Does it replace the CRM?

No. The CRM stays the system of record. Nemosyne reads it alongside the documents and reports where the two disagree — a stage reading “approved” against minutes recording an outstanding condition, for instance.

What about custom objects?

Records are records. What a typed field means comes from your ontology, which is a per-firm composition rather than a fixed schema.

Source

Salesforce, joined to the documents

The CRM knows the pipeline and the documents know the reasoning, and the two have never been joined. Nemosyne reads Salesforce records as records — typed fields stored exactly, not interpreted by a model — and resolves them to the same companies and people the documents talk about.

Start free →

What it captures.

Accounts, contacts and opportunities — the companies and people your documents also name.
The stage history, which is what makes a CRM row disagree with a set of minutes in a way worth surfacing.
Notes and descriptions, which are prose and do get read.

How it behaves.

Authentication: OAuth, against your Salesforce org.
Permissions: Firm-wide — see below.
Scope: The org the credential belongs to.
A document: A record. An account, a contact or an opportunity each count as one, which is why a CRM is cheap to read compared with a document archive.
Sync: Continuous; only what changed is re-read.
Writes: None. Salesforce stays the system of record for the pipeline.

What this means for permissions.

Salesforce is read firm-wide, not per person. The connector reads at organisation scope rather than reproducing Salesforce's own sharing rules, so anything it captures is visible to everyone in your memory, regardless of who could see that record in Salesforce. If your CRM holds records some colleagues must not see, connect it with that in mind, or do not connect it.

What we enforce, and the limits we publish →

Questions this raises.

Does it replace the CRM?: No. The CRM stays the system of record. Nemosyne reads it alongside the documents and reports where the two disagree — a stage reading “approved” against minutes recording an outstanding condition, for instance.
What about custom objects?: Records are records. What a typed field means comes from your ontology, which is a per-firm composition rather than a fixed schema.

Connect it to what you already use.

Free for three people and about a hundred documents. No card, no expiry.

Start free →

See also SharePoint · Google Drive · Slack · Gmail · HubSpot