Know who’s doing what.

It overlaps with one, but that is not how it is built. HR software stores employees. Orglume models the operating structure behind them — roles, responsibilities, authority, skills, access and who covers whom — so it can answer what happens if somebody leaves rather than just telling you their start date. If what you need is payroll processing and a document store, there are cheaper products that do exactly that.

Roughly five to a hundred people. Below five you can genuinely hold it all in your head. Above a hundred you will start wanting things this deliberately does not do, such as workforce analytics and complex approval routing. The sweet spot is the stretch where the company has outgrown one person's memory but cannot justify an HR department.

An afternoon to something useful. You bring people in from a spreadsheet or by invitation, confirm the reporting lines and confirm what each team owns. Recording responsibilities properly takes longer — a few weeks of noticing things — because much of it has never been written down anywhere.

The connected model underneath everything: a person holds a role, a role carries responsibilities, a responsibility has an owner and a backup, a person has skills and access, and each of those relationships is a real record. It is what lets the product answer a question nobody anticipated, such as who has the skill, the authority and the availability to cover Friday.

A task finishes. A responsibility continues — the monthly report, supplier onboarding, the security review. Treating them as the same thing is why handovers miss the recurring work: it never appeared on a list because it was never a ticket. Responsibilities are objects with owners, backups, a rhythm and evidence.

No, and that is a structural limit rather than a setting. It will not say somebody is underperforming, should be promoted or is not suited to a role, and it produces no score intended to inform those decisions. It reports facts, labels its own interpretations as interpretations, and leaves employment judgements to the manager who is accountable for them.

No. Questions run with your identity through the same permission checks as every screen, so there is no separate AI permission model that could drift out of step. Compensation, health-related absence and private manager notes are excluded from assistant answers by default even for people who could open them directly.

Content is treated as data, never as instruction. Text inside an uploaded contract, a form submission or an imported record describes the business; it cannot change how the assistant behaves. This is a design property of how the content is handled rather than a filter that tries to spot bad input.