Where guests drop off in online check-in

Online check-in saves the desk minutes per guest, but only for guests who finish it. Knowing where they stop is the whole job.

That is the whole economics of it. A form that half your arrivals abandon does not save you half the time. It costs you more, because now the desk has two processes to run: the normal one, and the one where they open a half completed record, work out what is missing, and ask for it again at the counter while the guest stands there.

So the question worth asking is not "do we have online check-in". It is "where exactly do people stop", and almost nobody can answer it.

The drop-off points we see

• Passport entry. This is the biggest one, and the least forgiving. Passport numbers are long, they mix letters and digits, and the field usually rejects a space, a hyphen or a lower case letter without saying which. One guest cancelled a booking outright after struggling to enter passport details. Make the field forgiving: strip spaces yourself, accept any case, and explain the format next to the box rather than after the error. Better still, let the guest photograph the passport instead of typing it. A phone camera is faster than a keyboard and more accurate than a tired traveller.
• The payment step. An unfamiliar payment page scares people, and it should. Guests have been trained to be suspicious of a payment screen that looks nothing like the hotel they booked. If your provider's page is unbranded, or in the wrong language, or asks for details in an order nobody expects, you will lose people who fully intended to pay. Use a page they recognise, with Apple Pay, Google Pay and the local wallets your guests actually use. Every tap you remove is a guest you keep, which is the longer argument in the payment field that quietly kills your digital check-in.
• Deposits. A card form that appears without warning reads as a charge. Explain the deposit before the form, not after: what it is for, how much, when it is released, and whether it is a hold or a payment. One sentence in plain language prevents most of the abandonment and most of the angry messages afterwards.
• Companions. Asking for every companion's passport, date of birth and address up front doubles or triples the form for a family. Ask for the lead guest first, complete the check-in, and collect the rest afterwards if you need it. A finished check-in with one guest on it is worth far more than an abandoned one with four.
• The link itself. Some guests never reach the form. The link arrived in an email they did not open, or it expired, or it opened in an in-app browser that broke the camera. If you can send it in the channel the guest already uses, do.

What helps

• Send the link 3 days before arrival, with a one line expectation: "this takes 2 minutes". People decide whether to start based on how long they think it will take, and they are usually guessing high.
• Pre-fill everything the property system already knows. Name, dates, room type, number of guests, phone number. A form that already knows who you are feels like a service. A form that asks for your booking reference feels like paperwork.
• Show the upsells after the required fields, not before. An early room upgrade offer turns a task into a sales page, and some guests close the tab. Once the required part is done, the same offer converts better, because the guest has already committed.
• Let the guest stop and resume. Save on every step. A check-in that has to be restarted from the top will not be restarted at all.
• If the guest stops, follow up once. One reminder, in the same chat, that answers questions rather than repeating the link. Most stalls are a question, not a refusal: is the deposit refundable, do I need my son's passport, can I do this at the desk instead. Answer it and the guest finishes in the same conversation.
• Keep the desk path open. Say clearly that check-in at the counter is still fine. Guests who feel cornered abandon; guests who feel offered a shortcut take it.

What to measure

Three numbers, per property, per month.

Started. Of the guests who received the link, how many opened the first screen. If this is low, the problem is the link, the timing or the channel, not the form.

Finished. Of those who started, how many completed. This is the number that determines whether the desk saves any time at all.

Where they stopped. The step, not just the count. This is the one most properties do not have, and it is the only one that tells you what to fix. If a third of your abandonment happens on one screen, that screen is your next week of work.

Track it for a month before you redesign anything. Properties that guess usually rewrite the welcome text, which was never the problem, and leave the passport field exactly as it was.

Then fix one thing and measure again. Online check-in improves in small increments: a forgiving field here, a clearer deposit sentence there. It rarely improves through a redesign.

One more habit worth borrowing from operations: treat each stall as a ticket with an owner rather than a statistic, the same way a guest request works better as a ticket than as a message in a group. And if your check-in has to pre-fill or update anything from another system, whether it can is decided by that system's API, which is the same argument as choosing a spa system for its API.

Takeaway

The form is a conversation. Ask less, explain the money, and let people finish on their phone in two minutes.