Is this a two-way sync?
No, and it is worth being blunt about it. Data comes in when you import, and one specific thing goes out: creating a contact from an automation. Edits made in RowFold do not flow back to HubSpot, and edits made in HubSpot do not appear in RowFold until the next import, which Keep in sync can run hourly or daily. Treat HubSpot as the source for contact and deal facts, and RowFold as the source for everything you build around them.
Am I supposed to replace HubSpot with this?
No. If your problem is the pipeline, sequences and marketing, HubSpot is better at that than a general workspace will be. This connector exists because the other half — what your team actually delivers after somebody says yes — is usually in a spreadsheet, and that is a problem RowFold is good at.
What happens if the contact already exists?
The step fails for that record rather than creating a duplicate or quietly overwriting the existing contact, and the failure is visible in the automation's run log. That is the safer direction: a CRM full of near-identical contacts is a much more expensive problem than an automation you can see refusing. Put a condition in front of it if duplicates are expected.
Do custom properties come across?
The import brings the objects and their fields as tables you own. Where something does not map cleanly, the answer in RowFold is usually not to force it into the imported table but to put it on the table next door — the one modelling your delivery — where you control the shape and are not constrained by anybody's object model.
Which HubSpot plan do I need?
Any plan whose account can authorise an app, including the free tier. The permissions requested are read access to contacts, companies and deals, plus the ability to create a contact — none of it gated behind a paid HubSpot feature. If your account can install an app at all, it can connect this one.
Can I import again to pick up CRM changes?
Yes. Importing again updates the tables the first import made, matched on HubSpot's record IDs: new contacts, companies and deals are added, changed ones updated, and the links your own tables made to them stay put. Keep in sync does the same hourly or daily. Nothing goes back to HubSpot, so a value edited here in a column the import fills is replaced by HubSpot's on the next run. Keep your own facts in columns or tables you add, which the import never touches.
Do HubSpot's associations come across?
No. Each object is imported on its own with the properties HubSpot returns for it, so the contact-to-company and deal-to-company associations are not rebuilt as links here. A contact does carry its company NAME as text, and Convert to Relation… turns that column into real links to the imported companies in one pass, after which later imports keep filling it for the contacts they add. A deal carries neither its company nor its contacts, so deals are linked by the tables you build around them — the install, the onboarding, the casework that points back at the deal — rather than the other way round.