Teamhood AI

Teamhood AI gives you a way to ask questions about your workspace, generate reports, and create project structures directly from natural-language instructions. You can use Teamhood AI to analyze work-in-progress, build new project boards, generate tasks from documents, and model processes as Kanban workflows.

How Teamhood AI Works

⚠️ We do not share your data with any AI LLM’s, everything happens inside Teamhood.

Teamhood AI processes workspace data and user inputs inside the Teamhood platform. You can query tasks, boards, time tracking entries, or custom fields without building filters manually. You can also create project structures by uploading documents or pasting notes. Teamhood AI returns structured output in the form of tables, task lists, or full project boards.

Teamhood AI supports queries and creation actions on both single and multiple workspaces level. When you work with several workspaces, you can specify the source by including the workspace reference in the question (@workspace). You can direct Teamhood AI to work in a specific board (@board) when preparing new tasks or workflows.

Asking Questions About Your Data 

You can ask about overdue tasks, progress trends, user workload, or time tracking performance. The output appears as a structured table that you can sort, filter, or export. 

Example questions include: 

• Show overdue tasks grouped by board and assignee. 
• List tasks with significant difference between estimate and tracked time. 
• Display all tasks completed last week.
• Summarize workload for all team members in the Workspace Alpha board. 

When you reference a custom field, use the exact field name. Teamhood AI reads custom fields as they appear in your workspace. When your workspace contains several boards with similar names, specify the exact location to ensure accurate results. 

Finding Things and Asking About Teamhood

Teamhood AI is not only a reporting tool. It also works as a search and help layer, so you can stay in one place instead of switching between the app and documentation.

• Find items across your workspaces and open them straight from the results — useful when you remember a task but not where it lives.
• Search this knowledge base. Ask how a feature works and Teamhood AI answers from our product documentation, including workflow tips and best practices.
• Get advice on workflows and general project-management questions, such as how to structure a board for a particular kind of work.

In practice this means you can ask Teamhood AI to find a task, explain a workflow, or answer a product question without leaving Teamhood.

Turning AI Features Off

Some organisations need to control which capabilities are available to their users. Account Owners can enable or disable specific features account-wide from Policy settings, including AI features and attachments.

If Teamhood AI is switched off for your account, the assistant will not be available to any user in it. See roles and access rights for who can change account-level policy.

Demo video

Creating Tasks and Project Structures 

Creator tool lets you generate project content from documents, pasted notes, or written descriptions. You can create new boards, prepare tasks and sub-tasks, define dependencies, and map phases of a process into workflow steps. 

Creating a Project Board From a Document 

You can upload a project charter or requirements PDF and ask Teamhood AI to generate a project board. Teamhood AI extracts actionable work items, organizes them into groups, and prepares dependencies based on sequence or prerequisite relationships. 

Example workflow: 

• Open Teamhood AI. 
• Upload the PDF containing project scope or planning information. 
• Ask to create a project board from the document. 
• Review the generated groups, tasks, and dependencies. 

The resulting board contains tasks grouped by topic or phase. When the document describes timelines, Teamhood AI assigns estimates where possible. 

Creating Tasks From Notes 

You can paste unstructured notes from a workshop or meeting. Teamhood AI identifies discrete work items and places them in your selected board. 

Example workflow: 

• Copy the notes from your workshop document.
• Paste the text into Teamhood AI.
• Ask to convert the notes into tasks and place them in a specific board.
• Review task names, descriptions, and proposed sub-tasks. 

Teamhood AI prepares tasks based on headings, bullet points, or free-form sentences. When notes contain implicit sequencing, Teamhood AI creates dependencies between tasks. 

Modelling a Process as a Kanban Board 

You can describe a workflow, and Teamhood AI generates a board structure with relevant statuses, swimlanes, and optional grouping. 

Example workflow: 

• Describe your process, such as approval flow, procurement steps, or development pipeline.
• Ask Teamhood AI to model the process as a Kanban board.
• Select the workspace where the board should be created.
• Review generated statuses and swimlanes. 

Teamhood AI interprets process phases as statuses. When the process separates responsibilities, Teamhood AI creates swimlanes that reflect the roles or work categories. 

Working With Generated Data 

When Teamhood AI returns requested data, you can work with it further by: 

• Exporting it as .csv to 3rd party systems (click save icon in the toolbar) 
• Creating charts to visualize numbers (select data and then use toolbar to insert a new chart)
• Add additional data such as formulas (click any cell and type in ‘=’, just like you would do in excel)
• Save it as a company report and share with others (click Save as report) 

For complex queries, you can refine the question without clearing the previous output. This approach helps you adjust the scope of a report or extend the generated view. 

Improving Accuracy of AI Queries 

You can improve accuracy by providing clear task names, exact board titles, and specific workspace context: 

• Use the exact custom field title when requesting values. Spelling and spacing must match the Teamhood field name. 
• To specify where the data should be taken from, use the @ symbol to select workspaces or boards. 
• Shape your prompts using the structure “what to take, from where to take, and how to display.” 
• Avoid broad requests such as “analyze my project data,” because they do not define a clear goal. 
• Use Teamhood terminology in your prompts. Prefer Item instead of task, Board instead of project, Workspace, Tracked time, and other Teamhood-specific terms. 

Practical Example: Building a Multi-Phase Project 

A user preparing an implementation project can upload a planning document, ask for a new board, and receive a complete structure with discovery, preparation, execution, and launch phases. Teamhood AI reads the document sections and creates groups that match the phases. Tasks appear under each phase, sub-tasks reflect actions or deliverables, and dependencies connect sequential items such as design, handoff, and deployment tasks. 

Practical Example: Process Setup for a Procurement Team 

A procurement manager can describe the sourcing and approval process. Teamhood AI produces a board with statuses such as Request Received, Review, Supplier Contact, Negotiation, Approval, and Purchase. Swimlanes group tasks by request type. The team can start using the board immediately and refine naming as needed. 

Feedback

Hit like or dislike icons after response to give feedback whether Teamhood AI is going the right direction. Also, share your requests and improvement ideas in our customer portal.