Short answer
To migrate from HubSpot to Krokul, create a private app with read scopes, connect its token, match pipelines, stages and owners, run a dry run that writes nothing, then import. Krokul reads companies, contacts, products, deals and tasks with their associations, writes them in a fixed order and keeps every HubSpot ID. A rollback removes what a run created.
- Read from HubSpot: companies, contacts, products, deals, tasks
- Written in this order: companies, contacts, products, deals, tasks
- Stays in HubSpot: deal line items, email threads and call logs
On this page
What transfers when you migrate from HubSpot to Krokul#
Migrating from HubSpot moves five kinds of records and keeps the links between them. The method behind the order is described in the guide on how to switch CRM without losing data, and this page applies it to HubSpot.
The HubSpot connector reads through the HubSpot CRM API version 3 and turns every HubSpot object into a Krokul record. The table shows what each object becomes and which of its fields arrive.
| HubSpot | Becomes in Krokul | What arrives |
|---|---|---|
| Companies | Client companies | The name, or the domain when the name is empty, plus the description and the address or the city in the notes. Every other simple property is kept in the record’s custom data. |
| Contacts | Contacts | The first and last name, the email, the phone and the mobile phone, the job title and the link to the first company associated with the contact. |
| Products | Products | The name, description, SKU, price and the cost of goods as the purchase price. The currency is always UAH, and an archived product arrives inactive. |
| Deals | Deals | The deal name, the amount as the expected amount, the currency, the close date as the expected close date, the pipeline and stage, the owner and the links to the first associated contact and the first associated company. |
| Tasks | Tasks | The subject as the title, the body and a note that the task is completed in the description, the due date with its time, the owner as the assignee and the links to the first associated deal and contact. |
| Deal pipelines, stages and owners | Reference data | Read together with the records, then matched to your Krokul pipelines, stages and users by mapping rules. |
| Emails, calls and notes | Not transferred | Export them from HubSpot and keep the archive. |
| Deal line items | Not transferred | HubSpot keeps them as a separate object with its own rights. Export them and add the items to the deals in Krokul as needed. |
Sources and tags are not read either, and neither are the tickets of a HubSpot account.
In what order does a HubSpot migration run#
A HubSpot migration runs in six steps, and Krokul writes the records in a fixed order: client companies, contacts, products, deals and then tasks, so every association finds its target already in place. The general list of checks is in the CRM migration checklist.
Count and clean the records in HubSpot
Write down how many companies, contacts, products, deals and tasks you have, and note the currency of the deals. These numbers are the totals you compare with at the end.
Merge the obvious duplicate companies and contacts first. A duplicate in HubSpot arrives as a second record or as a skipped row in Krokul.
Prepare users, pipeline and stages in Krokul
Invite the people who own HubSpot records, because the import never creates users, and a deal with an unmatched owner gets its responsible person from the usual Krokul rules: the manager of the client company or contact, or the pipeline pool. Prepare the pipeline and its stages, so each HubSpot deal stage has a Krokul stage to land in.
The stages are matched in step 4 by rules, so prepare them before that step.
Create a private app in HubSpot and connect its token
In HubSpot open Settings, Integrations, Private apps and create an app with read scopes for companies, contacts, deals, products and owners. Tickets are not needed. Copy the access token.
Open the Migration screen in Krokul as an administrator, choose HubSpot as the source, paste the token into the private app token field and click Check connection. The server reads the token, and the screen does not show it again after the check. The connector only sends read requests, so nothing changes in HubSpot.
Tip. The API base field can stay empty. It defaults to the HubSpot cloud address.

Connecting HubSpot in a Krokul demo workspace: the token field, the hint with the read scopes and the connection check. Match pipelines, stages and owners
HubSpot deal stages become Krokul stages by mapping rules, and HubSpot owners become Krokul users by rules as well. The Migration screen has no step for these rules today, so they are written to the run through the migration API, by an implementation partner or by an administrator who works with the API.
A deal whose pipeline has no rule lands in the first mapped pipeline, and a deal whose stage has no rule lands in the first stage of its pipeline. With no pipeline rule at all and no default pipeline, a deal is reported as a failed row with the reason that no pipeline is mapped and no default pipeline is set.
Run a dry run and read the report
Start the dry run from the check step. It takes the token from the connection check on the server, reads the whole HubSpot account at 100 requests per minute, applies the rules and writes nothing in Krokul.
Compare its totals with the counts from step 1 and repeat the dry run after every change to the rules, until each failed row has a reason you accept and the skip counts match your audit.
