Lori Schafer in Forbes Tech Council: Agentic AI Only Works When Your Data Is Unlocked

Read the original article in Forbes Technology Council here.

To be candid, none of the promises of agentic AI are possible if a company’s data is locked away inside systems that do not connect or communicate. Agent-to-agent communication can only work at a high level when data is easy to find, easy to use and ready for action.

This is a core principle as enterprise companies explore and look to scale agentic AI within their operations. Gartner predicts that "by 2028, organizations that leverage multiagent AI for 80% of customer-facing business processes will dominate." Their research also predicts, "By 2028, 90% of B2B buying will be AI agent intermediated, pushing over $15 trillion of B2B spend through AI agent exchanges."

The message is clear. As the technology becomes more capable, it will evolve into a responsible, governed part of the organization. Agentic AI is not simply an assistant. It powers business decisions, completes tasks, coordinates across systems and helps teams move faster to make better choices. All of this depends on whether the data is liberated, unlocked and ready to go.

Data fuels agentic AI's role in strengthening operations.

A recent report from Salesforce finds that 63% of business leaders describe their organizations as data-driven, yet 7 in 10 say their most valuable insights are trapped inside unstructured or inaccessible data. Companies can benefit now if they prepare their data and make it usable for people and for AI.

Data fuels innovation. It strengthens workflows. It supports better decisions. To achieve all of this, companies need to remove barriers around their data and govern it carefully.

These four steps can help.

1. Help systems and agents speak the same language.

Most large companies use many different platforms for merchandising, e-commerce, marketing, logistics or operations. These systems rarely connect in a smooth way. Each may use different formats, rules or terminology.

Agentic AI can help bridge these gaps by creating a shared layer of understanding. With this in place, information can flow more clearly across departments and systems. The goal is for every team to work from a common set of definitions and business rules. This leads to consistent decisions, faster problem-solving and clearer communication.

2. Make data ready for both people and AI.

Good decisions require good data. It must be complete, accurate and easy to access. It must also be enriched with the right context, such as language, product details and business rules.

In retail, teams use data every day. Marketers, merchandisers, planners and supply chain teams all need information to solve problems. When people interact with AI and validate or improve the data as they work, they make the entire system smarter. Each question asked and each task completed strengthens the intelligence of the organization.

To support this, companies should treat their data platform as a center for innovation. It cannot be a closed or isolated system. The more tools and users that connect, the more value the organization can create.

3. Build strong governance to create trust.

As AI takes on more responsibility, oversight becomes more important. Governance sets the rules for how data enters the system, how it is used and how it is protected.

Every AI agent should follow the same permissions and restrictions that human users follow. Every action taken by an AI agent should be trackable. No AI agent should be able to expand its access, change permissions or make sensitive decisions without approval.

Trust allows teams to adopt new tools with confidence. Strong governance is the foundation of that trust.

4. Create agents that understand the business.

The most useful AI agents are not the ones that simply follow commands. They are the ones trained to understand how the business works. These agents can help teams:

• Build and adjust business plans.

• Run pricing, assortment, scheduling or demand simulations.

• Recommend actions based on data from across the business.

• Support leaders with scenario planning.

In retail, this may involve planning promotions or forecasting demand across stores and channels. When agents are trained with real business knowledge, they become trusted partners who support daily work. They help people think more clearly, move more quickly and make stronger decisions.

Agentic AI helps remove barriers in daily work.

The greatest value of agentic AI today is its ability to reduce friction. In retail and consumer brands, teams can instantly verify product data across channels, confirm content accuracy, audit supplier information or check inventory details without manual research. IT teams can focus on innovation instead of routine troubleshooting. Operations teams gain stronger accuracy, faster response times and clearer insights into what is happening across the business.

To achieve this level of performance, companies must unlock their data, connect their systems and govern everything with precision and care. They also need to help employees ask questions, test ideas and let AI handle more of the routine work.

Agentic AI is a new operational layer that supports modern business. Companies that prepare their data now can achieve stronger growth, higher efficiency and innovate with confidence.