ParleHub – Enterprise AI Collaboration and Cost Control – Launch Today on PH

ParleHub launched today on Product Hunt, offering enterprises a platform that ties AI usage to client projects with shared threads, project-specific budgets, and expense codes, while integrating with Microsoft Entra ID and SharePoint. The tool provides timesheets for AI agents, pre-flight budget enforcement with 80% warnings, and audit logs for regulated practices.

Every hour has a client code . Your AI spend has none. ParleHub gives every project a shared thread, its own budget, and your own expense code. AI cost lands on the engagement that caused it, and exports straight into the system you already bill from. Your team is already using AI on client work On personal accounts, with client documents, under terms nobody reviewed. You cannot say which clients, which files, or whose agreement covered it. That is also the honest answer to the AI section of every client security questionnaire you sign. - check circle Access tied to Microsoft Entra ID, so offboarding actually removes it - check circle Project files stay in your own SharePoint site, under per-site permissions, never leaving your tenant - check circle Model calls run on your provider keys and your agreement, or on models we host under ours, never a personal consumer account - check circle Four levels of role, and projects only their members can see Timesheets for your AI agents Your people book their hours to a client. Your AI books nothing. It arrives as one provider invoice with no client on it and lands in overhead, where the practices that barely touch AI quietly subsidise the ones that live on it. - check circle A budget scoped to the project, not the person, because your people work across six of them - check circle Your own project code on every project, carried through to the export - check circle Spend and token usage broken down by project, person, provider, or model, over any range - check circle One canonical token ledger behind both the report and the enforcement, so there is a single number Every project carries your own expense code, straight through to the export. AI cost lands where every other cost already lands, so you can recharge it, price it into a fixed fee, or cut the engagement type that does not pay. Find out before the invoice, not after Metered spend with no ceiling is the opposite of how a firm plans. One enthusiastic user, one runaway loop, and the number turns up at month end with an explanation attached to it. - check circle Pre-flight enforcement blocks the send before it exceeds the project budget - check circle An 80% warning state, so the team sees it coming - check circle Personal and org-wide caps stacked as backstops above every project - check circle Blocked? Request an increase in one click, routed straight to an admin The ceiling becomes a number you set rather than a number you discover, and it does not stop the work, because there is a path through it. The best work your firm did last week is in a closed tab Private chat histories mean context rebuilt from scratch at every client switch, by every person. Two people on the same engagement ask the same question a day apart, against different documents, and get different answers. When someone rolls off, what they worked out leaves with them. - check circle Shared project threads every member can see, live - check circle Presence indicators for who is viewing, typing, or waiting on a reply - check circle Keyword and semantic search across every conversation in the project - check circle Promote a personal chat into a shared project when it turns out to matter - check circle Branch by editing and resending, without losing the original When a partner asks where the number came from Work nobody can review gets re-done by whoever has to sign it, at their rate. In regulated practice it cannot be used at all, so the engagement pays for the analysis twice: once quickly, once defensibly. - check circle The whole exchange stays in the thread: the question, the reply, and which model produced it - check circle Real tool calls against the project's own files, so the work is grounded in your documents - check circle Exact find-and-replace and diffs, so a reviewer sees what changed, not a silent rewrite - check circle Sensitive actions in an audit log enforced append-only at the core platform level You still check the work. The difference is that there is something to check, instead of a paragraph that appeared from nowhere. Built for the person who fills in the security questionnaire Enterprise SSO, files that never leave your own tenant, and an audit trail that cannot be quietly edited. Enterprise SSO Microsoft Entra ID & Google Workspace Native storage SharePoint & Google Drive, in your tenant Tamper-evident audit log Append-only, enforced at the core platform level Role-based access Org admin, project admin, and member roles Fixed headcount, variable demand Three months a year, or every time a bid lands, you cannot take more work at any price. Leverage is the only capacity lever that does not require hiring. Works the actual files Search and read project documents, query spreadsheets, and convert PDF, Word, and PowerPoint to searchable markdown, as real tool calls. No code execution anywhere near your client data. Ramp in a day Someone joining an engagement reads the thread the team already built instead of rebuilding it. The context belongs to the project, not to whoever happened to open the chat. Every major model Anthropic, OpenAI, Google Gemini, and Azure OpenAI on your own keys. Curate which models a project may use, and switch mid-chat without losing the thread. Free to start. Prove it on one engagement. Free $0 3 seats, 3 projects, $10/mo of model spend. Team Flat monthly 10 seats, 10 projects, SSO, uncapped model spend. Enterprise Per seat Unlimited seats & projects, private projects, SharePoint/Drive storage. Put a client code on every dollar of AI your firm spends. Start free at app.parlehub.com arrow forward https://app.parlehub.com