How to migrate from HubSpot without losing deal history
Updated
The short answer. To leave HubSpot without losing deal history, import the current state of your companies, contacts and deals from a CSV, let the conversations behind each deal come back from your email, calendar and meetings, and keep HubSpot's own logged notes and activities as an archive; moving the deals over takes about a week. Day AI does this for teams that run it in place of their CRM, or on top of Salesforce or HubSpot: its agent imports HubSpot's companies, contacts and deals from a CSV, and the record builds itself from your team's email, calendar and meetings.
Say you run sales at a B2B payments company, and it's the first morning after you moved pipeline review off HubSpot. You run Day AI in place of your CRM now, and your morning briefing is waiting, as is every rep's. Nobody retyped the old deals: the open ones came over from HubSpot's CSV export, and the emails and meetings behind them came back from your team's Gmail, Google Calendar and recorded calls. Since yesterday, the agent has read the new emails and meetings on those deals, created the deals that were new and moved the stages that changed. It stopped short of closed-won and closed-lost, because that's the guardrail you gave it: those two calls stay yours. You open a deal it moved and see the email the change came from. Last night your reps found follow-up drafts waiting at the end of the day. This morning you know what's going on in the pipeline, and none of it depended on what someone remembered to type into HubSpot.
Where the usual answer goes wrong. That's the deal history most teams are afraid to lose when they leave HubSpot. Most migration guides treat deal history as rows to move: pull every logged call, email and note out of HubSpot through its API, map them into the new CRM's activity tables, and hope the associations survive. That's the step that eats the project. HubSpot's timeline is mostly a copy of conversations that happened in email and meetings, and the conversations are the record. Day AI rebuilds that history from the source and leaves HubSpot's copy in an archive. If your plan so far is a spreadsheet of fields to map and a renewal date, that was the right start.
The three moves of a one-week HubSpot migration, with Day AI
A migration that keeps deal history has three moves, in this order. Each one has a different source, so each one is done a different way, and together they fit in a week.
- Connect the conversations. Connect email, calendar and call recordings first, so the history behind every deal comes back from where it happened. In Day AI, each person connects Gmail and Google Calendar, the Day AI notetaker records meetings, and Gong brings your call history with transcripts.
- Import the current state. Export companies, contacts and deals from HubSpot as CSV files and bring them in with the stages, amounts, close dates, owners and people on each deal. In Day AI, the agent's Import Data skill maps the columns and checks the file before it writes anything.
- Cut over and archive. Export HubSpot's notes and activities as an archive and move pipeline review to the new system. In Day AI, every value from then on shows its sources, and you can ask the agent what changed and why.
How to migrate from HubSpot with Day AI
Step 1: Connect email, calendar and meetings to Day AI before you import anything
Start with the conversations, not the CSV. Each person connects their own Google Workspace account (Gmail and Google Calendar) and sets their sharing rules: share everything or nothing by default, with exceptions by domain, address or Gmail label. The first sync finishes within hours, and agents work from your history from then on. Turn on the Day AI notetaker for meetings going forward. If your team records calls in Gong, connect it and Day AI brings in your call history with transcripts; Granola connects with transcripts and notes.
Doing this first means that when the deals arrive in step 2, the emails and meetings behind them are already in Day AI.
Step 2: Import HubSpot companies, contacts and deals into Day AI from CSV
In HubSpot, export each object from its view. HubSpot's own help page says a record export "downloads their current property values and associations" (HubSpot Knowledge Base, read October 29, 2026). Choose "All properties and associations on records" so each deal keeps its company and contacts. HubSpot emails a download link, and the same page notes that "Download links to export files will expire after 30 days," so save the files as soon as they arrive.
Then, in Day AI, open Imports in the left navigation and click Start New Import, or upload the CSV in any chat. The agent's Import Data skill takes it from there, in this order: companies, then contacts, then deals.
- Columns to fields. The agent maps each column to a Day AI field and shows you the mapping. For columns with no matching field, like closed-lost reason, lead source or deal type, it creates a custom property so nothing is dropped.
- Stages. It maps every HubSpot stage in the file to a stage in your Day AI pipeline. If you don't have a pipeline yet, ask the agent to set one up first.
- Owners and people on the deal. It matches deal owners to your teammates and brings the contacts on each deal over as the people on that deal.
