Your guest has checked in, dropped their bags, and discovered they'd quite like a late check-out tomorrow. They'd pay £25 for it without blinking.

They won't get it. Not because you wouldn't sell it, but because there's no way for them to ask that doesn't involve getting dressed and going back down to reception. So the request never happens, and the £25 stays in their pocket.

This is the part of hotel upselling that most software doesn't touch. The pre-arrival market is mature and genuinely good — if you're not running upgrade offers before arrival, start there. But pre-arrival is the easier half of the problem, and it's the half where guests know least about what they actually want.

This guide covers what the main categories of hotel upselling solutions do well, where they stop, and what it takes to sell into the moment a guest is actually ready to buy.

What Upselling Is Worth At Your Size

Most articles on this topic quote a revenue figure with no working shown. Here's the arithmetic instead, so you can swap in your own numbers.

Take a 40-room property at 70% annual occupancy with an average stay of 1.8 nights:

• Room nights per month: 40 × 0.70 × 30 = 840

• Stays per month: 840 ÷ 1.8 = roughly 467

At a 5% upsell conversion rate and a £20 average upsell, that's £467 a month, or about £5,600 a year.

At 12% conversion and a £25 average, it's £1,400 a month, or about £16,800 a year.

The gap between those two lines is the entire argument. It isn't about having offers — both scenarios have offers. It's about how many chances you get to put the right one in front of the right guest, and whether the guest can act on it in the ten seconds they're thinking about it.

Run the same maths on your own room count and ADR before you talk to any vendor. It tells you what the problem is worth, which tells you what you should be willing to pay to fix it.

Upselling Vs Cross-Selling, Briefly

Upselling moves a guest up within what they've already bought: a superior room, a better view, a flexible check-in time.

Cross-selling adds something alongside it: breakfast, parking, a spa treatment, a dinner package, an airport transfer.

In practice, teams use "upselling" for both, and so does most of the software. What matters commercially is the same in either case — total revenue per stay goes up, and your fixed costs don't move.

The Four Moments, And Which Ones Are Actually Covered

Every upsell opportunity in a stay falls into one of four windows. The vendor landscape covers them very unevenly.

1. Booking confirmation, minutes after purchase

Well covered. This is where low-friction "complete your stay" add-ons work: breakfast, parking, cancellation flexibility. Guests are still in transaction mode. Keep it to one or two options — a long menu here reads as a checkout funnel and kills conversion.

2. Pre-arrival, two to three days out

The most contested window, and the best served. Guests are thinking about arrival times and plans, so room upgrades and early check-in convert well. Nearly every upsell platform on the market is built primarily for this moment. Set a clear deadline — "upgrade available until 18:00 the day before arrival" — so housekeeping can plan.

3. Check-in, desk or online

Covered, but constrained. Your priority at the desk is speed, so you get one offer, based on what's actually available: "We've got a quiet courtyard room for £15 extra a night — would that help?" One sentence, then stop talking.

4. In-stay, day one onwards

Barely covered, and it's where the intent is highest. This is when the guest has actual information: they know the room is a bit noisy, they know they're too tired to go out for dinner, they know tomorrow's train isn't until four. Every one of those is a purchase decision waiting for a channel.

The reason window four goes unsold isn't that hotels don't want the money. It's that the channel is broken. The in-room phone is a dead object to anyone under fifty. QR codes get scanned once and ignored. Nobody downloads an app for a two-night stay. And the front desk, at 7pm on a Friday, has neither the bandwidth nor the visibility to call thirty rooms and ask whether anyone fancies dinner.

Comparing Hotel Upselling Solutions By Category

Rather than ranking vendors, it's more useful to understand what each category is built to do — because that determines which of the four windows you're buying.

Pre-arrival upsell platforms

Examples: Oaky (Plusgrade), UpsellGuru, UpStay, GuestJoy.

Covers: Booking confirmation and pre-arrival. Automated offers by email, often with auction or bidding for upgrades.

