Donations and money

What each CSV column means

Eight columns. Two of them can be empty on purpose, and this explains when and why.

Column	What it is
Date	The day the donation belongs to, in your organization’s time zone.
Amount	The gift. This is the column that sums to what you raised.
Donor covered	The processing fee this donor chose to absorb. Never yours, never counted as raised.
Charged	What actually hit the card. Amount plus Donor covered.
Stripe fee	What Stripe took for this donation. Empty when it has not settled yet.
Net to bank	Charged minus the Stripe fee. What this donation put into your account. Empty when the fee is not known yet.
Status	completed or refunded. Nothing else appears in the file.
Athlete	Who it was credited to, or General Donation if nobody.
Stripe payment ID	The payment in your organization’s Stripe account. Empty until a gift finishes settling.
Donation ID	Our own reference for the gift. Always filled in.

The last two columns are for your treasurer

They are the only columns in the file meant for a machine rather than a person, which is why they sit at the end where they stay out of your way.

Your organization can export its own payments from Stripe. Stripe payment ID is the first column of that export, so the two files line up row for row. Before this existed, matching them meant going by date and amount, which stops working the first time two people give the same amount on the same day.

Why a blank is not a zero

Stripe fees settle a little after a charge does. Until a fee is known, Stripe fee and Net to bank are left empty rather than filled with 0. A zero there would read as "Stripe processed this one for free" and would overstate your deposit.

So if you sum Net to bank and it comes up short of your Stripe deposit, look for the empty rows. Those are the ones still settling, and they will fill in.

Reconciling against your bank

Your bank deposit matches the sum of Net to bank, for the donations in that payout.
A payout usually covers a couple of days of donations, so it will rarely match a calendar month exactly.
Refunded rows show what left again. They are in the file so a number never changes silently between two exports.