Email marketing for hotels: turning a booking into a returning guest

A hotel that relies heavily on OTA channels for its bookings can say something less comfortable in the same breath: for most of those guests, it has no reliable way of emailing them again once they've checked out. The booking confirmation and the receipt both came from the platform, not the hotel. And on a booking routed entirely through the messaging system of the OTA (online travel agency, the booking platform), the address sitting in the hotel's reservation record is frequently the platform's own masked, single-stay address rather than the guest's real inbox; the masked address can stop forwarding messages once the review window closes. The guest stayed, paid, and then disappeared from the hotel's own records the moment they checked out.

TL;DR: Email marketing for hotels isn't a monthly newsletter. It's three separate, trigger-based sequences built on a genuinely opted-in list: a pre-stay sequence (sent before arrival, built to earn a small upsell or seed a direct rebooking), a post-stay sequence (sent shortly after checkout, built to earn a review and plant a return offer), and a winback cadence (sent once a guest has gone quiet for a set stretch, built to bring them back while the stay still counts for something). None of it works without a real, consented email address, which is exactly what a booking made through an OTA rarely hands the hotel directly. This article covers what each sequence should actually contain, what genuine consent looks like against the performative version, the single most common way hotels waste a working list and conclude email doesn't work, and when a hotel isn't ready to run any of this yet.

This is written for the owner or general manager of an independent hotel of roughly fifty to two hundred rooms who already has some guest data sitting in a booking system or PMS and wants to know what to actually do with it. It's a working guide to the three sequences, not a setup tutorial for any particular piece of software.

The guest email an OTA booking rarely hands you

Every other channel a hotel uses to reach a past guest depends on the platform that guest is currently standing on. Paid search and social both require paying again for attention the hotel already earned once. The guest email is different: it is the one channel a hotel owns outright, sitting in its own system, reachable at close to zero marginal cost, unaffected by an algorithm change or a bidding war. That's the whole case for treating it as a retention asset rather than a marketing afterthought.

The catch is that a hotel only owns that asset on the bookings where it actually captured a real address with real consent, and that is disproportionately the direct channel. How to increase direct bookings is the piece for growing that channel itself; this one assumes some direct volume already exists and asks what a hotel does with the contact details that volume produces once the stay is over.

What actually counts as email marketing here, and what doesn't

Three sequences do the commercial work, each triggered by where a guest sits in their own journey rather than by the calendar:

• Pre-stay, sent in the days before arrival, aimed at a small upsell (a room upgrade, an early check-in, a dinner booking) and quietly seeding the idea of booking direct next time.
• Post-stay, sent a day or two after checkout, aimed at a genuine review request and a return offer while the stay is still fresh.
• Winback, sent once a guest has gone quiet past a defined point, aimed at getting them to book again before the hotel fades from memory.

A newsletter, the update on renovations, events or seasonal offers that goes to the whole list on a fixed monthly schedule, is a different thing. It has a place as brand maintenance, but it isn't a substitute for any of the three sequences above, because it ignores where each guest actually is.

When the problem isn't email at all

None of the above matters if the list underneath it barely exists. A hotel with no capture field on its own booking flow and no genuine opt-in prompt at check-in has a collection problem, not a sending problem, and no amount of sequence design fixes that. This is the first thing worth checking before designing anything, and it's the one hotels most often skip, because it feels like housekeeping rather than marketing. The fix is mechanical, not creative: add the field to the booking flow, add a real opt-in moment at check-in, and give staff a reason to ask for it that isn't "for the newsletter." Building sequences on top of a list that is mostly missing, stale or never actually consented is solving the wrong layer of the problem, and it's worth checking honestly before reading any further.

Consent done properly, not performatively

A guest email captured without a clear opt-in is a liability rather than a marketing asset. In practice, the consent worth having looks specific: a genuine choice at the point of capture rather than a pre-ticked box, a record of when and where that choice was made, and a working unsubscribe on every send. Mailbox providers police the consequences from the technical side regardless of what any regulator requires: Google's guidance for bulk senders gates inbox placement on the spam-complaint rate, kept below 0.3% in its Postmaster Tools, and is explicit that it does not track open rates at all (Google, "Email sender guidelines," accessed August 2026). A list built on weak consent generates complaints for the obvious reason that the people on it never really asked to be there, and a rising complaint rate drags down deliverability for every email the domain sends afterwards, including the transactional ones a guest actually wants.

