A project manager that knows what you agreed
Tresmeta reads the documents, the meetings, the emails and the context of a project, keeps a living knowledge base from them, and surfaces the tasks, commitments and scope changes they contain.
What it solves
Projects are not lost for lack of information. They are lost because the information is spread across minutes, emails, proposals and conversations nobody has time to read again.
The result is always the same. What was agreed in a meeting three months ago is no longer remembered by anyone. A 'we'll send it on Friday' said in an email never reached any task list. And scope grows through requests that are reasonable one by one and add up to a month of work nobody budgeted for.
Tresmeta goes after that: it turns the documents that already exist into a knowledge base of the project and into concrete proposals for tasks, commitments, decisions or scope changes. Each proposal says which document it came from, and nothing enters the board without a person approving it.
What it does
- Reads minutes, proposals, contracts, emails, spreadsheets and transcripts, in the usual formats
- Builds a project knowledge base that updates as new documents arrive instead of freezing in time
- Proposes tasks, commitments, decisions and scope changes in an inbox where someone accepts, corrects or discards
- Keeps the origin of every task and decision: the document and the fragment it came from
- Flags duplicate tasks, tasks outside the agreed scope, and drift from what was decided in earlier meetings
- Measures its own work and shows its failures on a control panel instead of hiding them
- Configures to how you work: your preferences and your way of managing. And to ask it something or pull information out you do not need a dashboard, you just ask in a conversation
Our part in it
Tresmeta is a product built together with Scalabble. Our work is the tuning, and it is worth explaining why that is needed.
A system like this is not deterministic: it interprets the meaning of the context you give it, and that context is no longer only text. It is a recording of a meeting, a photo of a whiteboard, a video, an audio file. A system of this kind does not work fully unless its ontologies and skills are tuned to a concrete context: the vocabulary of an architecture practice is not that of an agency or an engineering firm, and what counts as a scope change in one place is the normal way of working in another.
That tuning is what lets the system tell a duplicate task from a similar one, a request outside scope from a clarification, or a real drift from what was agreed in a project's meetings.
And there is a part of the context no form ever captures, which is exactly the part that decides whether a project goes well: people's real preferences rather than the ones they declare; someone's leadership style when nobody is watching; how they decide under uncertainty; the biases they do not know they have; and the emotional situation surrounding each choice. Capturing and modelling that living context is where the tool either gets it right or gets in the way.
Tresmeta is not a download: it is implemented and supported. You start with a real project and watch it run on real documents, which is the only way to know whether the tuning is right.
Got a process that hurts?
Tell us about it in half an hour. No forty-page deck. If we are not the right people, we will say so and point you towards someone who is.