Workday has agreed to acquire Honeydew, a semantic-layer technology company that creates consistent business definitions for artificial intelligence agents and business intelligence tools.
Workday plans to integrate Honeydew’s semantic compiler with Workday Data Cloud so agents, analytics tools, and employees can use the same governed definitions when querying enterprise data.
The transaction is expected to close during Workday’s fourth fiscal quarter, subject to customary conditions.
Creating Consistent Answers Across Tools
Different business systems may calculate the same metric in different ways.
For example, one headcount calculation may include only full-time employees, while another includes contractors or approved but unfilled positions.
All of those answers can be technically valid under different definitions, but the inconsistency can create operational and financial problems.
Honeydew captures multiple business definitions and selects the relevant one based on the requester, use case, permissions, and context.
Its compiler converts those definitions into deterministic database queries before a request reaches an AI model.
This reduces the need to place extensive data dictionaries, SQL rules, and business logic into every prompt. It can also preserve definitions when a company changes analytics tools or underlying AI models.
Applying Semantics To Live Workday Data
Many semantic layers operate on information that has already been copied into a data warehouse or lake.
Workday plans to apply Honeydew definitions directly to live data in Workday Data Cloud. This could allow an agent to work with more current information while reducing the need for duplicate extracts.
Workday also intends to carry user and agent permissions into the query process. Unauthorized or invalid requests could be blocked before the database query runs.
The acquisition is intended to complement Workday’s existing agent-governance capabilities involving identity, access, behavior, and verification.
Honeydew’s semantic technology would add governance at the point where a natural-language request becomes a data query.
KEY QUOTES:
An AI model shouldn’t be improvising the calculation behind a payroll number. With Honeydew, the model says what it needs and Workday builds the exact query, the same way every time, within each person’s or agent’s permissions.
Gabe Monroy, Chief Technology Officer of Workday
Getting AI to write database queries was never the hard part. Getting it to understand the true definitions behind your company’s data is. We built Honeydew to solve that problem once, rather than in every tool and every prompt. With Workday, we plan to bring that shared understanding to every agent working with a company’s most important data.
David Krakov, CEO of Honeydew
The greatest bottleneck in enterprise AI today isn’t reasoning, it’s context. Foundation models know language, but they don’t naturally understand the physics of your business. By embedding Honeydew into Workday Data Cloud, we will provide a living semantic graph to give autonomous agents the exact, governed business context they need to safely execute at the core.
Yasmeen Ahmad, Senior Vice President and General Manager of Workday Data Cloud

