this imbalance the “great divide.”

The consequence also has a name: pilot purgatory. Use cases are multiplying without the end-to-end processes, data, and systems that would allow them to scale.

The finance function is a perfect example of this phenomenon. According to the CFO Connect State of AI in Finance 2026 report, 56% of finance teams are now using AI, but 45% are still in the pilot phase and only 17% have integrated it into their core processes.

What does an AI-ready ERP actually look like?

An AI-ready ERP is identified by four signs: reliable master data, documented end-to-end processes, up-to-date approval rules, and data accessible via APIs. Nothing spectacular, and that’s the point: AI relies on what the ERP already structures.

McKinsey points out that the ERP contains the company’s “operational DNA”: the process knowledge, data structures, and business logic accumulated over years. AI agents don’t replace this foundation; they extend it.

And they are coming, ready or not. Gartner predicts that 40% of enterprise applications will integrate specialized AI agents by the end of 2026, up from less than 5% in 2025. We have detailed what SAP, Microsoft, and Oracle are already delivering in their ERPs.

How do you measure the gap between your ERP and your AI ambitions?

By comparing your situation to published benchmarks rather than vendor demonstrations. The 2026 figures paint a market that is accelerating on the surface while stalling in the depths:

• about 40% of companies report an impact from AI on their operating profit, with most quantifying it at under 5% (McKinsey, January 2026);
• 40% of enterprise applications will integrate specialized AI agents by the end of 2026, up from less than 5% in 2025 (Gartner, August 2025);
• more than 40% of agentic AI projects will be abandoned by the end of 2027 due to a lack of clear value or sufficient risk controls (Gartner, June 2025);
• 43% of SAP customers have AI use cases in production, but 77% of those scenarios are running on non-SAP solutions (DSAG, February 2026);
• in Canada, 19.2% of businesses were using AI in the second quarter of 2026, with the rate reaching 40.4% in finance and insurance (Statistics Canada, June 2026);
• in Quebec, 12.7% of businesses were using AI in production in the second quarter of 2025, and the forecast for usage over the following 12 months (13.1%) is growing more slowly than in Ontario (16.5%), according to the Institut de la statistique du Québec (November 2025).

Methodological note: these figures come from public surveys and analyst firms, with different scopes from one study to another. None describe your specific situation.

The complete picture of Canadian figures can be found in our analysis of business AI adoption in Canada.

The bottleneck in 2026 isn't the power of the models. It’s the state of the system that serves as their memory.

Where do you start to prepare your ERP for AI without rebuilding everything?

One domain at a time: a bounded process, its validated master data, its documented governance rules, and then a measured pilot before any rollout. McKinsey observes that companies achieving 5% or more in financial impact from AI transform an entire domain rather than stacking isolated use cases.

In the ERP mandates PlanAxion carries out for B2B distributors and mid-sized organizations in Quebec, the same diagnosis keeps coming up: customer and supplier data hasn't been reviewed since implementation. Duplicates, outdated payment terms, and approval hierarchies that no longer reflect the organizational chart.

One process is particularly well-suited as a first domain: cash application. It’s bounded, measurable, and fueled by ERP history. This is the area targeted first by the cash application solutions PlanAxion implements, and the same reasoning guides choosing the right automation layer.

To turn this assessment into a roadmap, we use our AI workshop approach: 4 weeks, five steps (prepare, identify, prioritize, validate data, decide and deliver), one process at a time. The investment varies based on scope and is confirmed during a short exploratory call.

Should you pause your AI projects while you modernize your ERP?

No. Both initiatives should move forward together, provided you choose one domain and finish it. In fact, McKinsey notes that AI agents themselves are becoming a lever for modernizing ERPs, particularly for documenting processes and cleaning up data. The right approach isn't to choose between AI and the foundation: it’s to reject any agent project that doesn't clearly define the data it depends on.

Frequently asked questions about AI-ready ERPs

What is an AI-ready ERP?

An AI-ready ERP provides agents with reliable master data, documented processes, up-to-date approval rules, and data accessible via API. AI doesn't require a new ERP: it requires a foundation where the data and business logic are trustworthy.

Do you need to replace your ERP to adopt AI agents?

No. Major vendors are delivering their agents within existing suites, and a replacement is only justified by other criteria: obsolescence, maintenance costs, or structural functional limitations. Preparation involves master data, process documentation, and governance, not a complete reimplementation.

Why do so many AI agent projects fail?

Gartner predicts that more than 40% of agentic AI projects will be abandoned by the end of 2027 due to rising costs, unclear business value, or insufficient risk controls. An agent connected to outdated data accelerates errors instead of correcting them.

How long does it take to prepare an ERP for AI?

It depends on the scope. A bounded domain, such as supplier data or cash application, can be prepared in a few weeks to a few months. A 4-week prioritization workshop is often enough to identify the first domain, measure the data gap, and sequence the work.

Qualifying your first AI use case in your ERP

Before adding an AI agent or feature to your ERP, you must define the process, the accessible data, the controls, and human accountability. Our enterprise AI workshop helps prioritize use cases; our artificial intelligence consulting services then cover the pilot and its integration. Describe the process you want to improve and your current ERP to determine where to begin.