Use Case: Data Engineering

Three vendors. Three schemas. One canonical record.

Vendor A uses snake_case with USD amounts. Vendor B uses camelCase with EUR amounts. Vendor C uses inconsistent date formats. One normalization config handles all three, with schema version tracking when any vendor changes their format.

Connect your first feed View docs

The multi-vendor problem

When vendors have no shared schema standard, your data engineering team absorbs all the reconciliation work. Here is what three vendor feeds actually look like before normalization.

Vendor A (snake_case USD)

company_name,gross_revenue_usd,last_updated,region
"Acme Corp",142500.00,03/14/2026,APAC

Vendor B (camelCase EUR)

CompanyTitle,TotalRevEUR,UpdatedAt,Territory
"ACME Corp.",131280.00,2026-03-14,Asia Pacific

Vendor C (mixed)

org_name,revenue,update_ts,area_code
"Acme Corporation",null,14-03-26,AP

Without normalization: the join on entity name fails to match across vendors, the revenue field is in different currencies with different type encodings, and the date field has three different format conventions. Downstream aggregations either silently produce wrong results or fail entirely.

HyperNorm's normalization config

One YAML config defines how each vendor's fields map to your canonical schema. HyperNorm applies it on every run. When Vendor B changes their schema in week six, schema versioning detects the drift and alerts you before the job fails silently.

version: 1
canonical_schema: canonical_v1.yml

sources:
  vendor_a:
    fields:
      company_name -> canonical_entity
      gross_revenue_usd -> revenue_usd
      last_updated -> updated_at  # format: MM/DD/YYYY
      region -> region_code

  vendor_b:
    fields:
      CompanyTitle -> canonical_entity
      TotalRevEUR -> revenue_usd  # currency: EUR -> USD
      UpdatedAt -> updated_at   # format: ISO8601
      Territory -> region_code

  vendor_c:
    fields:
      org_name -> canonical_entity
      revenue -> revenue_usd  # null_strategy: inherit_from_vendor_a
      update_ts -> updated_at  # format: DD-MM-YY
      area_code -> region_code

Before and after HyperNorm

Before HyperNorm

3 days per sprint reconciling schemas manually
Revenue joins failed silently when Vendor B switched to EUR
Date format changes broke downstream aggregations on first run
Entity duplicates split revenue attribution across phantom records
No audit trail for schema version changes

After HyperNorm

20-minute config review per sprint
Currency conversion applied automatically at ingestion
Schema drift detected before the job runs silently wrong
0 unit mismatch errors in downstream model since integration
Full schema version history with diff tracking

Connect your first vendor feed.

Start with the quickstart guide, or connect directly if you already know your schema.

Start Free View quickstart