Skip to content
woodfrog
Data Foundations
BFSI · DPDP Act 2023Use case, no client result yet

Keep personal data in India and out of the wrong hands.

Customer names, mobile numbers and PAN details sit in your core system, your CRM, your loan files and a dozen spreadsheet exports. Nobody can say where all of it lives or who has looked at it this month.

Every field, classed and maskedSchematic
Masked by default. Unmasked only for a named role, and every look is logged.
3.2×Target return in 12 months, on your own numbers
Week 1Fixed-fee audit
6 weeksTo the first live result
3Senior people on the work
TypeUse case: what we build, not a client story
ForBanks, NBFCs, insurers and fintechs that hold customer personal data
EngagementWeek 1 audit, then build in about six weeks, fixed scope, fixed fee
Runs onAntvia lakehouse in your own cloud account, in an Indian region you choose
Before and after

From scattered copies to one controlled record.

Before
  • Personal data copied into exports nobody tracks
  • Everyone with report access sees full names and numbers
  • No list of where personal data lives
  • No record of who looked at what
After
  • Every personal field found and classed
  • Masked by default, unmasked only for a named role
  • Data kept in an Indian cloud region you choose
  • Every look at personal data logged, with a name against it
The problem

Personal data leaks sideways, not through the front door.

The risk is rarely a hacker. It is a report that shows full mobile numbers to fifty people, an extract emailed to a vendor, a test copy of production left open. India's Digital Personal Data Protection Act 2023 puts duties on any business that holds personal data. You cannot meet a duty for data you cannot find.

Server racks holding customer records
Customer data, spread across more systems than anyone listed
What was broken
  • Nobody holds a list of where personal data lives, field by field.
  • Reports and dashboards show raw names, numbers and IDs to anyone with access.
  • Copies multiply: exports, test databases, vendor extracts, laptops.
  • Some data sits outside India, on services nobody chose on purpose.
  • When someone asks who saw a record, there is no log to check.
You might be here if
  • Your answer to "where is our customer data?" is a list of system names.
  • A new analyst gets full customer details on day one.
  • Test environments are filled with a copy of production.
  • Your legal team asked a data question and is still waiting.

Why it mattered. Governance added after the build means going back through every pipeline. Built in from the first pipeline, it costs far less and stays in place.

The approach

Find it, class it, then lock it by default.

We start by finding every personal field, not by writing a policy. We scan your sources, sample the columns and list what holds personal data, then your team confirms each one. Your legal and compliance people decide what the Act means for you. We build the controls that make those decisions hold, in the pipeline, not in a document.

A hand holding a padlock
Locked by default, opened on purpose
Where personal data sits
Core system and CRM
Loan and policy files
Exports and spreadsheets
woodfrog method
Find every fieldScan, sample, confirm
Class and maskPersonal, sensitive, business
Grant by role, log every lookOne owner per dataset
What changes
Data stays in India
Masked unless needed
A log you can show
Personal data under control
Deliberately not doneTelling you that you are compliant. That is a legal judgement, and it is yours to make.
Your team confirms every class. Claude helps scan and draft.schematic
The solution

Governance in the pipeline, not in a binder.

What we build
  • A personal data map: every source, table and field that holds personal data, classed as personal, sensitive or business, and confirmed by your team.
  • Masking and tokenising on the way in, so names, numbers and IDs are hidden in the cleaned and reporting layers by default.
  • Role-based access: unmasked data only for named roles, granted and removed in one place.
  • An access log: who opened which personal data, when, and through which report or query.
  • Storage and processing in an Indian cloud region you choose, on open formats in your own account.
  • Retention rules per dataset, so personal data past its purpose is flagged for your team to delete.
It runs on Antvia, our lakehouse with bronze, silver and gold layers and built-in PII masking, inside your own cloud account. Everything is documented and stays with your team. We make no claim that any of this makes you compliant. It gives your compliance team the facts and the controls to decide.
How it would runWeek 1 ends with the personal data map, so you know the size of the job before the build starts. Operate is open-ended and drawn at one week only to fit.
The impact

What we would measure.

Who can see personal data, layer by layerFewer people can read it the further it travels.
Schematic, no values
Width shows how many people can open the layer. Darker means less personal data inside it.
What we would measureNo client values yet
Personal data fields found and classedAgainst the list your team held before
People who can see unmasked personal dataBefore and after, by role
Copies of personal data outside the platformExports, test copies, vendor extracts
Time to answer "who saw this record?"From days of asking to one query
Return on the engagementTarget: at least 3.2× within 12 months, measured on your own numbers
Know where your customer data is, and who sees it.A short call, or record a message. A person replies within one business day.
Contact usCertified Services Partner, Claude Partner Network