Public Sector Equality Report De-Identification with anonym.plus

Strip identifiers from a PSED monitoring return before a draft is reviewed.

In simple terms, PII redaction is the on-device process of finding and masking personally identifiable information in a document before it is shared.

This de-identification removes personal identifiers from a monitoring return behind the public sector equality duty (PSED). Section 149 of the Equality Act 2010 sets three aims for a listed public authority. It must eliminate discrimination, advance opportunity, and foster good relations. Progress is measured across the nine protected characteristics s.4 lists. anonym.plus marks names and IDs on your device, so the counts stay usable while named staff are shielded.

When this applies

The Specific Duties and Public Authorities Regs 2017 require the information behind s.149 to be published each year. Objectives must be set at least every four years. The working return is a different animal. It holds one row per employee, tagged with race, religion or belief, disability, or sexual orientation. Those tags are special category data under UK GDPR Art. 9(1). The condition most bodies rely on is DPA 2018 Sch. 1 Part 2 para 8.

How anonym.plus handles it

  1. Open the document in anonym.plus on your device.
  2. The tool flags named staff rows, the compiler, and contacts.
  3. Local OCR reads a scanned analysis page.
  4. Keep the characteristic counts and category codes.
  5. Swap or black out the confirmed identifiers.
  6. Save the clean draft locally.

What you need to provide

PII entity types detected

Categoryanonym.plus entity typeExample
NamesPERSONstaff row → [WORKER]
IdentifiersNATIONAL_IDstaff no. 41203 → [STAFF_ID]
NRPNRPcharacteristic group → [GROUP]
ContactEMAIL_ADDRESSwork email → [EMAIL]
DatesDATE_TIMEperiod 2026 → [YEAR]
OrganisationORGANIZATIONdepartment → [DEPT]

Compliance achieved

Anonymise public sector equality reports offline — see plans & start free →

Limitations & cautions

The return relies on aggregate counts. A small characteristic group can still re-identify a person. Where one or two staff fill a group, the cell names them. Band or suppress small groups before review. What you publish is a separate decision from what the draft holds.

Frequently asked questions

Does an aggregate monitoring return hide individuals?

Not in tiny groups. A group of one or two re-identifies a person. ICO anonymisation guidance expects small cells to be suppressed or banded before publication.

What condition covers this monitoring data?

Most public authorities rely on DPA 2018 Sch. 1 Part 2 para 8, on equality of opportunity or treatment. It requires an appropriate policy document under Sch. 1 Part 4.

Is the file uploaded?

No. The app is fully offline, so the draft stays on your machine.