Where it stops: Effectively over once the guest arrives. Strong ROI inside its window; not designed for in-stay.

Guest experience and messaging platforms

Examples: Duve, Canary, SiteMinder.

Covers: Pre-arrival, online check-in, front desk, plus in-stay via a guest app or web portal.

Where it stops: In-stay depends on the guest opening the app. Adoption is the constraint, not features.

Front desk performance and training

Examples: Frontline Performance Group (IN-Gauge).

Covers: Check-in conversion, staff coaching and incentives.

Where it stops: Only works where you have staff, at the desk, with time. Doesn't scale to in-room moments.

PMS-native upsell modules

Examples: Oracle Hospitality, Mews, Cloudbeds and similar.

Covers: Rate and room-type upgrades tied to live inventory.

Where it stops: Usually basic; rarely covers non-room ancillaries or guest-initiated requests.

In-room voice concierge

Examples: ButlerIQ.

Covers: In-stay, guest-initiated, with no app or download.

Where it stops: Newer category with a shorter track record than the categories above. Doesn't replace pre-arrival email upselling.

The honest reading: if you buy one thing, and you're not already doing pre-arrival offers, buy a pre-arrival platform. It's proven, the vendors are mature, and the payback is fast.

But recognise what you've bought. You've bought windows one and two. Windows three and four are still open, and window four is where the guest knows most about what they want.

What To Ask Vendors Before You Sign

PMS integration and real-time availability. Non-negotiable. Selling an upgrade you can't honour at 7pm on an arrival day costs you more than the upsell was worth.

Rules and targeting. Can you stop late check-out being offered on fully booked days? Can you restrict family add-ons to bookings with three or more guests? Can you set a floor so upgrades never go below a set price?

Operational control. Can your duty manager change a price, pause an offer, or shift the timing themselves — on a Tuesday afternoon, without raising a support ticket and waiting a week?

Reporting by product. You need upgrade revenue separated from breakfast separated from late check-out, or you can't kill the weak offers.

Channel delivery. If you use WhatsApp, SMS or email, can offers go out there without someone copying and pasting?

Payments. Pre-authorisation or payment links, and which of those fits your guest profile.

Pricing model. You'll see per-room, per-property, and revenue-share. For smaller properties revenue-share lowers early risk, but make them show you a worked example: "If we do £2,000 a month in upsells, what's your cut and what is our net?" Confirm contract length and whether you can export your data if you leave.

Making The Offer Without Sounding Like A Salesperson

Guests don't object to upsells. They object to feeling worked on. The difference is almost entirely down to three things.

Use what you already know. You don't need a CRM. Your PMS already tells you party size, length of stay, arrival day, and any flagged occasion. A two-night couple's break is a natural late check-out. A one-night business stay wants early check-in and a quiet floor. That's enough personalisation to double your hit rate.

Show the incremental price. "Add £20 a night for a superior room," not the full restated stay value. Don't make the guest do arithmetic and don't bury the number.

Give your team one question, one offer, one exit.

• Ask: "How was the journey in today?" This tells you their mood before you offer anything.

• Offer: "If you'd like a bit more space, we can move you to a superior room for £15 a night."

• Exit: "No problem at all — just wanted to mention it."

The rule that matters more than the script: if a guest declines once, don't ask again unless they raise it. This is also the single hardest thing to get right in automated systems, and worth testing explicitly before you buy one.

If your team is too stretched to have these conversations at all, the constraint isn't your script — it's admin load. Our guide on hotel staff efficiency tools in 2026 covers where that time usually goes.

Implementation: The SOP Matters More Than The Software

The biggest risk in upselling isn't guest complaints. It's inconsistency. If whether an upgrade gets offered depends on who's on shift, you'll never learn what works.

Write a one-page SOP covering what gets offered, when, and who delivers it. For example: "Late check-out is offered pre-arrival when next-day occupancy is below 80%. Housekeeping confirms capacity by 10:00. Front desk applies the fee in the PMS." This kills the classic failure where an offer is sold and nobody owns fulfilment.