- A check before anything is written. The agent runs the whole file through the import's rules first and tells you which cells would fail and why, then fixes what it can. Keep the email column on contacts and the domain or website on companies; when rows lack them, the agent tells you which ones may come in as duplicates.
- Progress and errors. Large files run in the background. Ask the agent how the import is going, and it lists any failed cells with the reason, so you can fix and re-import them.
The import covers companies, contacts and deals. For HubSpot custom objects, set up matching custom objects in Day AI and load the records through the Day AI API.
Step 3: Let Day AI build the context around each imported deal
Once the deals are in, you don't type their history back in. In Day AI, every email and meeting is organized by the same people, companies and deals, so opening a deal shows the conversations behind it next to its stage, amount and close date. Answers about a deal cite the call clip or the message they came from.
From here the record keeps building. Turn on the Opportunities: Creation and Stage Movement skill. It starts in learning mode: it tells you which deals it would create and which stages it would move, and asks for feedback before it's allowed to act. Tell it your guardrails, such as stopping before closed-won and closed-lost, and it edits the skill so later runs follow them.
If a HubSpot deal field never got filled in, a workspace admin can add it in Day AI under Settings > Schema with Add Property, or from a deals table with Columns > Create new property. Describe a property and an agent fills it from your conversations, and you can ask the agent to backfill it across the deals already in Day AI. Check a few deals before you rely on the values.
Step 4: Cut over to Day AI and archive HubSpot
Move pipeline review to Day AI first. From the day you switch, every value in Day AI shows its sources, and you can ask the agent what changed and why. You don't have to carry every old activity into Day AI; decide on purpose which ones you archive.
Some deal history exists nowhere but HubSpot: a note a rep typed after a hallway conversation, a pricing exception someone logged. The CSV import doesn't carry HubSpot's timeline of notes and activities, so before you cancel HubSpot, export that archive. HubSpot's help page says that to export activities such as calls and notes you can "Export individual contact data," "Export certain activity reports," or "Use the engagements API" (HubSpot Knowledge Base). Keep the deal export from step 2 with it: it records where every deal stood the day you left. If a note on an open deal still matters, add it to that deal in Day AI as a note.
A one-week HubSpot migration with Day AI
Moving the deals takes about a week. Start it whenever you're ready; you don't have to wait for your HubSpot renewal.
Day 1: Connect conversations to Day AI
Everyone connects Gmail and Google Calendar and sets sharing rules. Connect Gong if you use it. Turn on the notetaker. Set up your pipeline and stages in Day AI.
Days 2 and 3: Import into Day AI
Export companies, contacts and deals from HubSpot with all properties and associations. Import them through Imports in Day AI, companies first, then contacts, then deals. Fix the failed cells the agent lists.
Day 4: Check the deals in Day AI
Open the deals you work most and check that each one shows its stage, people and the conversations behind it. Turn on the opportunities skill in learning mode and correct it as it goes.
Day 5: Cut over to Day AI
Export HubSpot notes and activities as your archive. Run pipeline review from Day AI. Cancel HubSpot when you're ready.
Where a field-by-field HubSpot migration breaks
- The activity copy is the slow part. Pulling every logged email and note through an API and mapping it into a new activity table is where projects stall, and the associations are what break. In Day AI, the email and meeting history comes back from Gmail, Google Calendar and your recordings, so the import only carries the current state and the move takes about a week.
- The new system starts empty the day after. A migrated CRM holds what was typed up to the export date, and then waits for reps to type again. In Day AI, the record keeps building from every email and meeting after the switch, and agents create deals and move stages with the guardrails you set.
- Fields nobody filled in stay empty. Mapping a blank HubSpot field gives you a blank field somewhere else. In Day AI, describe the property and an agent fills it from your conversations, and you can ask it to backfill the deals already in Day AI.
- Cutting over without an archive loses the notes. Once HubSpot is cancelled, the notes that lived only there are gone. In Day AI, the conversations behind each deal are already there, so the only thing left to keep is HubSpot's own export, saved before you cancel.
What Day AI adds after the switch
Day AI provides the context graph and agent control plane to quickly and securely upgrade your company brain. After a migration, that matters because you don't rebuild HubSpot's workflows field by field: agents have a name, a job and skills, and skills run on a schedule, when a meeting is recorded or an email arrives, or on request.