Skip to content

Validate a Salesforce data migration

Compare the numbers before go-live. For every object, the record count in the source has to match the count in Salesforce, and every gap needs a reason. I map, clean, and deduplicate your data, load it with the Bulk API, and reconcile the counts before anyone switches over.

You work directly with me, Kiryl Kazlovich, a Solution & Technical Architect who can write code. Based in Grünwald near Munich. More than 12 years on Salesforce, for companies directly and inside Salesforce partner teams.

From process to handover

Process before code

I start with the business process, who owns which data, and how changes reach production.

Standard before custom

Standard features and Flow first. Custom Apex only where it is worth the upkeep.

Documented at handover

I write the decisions down, test the critical paths, and leave documentation your support team can actually use.

How I migrate and check

  • A mapping you sign off

    Each source field gets a Salesforce field, or a note why it stays behind. You see the mapping before the first load.

  • Duplicates found before the load

    I match accounts and contacts by name, address, and email, typos included, and you decide which record stays.

  • Loaded with the Bulk API

    Upserts with external IDs, so a second run updates the records of the first instead of creating copies.

  • A trial run in a sandbox

    The full load runs in a sandbox first, so errors show up before they reach production.

  • Counts reconciled

    For each object I compare the record count in the source with the count in Salesforce and list every failed record with its error.

  • Relationships checked

    Contacts sit on the right account and orders on the right customer. Lookups that point nowhere go on the list too.

Tell me about your project

Book a call, or send a few lines about your system and what should change.

A 30-minute call, in English or German.