Settld reads every call, thread and email chain, files each decision with an owner and a deadline, and warns you when a new one reverses an old one.
First 10 conversations free, for up to three people. No credit card. Nothing joins your calls and nothing sits reading your channels.
Move the mobile app redesign ship date from October 15 to November 5.
“ok — moving the redesign ship date to Nov 5. that replaces Oct 15, I’ll update the roadmap”
Contradicts “Ship the mobile app redesign on October 15” — decided 14 Jul in Q3 Planning. Needs review.
Switch the notifications service datastore from Postgres to DynamoDB.
In-app chat stays cut from v1 and remains in the v2 backlog.
Works with where your team already talks
The feature gets cut on a Tuesday call. Three weeks later it is quietly reinstated in a Slack thread six people can see. The deadline moves in a reply to a reply on an email chain the owner was never copied on. Each tool holds its own fragment, none of them holds the record, and nobody notices the three contradict — so six weeks on the same argument happens for the third time.
Notetakers made this worse, not better. A summary of everything said is not a record of what was settled — and it only ever covers the calls.
No bot joins your call. Nothing sits reading your channels. Every conversation reaches Settld because a person sent it there.
React to a Slack thread, forward an email chain, use Send to Settld on a Teams message, or drop in a meeting transcript. Settld detects the format and normalises it.
Every extracted decision arrives with an owner, a deadline, a confidence score, and the exact line it came from. You accept, edit, or reject.
New decisions are checked against everything already on record, whatever they arrived from. Contradictions are flagged for a human — never auto-resolved.
Extracting decisions is table stakes. The value is knowing when a decision you already made stops being true — and telling the person who owns it.
“We’re going with Postgres” and “switch the notifications store to DynamoDB” share no words. Settld compares meaning, so a reversal phrased entirely differently still surfaces.
A missed contradiction costs weeks. A false flag costs ten seconds. Settld errs toward showing you the pair and letting you dismiss it.
A model deciding which of two decisions is “right” would be a liability. Settld surfaces the pair with both sources and a person chooses.
A full replacement supersedes the original and links back to it. A partial change is recorded as a resolved conflict with a note, so the original still reads as the decision it was.
Each one links back to the message or the transcript line it came from. When someone disputes it, you open the line.
Pulled only when someone actually accepted the work. No owner is recorded when nobody took it.
Hedged discussion scores low and routes to review. Confident extraction of ambiguity is the failure mode Settld avoids.
One email: decisions past their deadline, decisions with nobody on the hook, and conflicts waiting for a human.
Filter by owner, project, status, or date. Find what was decided about billing in March without opening Slack, your inbox and six documents.
No bot in your meeting, no browser extension, no third-party notetaker holding your recordings — and nothing reading a channel you didn’t point it at.
The transcript your platform already produces. Nothing new attends the call.
.vttAn app you invite to the channels you choose. React with ✅ and it reads that thread — and nothing else.
Your workspace gets its own address. Settld never connects to a mailbox and sees only what is sent to it.
The … menu on any message. It reads the thread around that message, as you, and stops there.
For meetings, your workspace admin must have cloud recording and transcription enabled — Settld reads what those produce and can’t turn them on for you.
Settld has no ambient access to anything. In Slack it is an app you invite to specific channels, and it reads a thread only when someone reacts to it. The email address receives only what is forwarded to it. In Teams it reads the thread around the message you actioned, with your permissions, and stops there. No channel, inbox or calendar is synced in the background.
What you send is processed to extract decisions and is never used to train models.
Every row is filtered by your organisation at the database level, not by application convention. A query that forgets to scope returns nothing rather than someone else’s data.
Uploads live in private object storage and are never served from a public URL. They’re fetched server-side after an ownership check.
Single-use, fifteen-minute links. Nothing to reuse, leak, or store.
Removing a conversation removes its text, its decisions, and the stored file.
You pay for the people who send conversations in and keep the record straight. Reading it stays free for everyone else — a record half the organisation can’t see isn’t a record. The monthly allowance is shared across the workspace, so one person’s heavy week doesn’t block the rest of the team.
10 conversations, up to 3 people
Start free20 conversations per person, per month
Start free trial60 conversations per person, per month
Start free trialA conversation is one thing you sent: a meeting, a thread, a forwarded chain — a fifty-minute call and a six-message Slack thread each count as one. Allowances pool, so five people on Team share 100 a month however they fall between them.
No. Settld never joins a call and never records anything — it works from the transcript your platform already produces. In Slack it is an app you invite to specific channels, and it reads a thread only when someone reacts to it. In Teams it reads the thread around the message you send it, using your own permissions rather than tenant-wide access. Nothing is listening in the background.
A notetaker summarises what was said, on the calls it was invited to. Settld extracts what was settled, wherever it was settled — and keeps it, so a decision made in July can be checked against one made in October. A summary has no memory across conversations; that memory is the entire point here.
Nothing reaches your record without a human accepting it. Every extraction lands in a review queue with its confidence score and the line it came from. Hedged discussion scores low precisely so it surfaces for a person rather than being asserted.
It’s stored in private object storage, encrypted at rest, isolated per organisation at the database level, and never used to train models. Deleting a conversation deletes its text, its decisions, and the stored file.
It will tell you so. A status update, or a thread that argues and settles nothing, returns an empty result rather than inventing something to justify the trip. The empty case is tested explicitly.
When a conversation confirms something settled earlier — “that’s decided, two weeks instead of one” — Settld records it and marks it as previously decided, so the date it entered the record isn’t mistaken for the date it was made.
Send it one thread or one transcript and see what your team actually decided last week. Takes about a minute.
Start free