Ask how to fix a checkout and someone will say reduce the fields. The advice is not wrong, but it is measured wrong. Stores count boxes on the form, celebrate deleting two, and watch conversion stay flat, because the shopper's effort was never in the box count. It is in what she has to recall, decide, type, and correct along the way.
A checkout with fourteen fields where ten arrive prefilled or autocompleted is an easier checkout than one with eight fields that demand typing, thinking, and a trip to another room for a card. Counting effort instead of boxes changes which fields get cut, and the cuts that result actually move conversion.
Count information produced, not boxes rendered
Walk the checkout as a shopper and tally every unit of information she must generate: each typed string, each decision between options, each correction after an error. Autofilled fields cost nearly zero. A decision with a clear default costs little. Typing a full address on a phone keyboard costs more than three dropdowns combined.
This tally exposes the expensive fields that box counts hide. The phone number field is one box and a frequent abandonment point, because shoppers must decide which number, whether it will be marketed to, and why a download needs it. The country dropdown is one box and nearly free. Equal in the count, opposite in cost.
The fields that earn their place
Some fields are the transaction: payment details, a delivery destination, a way to send the receipt. These earn their effort and shoppers accept them. The fight is over the rest: middle names, address line two presented as required, company fields in consumer checkout, date of birth for ordinary goods, and the account password demanded before the order exists.
A useful test per field: what breaks if this is optional, deferred until after the order, or collected on the confirmation page instead? For most suspect fields the honest answer is nothing breaks. Marketing wants it, or the platform default included it, and the shopper has been paying for both with abandonment.
Errors multiply field cost
A field's true cost includes its error rate and the price of fixing mistakes. Strict formatting that rejects valid phone numbers, address validators that refuse real addresses, and card fields that clear on a single wrong digit turn one field into three attempts. Lenient parsing, inline validation that helps instead of scolding, and never clearing entered data on error cut the effective field count without removing a question.
Measure fields by abandonment adjacency: which field was last touched before exits. Stores often find one validator, not the form's length, sitting behind a disproportionate share of lost orders. Fixing that field outperforms a redesign.
Prefill is the invisible reduction
Returning shoppers should meet their own data: address book entries, saved payment, remembered preferences. Every prefilled field is a field removed from the effort tally while remaining on the form. Identity from the payment step, where wallets already know the shopper, can collapse the address section to a confirmation.
The conversion literature's famous field reductions usually smuggle this in: the shorter form also remembered more. Stores that cannot cut questions can still cut effort, by remembering everything they are allowed to remember and asking the shopper only to confirm. Confirmation is the cheapest field there is.
Related reading
These earlier guides go deeper on connected questions: Checkout Page Speed: The Seconds That Cost Completed Orders, Duties and Taxes at Checkout: Surprise Fees Are a Conversion Problem You Can Fix, and Stock Messages at Checkout: Honest Scarcity Versus Manufactured Urgency.
Frequently asked questions
What is the ideal number of checkout fields?
There is no universal number. The right measure is effort: information the shopper must produce. A longer form that is mostly prefilled converts better than a short form that demands typing and decisions.
Which fields should be cut first?
Ones the store can infer, defer, or skip: required address line two, company fields for consumers, passwords before purchase, and any field whose honest removal breaks nothing.
Why do validation errors matter more than field count?
Because a strict validator can turn one field into several attempts and a lost order. Abandonment often clusters at a single bad field rather than spreading across a long form.
How can a store cut effort without cutting questions?
Prefill and remember. Autofill friendly markup, saved addresses and payment for returning shoppers, and wallet identity turn production into confirmation.