Migrating to Zoho CRM: what moves and what breaks
The first decision in a migration is not technical. It is how much of your past you are willing to pay to carry forward.
Most clients want everything. Then they see the cleanup estimate for records from 2016 and discover they did not want everything after all. That conversation is better had in week one than week six, so I push it early.
What is usually worth moving
| Data | Move it? | Why |
|---|---|---|
| Open deals and current pipeline | Always | Your team needs these on day one |
| Active accounts and contacts | Always | The working core of the system |
| Closed deals, last 2 to 3 years | Usually | Reporting baselines and renewal context |
| Notes and activity history | Selectively | Valuable on live accounts, noise elsewhere |
| Email threads | Rarely | Expensive to move, and your mailbox already has them |
| Anything older than 3 years | Usually not | Archive the export; do not pay to clean it |
| Custom fields nobody fills in | No | Migration is the cheapest time to drop them |
That last row is the one clients thank me for later. Every CRM accumulates fields someone added for a campaign in 2019. Carrying them over means carrying over the confusion too, and nobody ever audits them once they are live in the new system.
What breaks, and why
Duplicates are the most common casualty. Imports match on a key, usually email, and if your source has three records for the same person with different addresses, Zoho will faithfully create three contacts. Deduplicate before import, not after. Fixing it afterwards means reconciling deals and activities that have already attached themselves to the wrong record.
Record ownership goes next. If the owner field does not map to a real Zoho user with a matching identifier, everything lands with the admin. Your reps open the CRM, see an empty pipeline, and conclude the migration failed. Technically it did not, but by then the damage to their confidence is done.
Then attachments and notes, which tend not to come out in a standard export and need either an API pull or a separate file migration. If nobody mentions them in the scope, assume they are not included and ask.
And relationships. Contacts belong to accounts, deals belong to both, activities belong to all three. Import them in the wrong order and the links do not form, leaving you with orphaned records that look fine individually and are useless in reporting.
Picklists quietly mangle things too. If your old system had free-text where Zoho has a dropdown, every value that does not match an option either fails or lands blank. Reconcile the option lists first.
Why the test import matters more than the plan
A test import is a small, representative slice, maybe 50 to 100 records chosen to include the awkward cases: the account with four contacts, the deal with a custom stage, the contact with a note attached. You import it, you walk the records in the interface, and you count.
What you learn in two hours of that is worth more than two days of planning, because it tells you what the real data does rather than what the field map says it should do. Then you fix the mapping and run it again. A migration that goes in once with no rehearsal is a migration where you find the problems in production.
When your source is a spreadsheet
In some ways easier, because there is no relational structure to preserve. The work moves instead into deciding which columns deserve to be fields, and into deduplication, which spreadsheets are particularly bad at hiding.
One thing to settle early: whether your sheet has one row per deal or one row per customer. Teams often maintain a hybrid without realising, and the split into Contacts, Accounts and Deals has to be decided by someone who understands the business rather than by whoever runs the import.
Questions I get asked
Can we keep the old CRM running in parallel?
For a week or two as a read-only reference, yes. Longer than that and people keep working in both, which means neither is accurate. Set a date for read-only and a date for switch-off, and tell everyone both dates at the start.
How do we know the migration was complete?
Reconcile counts by object, then spot-check. Counts catch the big misses, spot-checks catch the quiet ones: an account with the wrong owner, a deal with a blank stage. Ask your partner for the reconciliation numbers in writing rather than a verbal confirmation that it went fine.
Who should clean the data, us or the partner?
Split it. Your partner handles the mechanical work: format standardisation, deduplication rules, mapping. You handle the judgement calls, because only your team knows whether two similar accounts are the same company or a parent and subsidiary. Clients who hand the whole thing over pay for a consultant to guess at their own business.
Disclosure: This site is written and operated by Flowpeak Consulting, which sells Zoho implementation work. Weigh the advice accordingly.
Keep reading
What it costs
Why data cleanup is the line item that moves most, and how to get a real number.
Read the guide →