The legal side varies by market and this isn't legal advice: broadly, UK and EU rules lean toward requiring a clear opt-in for marketing email rather than assuming consent from the booking itself, and the detail differs by country and by how the address was collected. A hotel operating across more than one market should check what applies in each one with its own adviser rather than copy a single consent flow everywhere (see the ICO's guidance on direct marketing for the UK starting point).

The pre-stay sequence: earning attention before arrival

The booking confirmation itself is not the place for an upsell; it's transactional, and guests read it for the details they actually need. A separate pre-stay message, sent closer to the arrival date, is where a room upgrade, a late checkout or a restaurant reservation genuinely earns a look, because the guest is already planning the stay and receptive to anything that makes it better. This is also the quiet moment to seed a return: a single line noting that booking direct next time gets the guest the same room at the hotel's own rate, planted well before the guest has any reason to compare prices, works harder than the same message attached to a hard sell.

The post-stay sequence: the review ask and the return offer

Sent a day or two after checkout, while the stay is still front of mind, this sequence has two honest jobs. The first is the review ask itself, timed early enough that the memory is fresh and specific rather than generic. The second is a return offer, something concrete enough to act on later rather than a vague "come back soon," ideally tied to a date range or a reason (a local event, a season the hotel does well) rather than left open-ended. Splitting these into two short messages, rather than cramming both into one, tends to perform better than either alone in the accounts we run, because a guest asked for a review and sold a return stay in the same paragraph reads the review request as insincere.

The winback cadence: guests who have gone quiet

A guest who hasn't returned within six to twelve months (the exact window depends on how often that hotel's typical guest travels) is a different case from one who stayed last month. The winback message works because it treats the silence as a signal rather than ignoring it: a genuine "it's been a while" tone, paired with a specific reason to come back, does more than another generic seasonal push sent to the same list as everyone else.

Illustrative arithmetic, not a real result: take a hypothetical hotel with 120 rooms that has collected 500 verified, opted-in guest emails over a quarter. Three hundred of those guests haven't rebooked within nine months. A winback send to that segment at even a modest 2% response rate produces six bookings; at an average stay of two nights, that is twelve room-nights sold at no acquisition cost beyond the time spent writing and sending the email. That arithmetic doesn't need a large list to matter; it needs a list that actually consented and a message that treats the guest as someone who has stayed before, not a stranger.

The failure mode: one newsletter, sent to everyone, monthly

In the accounts we run, the most common way a working list gets burned is the same generic monthly update sent to every address on it, regardless of whether that person stayed last week or three years ago, then concluding from flat open rates that email doesn't work. It isn't that the channel failed; every guest received a message aimed at nobody in particular. A guest three days from arrival and a guest who checked out eighteen months ago have nothing in common except being on the same list, and a single monthly send treats them identically. Splitting the list by where each guest actually sits, even crudely, before writing anything, changes what "email doesn't work" usually turns out to mean.

Where the data actually comes from

All three sequences depend on data sitting somewhere the hotel can actually query it: who stayed, when, how often, and whether they consented to be contacted again. That's squarely PMS territory (property management system, the operational record of reservations and guest history), and a PMS that can't export clean guest data quietly caps everything above this section, however good the sequences themselves are. What a hotel PMS actually does covers what to check about that export before it becomes a problem nobody notices until the list needs rebuilding from scratch.

Fitting this into the rest of the marketing plan

Retention only earns its keep once the rest of the plan is pointed the right way; a strong winback cadence sending guests back to an OTA listing instead of the hotel's own site is solving half the problem while quietly funding the platform it's meant to reduce reliance on. Hotel marketing strategy covers where retention sits in the sequencing of everything else a hotel should be doing, alongside acquisition rather than as an afterthought once the rest is sorted.

If the list exists, the consent is genuine, and the sequences still aren't built, that's a reasonably small, mechanical project, and exactly the kind of thing a free audit puts a shape and an order on: whether the gap is the list, the sequences, or something further upstream.