The Filing And The Audit Counted Different Entities
A regulatory report counts customers, facilities or counterparties. Each source system has its own version of each one, under slightly different names. The report counts one set; the auditor rebuilds it from another source and counts a different one. Explaining the gap takes weeks.
Three facilities in the filing. The regulator’s list has one.
Reports Are Only As Consistent As Their Entities
Regulatory reports are counts and totals over entities: customers, facilities, counterparties, products. Every source system holds its own version of those entities, with its own spelling, its own address format and its own key.
When a report is built, someone decides which records are the same entity, usually in a spreadsheet or a one-off script. The next report, or the auditor, makes those decisions again and gets a different answer. Totals do not tie out, and the reasons are buried in manual work nobody can reproduce.
The cost is not only the reconciliation effort. It is not being able to show, line by line, why a number is what it is.
Totals Don’t Tie Out
The filing, the internal report and the audit each count a different set of entities.
Manual, Unrepeatable Matching
Which records are the same entity is decided by hand, differently each time.
Hard To Explain
Without recorded evidence, nobody can show why two records were treated as one.
Report On Resolved Entities
Resolve entities once, in a repeatable process, and give each one a stable ID with every source record listed behind it. Build every filing, internal report and audit extract on those resolved entities.
Keep the evidence for each match. When a reviewer asks why two records were combined, the answer is recorded, and rerunning the process gives the same entities again.
The filing and the audit both count this one entity, and can trace it back to every source record.
One Set Of Entities
Every report counts the same resolved entities.
Stable IDs Over Time
An entity keeps its ID from one reporting period to the next.
Evidence On Record
Each match can be traced back to its source records and the reason for it.
The Identity Layer Under Privacy
Zingg runs as a workload inside your data platform and gives every entity one persistent ZINGG_ID. Everything above it reads the same resolved entities.
The same layer powersCustomer Analytics & MarketingFinancial Crime & RiskSupply Chain & ProcurementProduct 360Master Data ManagementAI & Data Foundations
Numbers You Can Trace Back
Persistent ZINGG_ID
Each customer, facility or counterparty keeps one ZINGG_ID across periods.
Explainable Matches
See the evidence behind every match when a reviewer asks. Explainability
Compare Before You Change
Check which entities a model or rule change would move before it reaches a filing. Model diff
Learned Matching Plus Rules
Fuzzy matching trained on your labels, combined with exact rules on registration numbers and identifiers.
Compared to the previous approach, it’s obviously much better — probably more than 90% better.
Regulatory Reporting Questions
Why don’t regulatory totals tie out across reports?
Each report decides which source records are the same entity in its own way. Different decisions produce different counts, even from the same data.
Does Zingg produce regulatory filings?
No. Zingg resolves the entities your reports count and records the evidence for each match. Your reporting tools build the filings on top.
Can we see why two records were combined?
Yes. Explainability shows the evidence behind each match, and every resolved entity lists its source records.
Will a model change move last quarter’s numbers?
ZINGG_IDs persist, and model diff shows which records would change clusters before a new model is used.
Build Every Report On The Same Entities
Resolve entities inside your own data platform and trace every number back to its records.