Run the import and check the summary
Start the real run. It works on the server in the fixed order, so you can pause and resume it, or close the page and come back to the summary. The summary lists created, updated and failed rows, and each failed row comes with its reason.
Compare the totals with step 1, open twenty records chosen at random and compare every field with HubSpot.
How do HubSpot associations become links in Krokul#
HubSpot associations become links in Krokul, and the connector follows the first association of each kind. A contact is linked to the first company associated with it, a deal to its first associated contact and first associated company, and a task to its first associated deal and contact.
- A contact that belongs to several companies in HubSpot is linked to one company in Krokul. Decide in step 1 which company should stay the main one.
- A deal with several contacts shows one contact in Krokul. The other associated contacts are imported as separate contacts without a link to that deal.
What does a dry run of a HubSpot migration show#
A dry run of a HubSpot migration shows, for each kind of record, how many Krokul records the real run would create, update and skip, and lists the rows that would fail with the reason for each. The connection check alone cannot do this, because a HubSpot list returns no totals, so the connection check only proves that records exist.
| Records | Would be created | Would be updated | Would be skipped |
|---|---|---|---|
| Client companies | 310 | 0 | 12 |
| Contacts | 980 | 0 | 64 |
| Products | 42 | 0 | 0 |
| Deals | 415 | 0 | 0 |
| Tasks | 1,540 | 0 | 0 |
In the example nothing would be updated, because no earlier run exists. The report only counts the 12 companies and 64 contacts that would be skipped, because Krokul skips duplicates by default, so decide what to do with each one. The dry run skips write validation, so the real run can still fail on rows it passed.
How to roll back a HubSpot import that landed wrong#
A rollback of a HubSpot import removes the records that this import created and leaves everything else alone. It applies to one run at a time and is available only for a real run.
- Records the run created are removed, with the last kind of record going first.
- Records the run only updated keep their new values, so correct them by hand if needed.
- Records that someone changed after the import stay in place.
- A dry run created nothing, so there is nothing to roll back.
Krokul keeps the HubSpot ID of every record it writes, so a second run updates the records that already moved and creates no duplicates.
What stays behind or needs a decision when you leave HubSpot#
The HubSpot connector was written from the HubSpot documentation and tested on simulated responses. It has not been run against a live HubSpot account yet, so the dry run on your own data is the first real test, and a transfer without it should not start.
- Emails, calls, notes and the line items of deals are not transferred. Export them from HubSpot first.
- Products are written with the currency UAH, whatever currency HubSpot shows. A deal without a currency code is written with UAH too, so check the prices and the deal currencies after the dry run.
- A HubSpot task arrives as a Krokul task with the time of its due date. A task that is already completed in HubSpot gets a note about it in the description.
- Tasks are assigned only to the owners that you matched in step 4, and a task without a matched owner has no assignee, unless a default assignee is set through the API.
Who should run the move: your team or an implementation partner#
A company with a short list of contacts and a few users can connect HubSpot, run the dry run and read the report on its own. A business with a running pipeline, several users and years of deals gets the pipeline, stage and owner rules, the reading of the dry run and the final check from an implementation partner, who also trains the team.
The Krokul migration tool lists the sources it reads, what it moves and what it leaves behind, and Krokul compared with HubSpot puts the two systems side by side. Another source follows the same steps, for example migrating from Zoho CRM.
Which mistakes cost the most when you migrate from HubSpot#
- Starting the import before the pipeline rules exist. Deals then land in the first mapped pipeline and its first stage, or fail with the same reason when no pipeline is mapped, so write the pipeline and stage rules first and repeat the dry run.
- Skipping the dry run. The same mapping error repeats on every record, and only a dry run shows it before anything is written.
- Closing HubSpot on the import day. Emails, calls and notes cannot be exported later, so keep the account readable for a working week and export them first.
- Leaving owners unmatched. Deals then lose the HubSpot owner, so invite the users and match them before the run.
Frequently asked questions#
How long does a HubSpot migration take?
The connector reads HubSpot at 100 requests per minute, so a large account takes longer, and no timing has been measured on a live HubSpot account, so there is no estimate here.
Does HubSpot change during the migration?
No, the connector only sends read requests to HubSpot. Your HubSpot data stays as it is, which is why you can keep the account open until the check passes.
Are emails and calls moved?
Emails, calls and notes are not moved. HubSpot tasks do arrive, as Krokul tasks. Export the emails, calls and notes from HubSpot and keep the archive.
What happens to duplicates?
By default the import skips a record that duplicates one already in the workspace, and the dry run report counts the skipped rows.
Who can run a migration in Krokul?
Only an administrator runs a migration. The token you paste is read by the server and is not shown again after the connection check.