Learn / Playbooks / Bureau data and scoring checks

Bureau data and scoring checks

A new bureau feed arrives next week, and you must confirm the data and the score are fit to use. Clean bureau data, build usable variables and check the score still works.

8 steps · 10 tools (3 free) · for risk analysts at banks and NBFCs

Tools you will use

This playbook is free to read. The prompt for each step is inside its tool. See the Pro tools.

Schema profile builderFree

Definition libraryFree

Bureau data scrub and quality checksFree

Bureau variable derivationBureau and applicant analytics

Bureau score distribution and stabilityBureau and applicant analytics

Stability testsModel validation

Thin-file and new-to-credit analysisBureau and applicant analytics

Income, FOIR and obligations checksBureau and applicant analytics

Third-party or challenger model testingModel validation

Bureau reporting accuracyNPA, SMA and provisioning

Check these first

Each tool asks the same questions and gives you a prompt to build whatever is missing.

Bureau extract with account and enquiry level fields?

Have itOnly scoreDon't have it

Bureau score at application and at a later date?

BothOneNeither

Reference period fixed for comparing score stability?

YesNo

The steps, in order

Each step names the tool to open.

Set up

Tell the tools your table names

Bureau tables, keys and dates.

FreeSchema profile builderFreeDefinition library

Clean

Scrub the bureau data

Duplicates, impossible dates, negative balances, stale pulls.

FreeBureau data scrub and quality checks

Derive bureau variables

Enquiries in 6 months, worst DPD, utilisation and more, from raw fields.

Bureau variable derivation

Analyse

Score distribution and stability

Has the score mix shifted since you last looked?

Bureau score distribution and stabilityStability tests

Thin-file and new-to-credit customers

How many have no usable history, and how they perform.

Thin-file and new-to-credit analysis

Income and obligation checks

FOIR and obligations from bureau against stated income.

Income, FOIR and obligations checks

Test the bureau score on your own customers

Does the bureau score rank your good and bad accounts apart, or only theirs?

Third-party or challenger model testing

Report

Check what you report back to bureaus

Accuracy of your own submissions.

Bureau reporting accuracy

Check before you send

Row counts match between the raw extract and the cleaned table, with each drop explained.
No variable uses information from after the application date.
Score stability is measured against a fixed reference period, and you name it.
Thin-file customers are reported separately, not hidden inside the average.

Warning signs

Population stability moves sharply but the bad rate does not: a data feed change, not a real shift.

Thin-file share rises after a channel launch: check who is sourcing them.

What you hand over

A cleaned bureau table, a derived-variable sheet, and a one-page stability and thin-file note.

In your profile

Once you are signed in, your profile shows which steps you can run with the tools you own, and which tools you still need.

Other playbooks

Monthly portfolio review Delinquency tracking Underwriting policy change Scorecard build Model validation Early warning and leading indicators Collections strategy and agency performance Fraud and first-payment-default review Provisioning and ECL RBI reporting and asset classification Loss budget and stress testing Board and risk committee pack