Analytics and Tracking

begin_checkout

Also called Checkout start event

The GA4 ecommerce event fired when a shopper commits to buying and enters the checkout process.

Quick facts: begin_checkout

Category
Analytics and Tracking
Also called
Checkout start event
Level
Beginner
Affects
Funnel reporting, checkout diagnosis, remarketing audiences, abandonment emails
Where to see it
GA4 DebugView, GA4 funnel exploration, Google Tag Manager Preview
In this article4
  1. How begin_checkout works
  2. Why begin_checkout matters
  3. Common mistakes with begin_checkout
  4. How to act on it

How begin_checkout works

This event marks the moment a shopper stops arranging a basket and starts trying to buy it — the click on “Checkout”, or the first checkout screen loading. It carries the whole basket: every item with its identifier, name, price and quantity, the total value of the goods and the currency, and a coupon code if one has already been applied.

It should fire once per checkout attempt. If somebody backs out at the delivery step, returns to the basket and starts again, that is a second attempt and a second event. What it must not do is fire again on every screen of a multi-step checkout; the later steps have their own events for that.

Why begin_checkout matters

It is the anchor of the ecommerce funnel. Everything before it is interest; everything after it is a purchase in progress. The gap between checkout starts and completed orders is the single most valuable number a shop owner can look at, because it is the one place where lost revenue is both measurable and usually fixable without spending more on advertising.

It is also the sharpest audience you can build short of buyers themselves. Someone who reached the checkout and did not finish had a reason, and that reason is often practical rather than a change of heart — a payment method that failed, a delivery estimate that did not suit, a form that would not submit on a phone. In Nepal that last one matters more than most people allow for, because so much of the traffic arrives on modest handsets over patchy connections.

Common mistakes with begin_checkout

Firing it on every checkout step is the classic error, and it makes the funnel look as though checkout starts outnumber baskets. On a hosted platform, the opposite mistake is common: the checkout sits on a different domain, the tag is not present there, and the event never fires at all. Test the real checkout, not the theme preview.

Sending it without items is the third. The value alone tells you money was at stake but not which products were, so you cannot tell whether the abandonment is spread evenly or concentrated on a few lines. Keep the item identifiers identical to the ones used at the basket view and on the order confirmation, or the same product will appear as several in your reports.

How to act on it

Start by measuring the drop from checkout start to purchase, then watch it separately on mobile and on desktop. A gap that only appears on phones is a layout or a form problem, not a pricing one, and it is usually cheap to fix.

Then work through the checkout itself as a customer would, on a real phone: how many fields, whether an account is forced, whether the delivery charge and the payment options are visible before the last screen. Fix what you find, and use the checkout-starters audience for remarketing and for an abandonment email while you do. If the event is not firing reliably in the first place, that is the job to do first — an independent tracking audit will tell you quickly whether the numbers deserve any trust.

Do and do not

Do

  • Fire it once per checkout attempt, with the full basket
  • Test it on the live checkout, including a hosted domain
  • Compare checkout starts to purchases by device

Do not

  • Fire it again on every step of the checkout
  • Send the event without item identifiers
  • Assume abandonment is always about price

Questions people ask about this

When exactly should begin_checkout fire?

At the point the shopper commits to buying: the click on the checkout button, or the load of the first checkout screen, whichever you can catch reliably. Fire it once per attempt and send the full basket with it. Do not fire it again on the delivery or payment screens — those steps have their own events.

Why does my checkout start count look higher than my basket views?

Almost always because the event is firing on more than one checkout screen, so a single shopper is counted several times. Check in GA4 DebugView by walking through a real checkout and watching how often the event appears. The other possibility is a shortcut button that skips the basket entirely and goes straight to checkout.

Is a big drop between checkout start and purchase normal?

Some drop is normal everywhere, because people compare, get interrupted or change their mind at the payment screen. What is not normal is a drop that differs sharply between devices or payment methods. Segment the gap by device and by payment option before assuming it is a pricing problem, and test the checkout on a real phone.

Related terms

Found this useful?

Share it, or ask an AI to summarise it

Back to the glossary

Knowing the term is the easy part

Applying it to your own site and budget is the work. Book a call and I will tell you what actually applies to you.