Write two versions of each message — one for business guests (quiet room, early check-in), one for leisure (view, breakfast, celebration add-ons). Same offer, different framing.

Track three numbers weekly, not thirty:

• Upsell conversion rate (offers accepted ÷ offers made)

• Ancillary revenue per stay (upsell revenue ÷ stays)

• TRevPAR movement, if you already track it

Fifteen minutes every Monday. Kill one underperformer, change one variable — price, timing or wording — and let it run seven days. That's how the compounding happens, and it doesn't add a project to anyone's week.

Where ButlerIQ Fits, And Where It Doesn't

ButlerIQ is an in-room voice concierge. A guest speaks, and gets an answer or a request logged — in their own language, with no app to download and no QR code to scan.

The commercial logic is narrow and specific. Every other channel for reaching an in-house guest requires the guest to do something first: find the number, open an app, scan a code, get dressed and go downstairs. Each of those steps loses people, and it loses them precisely at the moment they'd decided they wanted something. Removing the step is the whole product.

You can test that claim without talking to us. Look at how many in-stay requests reach your front desk between 8pm and 7am, then ask your night team how many guests mention, at check-out, something they'd have liked but didn't bother asking for. The gap between those two numbers is the revenue nobody is currently capturing — and it's a gap you can measure this week with a notebook.

What ButlerIQ is not: it is not a pre-arrival email upsell engine, and it doesn't replace one. If you're not yet running automated offers before arrival, a dedicated pre-arrival platform will make you money faster than we will. The two sit alongside each other, covering different windows. We're also a newer category than the pre-arrival platforms, with a correspondingly shorter track record — worth weighing honestly against the fact that nobody else is covering window four without an app.

If you want to see how in-room voice works across the whole stay rather than just the upsell moment, our overview of an AI concierge for hotels covers the wider picture, and creating seamless guest journeys with AI voice assistants maps the touchpoints in detail.

Where To Start

Offer only what you can reliably deliver. Time it to where the guest actually is. Keep the choice to one thing.

Practically: run the arithmetic on your own room count first. If you have no pre-arrival offers running, fix that this month — it's the fastest money on the list. Once that's working, look at what's still unsold after check-in, because that's where the guest knows most about what they want and has the fewest ways to tell you.

Frequently Asked Questions

What upsell conversion rate is realistic?

It varies widely by property type, offer quality and timing, and any vendor quoting you a single benchmark figure without knowing your occupancy pattern is selling, not advising. The more useful approach is to measure your own baseline for four weeks before you change anything, then judge every change against that. Ask vendors for conversion data from properties of comparable size and segment, not a headline average.

Will upselling annoy my guests?

Only if you repeat the ask. The complaints come from guests being offered the same thing three times across email, check-in and in-stay, because three different systems don't know about each other. Before buying, ask specifically how a decline is recorded and whether it suppresses the offer on other channels.

Do I need hardware in the room?

Not for pre-arrival or check-in upselling — that's all software and messaging. For in-stay you need some channel the guest will actually use, which in practice means either a device in the room or an app you've persuaded them to install. Both have costs; the app's cost is adoption rather than hardware.

Do these tools integrate with my PMS?

Most major upsell platforms integrate with the common systems, but coverage varies by region and by PMS version. Get the integration confirmed in writing against your specific PMS and version before signing, including whether it writes charges back automatically or needs manual posting.

Can a small independent hotel make this work?

Yes, and often better than a chain, because you can change an offer the same afternoon you think of it. The constraint at smaller properties is fulfilment capacity, not demand — so start with add-ons you can deliver every single time, like late check-out and breakfast, before adding anything that depends on a partner.

How long before it pays for itself?

Run the arithmetic in the first section against the vendor's fee. If the sums only work at a conversion rate above roughly 10%, treat that as an optimistic case and ask what happens at 5%.