Treat it as a project
A CRM migration moves your customer records, history and settings from one system to another. It is rarely difficult in principle, but it is easy to do carelessly. Typical problems are missing notes, merged or split records, lost dates and fields that land in the wrong place. A little planning prevents most of these.
This article assumes you have already decided to switch. If you are still comparing products, the article on choosing a CRM covers what to ask, including how easily data can be exported, which is worth asking even before you sign up, since it also determines how easily you can leave later.
Step 1: Audit what you have
List everything the current system holds.
- Records: contacts, companies, deals and leads.
- History: notes, emails, calls, meetings and tasks.
- Attachments: documents and files.
- Configuration: custom fields, tags, pipeline stages and user permissions.
- Connected tools: forms, email software, calendars and reporting.
Note roughly how many of each there are, so that you can compare counts after the move. Also check what the old system can actually export. Some tools export contacts easily but not attached files or the full activity history. Find this out early, while you still have time to find a workaround, such as saving key documents separately.
Step 2: Decide what to bring
You do not have to migrate everything. Moving old, wrong or unnecessary data just brings the clutter along.
- Records with no activity for a long time may be worth archiving rather than importing.
- Data you have no reason to keep should be deleted according to your retention rules, in line with the principle of collecting only what you need.
- Some history may be kept as a read-only archive file instead of living in the new system.
Rules about keeping and deleting personal data vary by country and change, so check what applies to you. This is not legal advice.
Step 3: Clean before you move
Cleaning is much easier in a familiar system than during an import. Do the main tasks first:
- merge duplicates;
- fix obvious errors in names, emails and company names;
- standardise formats such as phone numbers, countries and status labels;
- remove test records.
The routines are explained in the article on keeping CRM data accurate over time. Make a full backup export before changing anything, and keep it safe, since it contains personal data.
Step 4: Map the fields
Field mapping means deciding which column in the old data goes into which field in the new system. Make a simple table:
| Old field | New field | Notes |
|---|---|---|
| Full name | First name + Last name | Needs splitting |
| Phone | Phone | Standardise format |
| Status | Lifecycle stage | Rename values |
| Notes | Notes | Check length limit |
Points to check:
- Different structures. One system may store a single “name”, another two fields.
- Field types. A dropdown in the new system accepts only fixed values, so free-text entries must be matched to the allowed options.
- Dates and time zones. Date formats can flip day and month if you are not careful.
- Unique identifiers. Keep the old record ID in a spare field. It makes checking and troubleshooting much simpler.
- Relationships. Make sure contacts remain linked to the right company and deals.
- Consent and unsubscribe status. Carry these over so that you do not email people who opted out. This is important, and rules differ by location.
Step 5: Test with a small sample
Never run the full import first. Import perhaps fifty representative records, including awkward ones such as long names, special characters, and contacts with many notes. Then check them against the originals, by hand.
Look for:
- scrambled characters in names;
- lost or shifted dates;
- fields that have been truncated;
- missing history or broken links between records;
- duplicates created by the import.
Adjust the mapping, and repeat until the sample looks right.
Step 6: Run the full import and verify
When the sample works:
- Pick a quiet time, and tell the team that changes in the old system should pause.
- Run the full import.
- Compare totals: contacts, companies and deals against your audit counts.
- Spot check a random selection of records, plus your most important customers.
- Test search, filters and reports.
- Confirm connected tools, such as forms and email sync, point to the new system.
Small count differences can have legitimate causes such as merged duplicates. Make sure you can explain every difference.
Step 7: Switch over carefully
Consider running both systems side by side for a short time, with the old one set to read-only. This gives you a fall-back if something was missed. Keep the old account, or at least a full export, for a period you decide in advance, then close it and delete data according to your retention rules.
Plan training too. A migration can be technically perfect and still fail if the team does not know where things are. Agree on the rules for entering data from day one, and note who owns what.
What can go wrong
- Underestimating the time needed for cleanup.
- Not checking that exports include history and files.
- Mapping all text into free-text fields, which loses the ability to filter.
- Cancelling the old subscription before verifying the new data.
- Forgetting to re-create integrations.
Outcomes vary, and some loss or reformatting is possible even with care, which is why backups and verification are the core of the plan.