GnuCash vs Pigment

The Agents are the story: Modeler and Analyst now launch from the homepage and hold one conversation across multiple Applications, with the Analyst comparing Metrics across them and handling Board search. Days earlier, the Modeler gained the ability to build a Pigment model — data, Blocks, formulas, and Boards — from a dropped XLSX file, with multi-file uploads of PDFs, CSVs, and images. The rest of the run is planning-platform craft: metric diffing, live deployment progress, selective Metric data sync across Test & Deploy environments, and a steady stream of View Template and chart formatting controls.

Pigment is moving its Agents from assistants scoped to one Application into a workspace-level layer that can see and act across the whole model estate. Pairing that with Excel ingestion targets the actual starting condition of most planning customers, whose current model is a spreadsheet. Meanwhile the environment tooling — deployment visibility and configuration data kept in sync between test and production — reads as a platform maturing for customers who now treat their Pigment models as software.

With Agents workspace-wide and able to construct models from files, the next step is putting them inside the change-management path — proposing model edits that flow through Test & Deploy rather than landing directly. Expect the Analyst's cross-Application reach to extend into scheduled or triggered analysis rather than only chat-initiated questions.