# Custom fields

> Add your own attributes to any record type — currency, selects, checkboxes, multi-selects and more.

_Collection: Companies, Contacts & Candidates (records) — Kepler Help Center. Canonical: https://keplercrm.com/support/articles/custom-fields/_

Every desk tracks something the standard fields don't cover — day rate, security clearance, right to work, languages. Custom fields let you add those attributes to **candidates, contacts, companies, jobs and placements**, and they behave exactly like built-in fields everywhere: tables, filters, kanban, imports and AI autofill.

## Field types

| Type | Use for |
| --- | --- |
| Text | Free-form values — references, IDs, notes-in-a-field. |
| Number / Currency | Day rates, salary bands, fee values. |
| Select / Status | One value from a fixed list — e.g. security clearance level. Select and status fields can power kanban columns. |
| Multi-select | Several values — languages, tech stack, sectors. |
| Checkbox | Yes/no flags — right to work, GDPR consent received. |
| Date | Visa expiry, available-from. |
| Timestamp | Date and time together. |
| Rating | Scored values. |
| Record / Relationship / User | Links to other records or to workspace users. |
| Location / Phone / Email / URL | Structured contact and web values. |

## Creating a custom field

  - Open the record list for the type you want to extend (e.g. **Candidates**).

  - Open the column controls and choose **Create New Column…**

  - Pick a name and type; for selects, define the options.

  - Save — the field is immediately available on every record of that type.

## Letting AI fill a column

The field dialog ends with **How this column gets filled**. A column has exactly one fill mode:

  - **Manually** — you type it. The default.

  - **From this record** — a value computed from other fields. Write a prompt and reference other columns by typing `{` to pick one (for example *"What industry is {Current employer} in?"*). Choose whether Kepler fills the column **if empty**, only **suggests** a value, or **always** recomputes it — *always* is the only unattended writer, and it locks the cell against manual edits.

  - **From source content** — a fact read from documents and conversations. Tick the sources it may read (**Primary CV** on candidates, **Emails and meetings** on any record type), then choose whether a newer source wins (**latest wins**) or values merge (**accumulate**).

Values that need a decision arrive in [Notifications](/support/articles/ai-suggestions/) for approval rather than being written silently. Which communication sources AI may read at all is a workspace-wide setting under **Settings → AI**, not a per-column one.

> **NOTE:** Built-in fields carry no per-column AI setup — their eligibility for AI updates is decided by Kepler, and the dialog tells you when a built-in field is not eligible.

## Where custom fields show up

  - **Record pages** — editable on the Overview tab.

  - **Tables** — add them as columns and edit inline.

  - **Filters & sorting** — filter and sort any view by them.

  - **Kanban** — a select/status custom field can be the grouping for a kanban view.

  - **Imports** — map CSV columns into them.

  - **AI autofill** — the CV parser and email analysis can populate them when the source material contains the answer.

> **WARN:** Renaming a select option doesn't retag old records automatically — prefer adding a new option and migrating with a filtered bulk edit if you need to restructure a list mid-flight.