The client is a global medical device company focused on musculoskeletal care, with a portfolio spanning spine, orthopedic trauma, joint reconstruction, regenerative biologics and surgical navigation and imaging technology. Its mission is to improve the quality of life of patients with musculoskeletal disorders, and its commercial teams work closely with surgeons and healthcare providers around the world. For a business like this, the relationships and history captured in the CRM are a working asset, not an archive.
After years on Salesforce, the client made a strategic decision to consolidate onto Zoho. Its Salesforce org held content in three generations of storage: Classic Attachments, the modern ContentDocument and ContentVersion file model, and Notes, each with its own object model, its own limits and its own quirks.
A CRM migration is judged by how much of the business history survives it. For a medical technology company, the documents and notes tied to accounts and customer records had to arrive in Zoho linked to the right parent records, with ownership and timestamps intact. That made a dependable Salesforce data extraction process a precondition for the whole programme, not an afterthought.
Modern extraction runs across all objects in a single pass and also captures files posted to Chatter, so the customer no longer has to run object-by-object or manage a separate feed export. To work within Salesforce’s query restrictions, the tool starts from the documents, applies the date filter there, and then reads links, feed attachments, posts, comments and latest versions by Id in safe batches. The key behaviours:
Notes are delivered as Notes.csv using Zoho’s own note-import headers, with owner, creator and modifier email addresses resolved from Salesforce users. Note content retains Salesforce’s original markup, so inline images were not migrated as files. They remain directly linked through their image links within the note content. Alongside it, Notes_Images.csv lists every inline image, classifies where it is hosted and whether it will display in Zoho, and states the recommended follow-up action. The client can see exactly which notes carry image links and which need attention, with no surprises after go-live.
Zoho can take only 4 GB at a time, so the toolkit packages everything into numbered buckets capped at 4 GB each (the cap is a parameter, set to 4,096 MB). Each bucket holds one Attachments.zip and a matching Attachment.csv listing only the files inside that ZIP, so every chunk can be imported on its own. Files are never split across archives, and files of 100 MB or more go into their own large-file buckets. With more than 20 lakh records, this meant the migration ran as a sequence of manageable, independently verifiable import batches instead of one unmanageable transfer.
When Zoho flags files that failed to import, an error file can be fed straight back into a dedicated re-extraction tool. It checks the input before touching Salesforce, re-fetches only the affected files with all their current parents, and produces a status for each file. A fetch-sample report documents every step, including queries, downloads and checksums, without exposing credentials.
Complete, Traceable Coverage
Classic attachments, modern Files, Chatter uploads and Notes are all covered by one toolkit with one operating model. Nothing is silently dropped: every file is either delivered or listed with the reason it was not.
20 Lakh+ Records Migrated in Import-Sized Chunks
More than 20 lakh (2 million) records were extracted and delivered as a series of 4 GB packages that respect Zoho's import limit, each self-contained and each reconciled before it was sent.
Output Zoho Can Ingest First Time
Folder names, CSV headers, ZIP size limits and compression method are all aligned to Zoho's import specification and enforced by validation, which removes a common source of rejected batches.
Confidence Before Cut-over
Dry runs, preflight API and disk checks, a built-in count check and per-run reports mean the team can judge readiness from evidence rather than assumption. Keeping images as direct links, backed by an image register for Notes, kept the transfer light and turned an unknown risk into a managed list.
Lower Operational Risk
Resumable downloads, clean stops at API and disk limits, and targeted re-extraction mean an interruption or a rejected file becomes a short, scoped task instead of a restart.
Flexible by Design
Decisions on naming, scope, user-linked files and date logic were handled through configuration and recorded in a decision log, so the customer's choices could change without code changes and every run records the rules it ran under.
Moving off a CRM is only as successful as the history that travels with it. Files and notes are where much of the commercial context lives, and they are the hardest data in Salesforce to extract faithfully. A few themes stand out for any organization planning a similar move:
Know the data model, not just the export button. Salesforce stores content across multiple generations of objects with real constraints on how they can be queried. Experience with those internals is what makes a complete extraction possible.
Scale demands engineering. Large attachment volumes call for batching, resumability and limit-awareness, not manual downloads.
Auditability builds trust. Stakeholders signing off a migration need proof of completeness. Reconciled counts and run reports provide it.
The target system sets the rules. A good extraction is shaped by what the destination can import, which is why close coordination between source and target teams matters.
Whether you are leaving Salesforce, consolidating orgs or planning a large-scale Salesforce data migration, GetOnCRM brings the platform depth and the tooling to protect your data on the way through. Talk to our team about your migration.
We are Salesforce Select Partner having 100% salesforce certified team.
Copyright © 2026 GetOnCRM Solutions PVT. LTD. All Rights Reserved.
