Help / Importing Transactions

Column mapping in detail

Step 3 of the import wizard is where you tell LedgerBear which CSV column is which. This page explains every field you can map and the three different ways of expressing amounts.

The mappable fields

Above each column in the preview table is a dropdown. Each column can be mapped to one of the following, or left unmapped:

  • Date — required. LedgerBear parses most common formats (MM/DD/YYYY, YYYY-MM-DD, Jan 5 2026, etc.).
  • Payee — optional. The payee text, typically a short label like Trader Joe's.
  • Memo — optional. Free-form description text. If your bank has both a "short description" and a "long description" column, Memo is the usual target for the long one.
  • Category — optional. If your CSV has a column with a category name, LedgerBear will try to match the text to one of your existing categories. It does a case-insensitive lookup by the subcategory Name. Unmatched values are left uncategorized.
  • An amount field — required, but there are three ways to map one (see below).

The three amount modes

Banks don't agree on how to represent amounts. LedgerBear supports three layouts:

Amount mode (one column)

The simplest layout: a single Amount column with positive values for money in and negative values for money out. Pick Amount on the right column.

Most modern bank exports work this way.

Debit/Credit mode (two columns)

Some banks split outflows and inflows into two separate columns — one that's populated when it's a debit (money out) and another when it's a credit (money in). Every row uses exactly one of the two.

Map Debit and Credit on the appropriate columns. LedgerBear fills the amount as -Debit or +Credit based on which column has a value.

Amount + Type mode (two columns, different layout)

A third common layout puts the raw amount in one column and a separate column that says whether the row is a debit or credit (values like DEBIT, CREDIT, debit, credit).

Map Amount on the amount column and Type on the type column. LedgerBear negates the amount when the type says debit (case-insensitive) and leaves it positive otherwise.

Auto-detection on first import

The first time you import to a given account, LedgerBear tries to guess the mappings:

  • Columns whose names contain "date" are scored against "post" / "posted" (preferring posted-date columns when both exist).
  • Columns whose names contain "payee", "memo", "description", or "category" get mapped accordingly — "memo" wins over "description" if both exist.
  • For amounts, if there's a column named "Amount" plus a "Type" column, the wizard picks amount+type mode. If there's a single "Amount" column, it picks amount mode. If there are separate "Debit" and "Credit" columns, it picks debit/credit mode.

If a guess is wrong, override it with the dropdowns.

Remembered mappings

After a successful import, LedgerBear saves your column mappings on the account. The next time you import into the same account, the wizard applies them automatically. If the bank changes its column headers (it happens), you may need to re-map — the wizard will fall back to heuristics for anything it can't match.

Unmapped columns

Any column you don't map is ignored. If your CSV has extra columns like "Type of Expense" or "Account Number" that you don't care about, just leave their dropdowns blank. They won't break anything.