Choose an upselling platform by checking how it attributes revenue, whether it knows real inventory before it sells an upgrade, and how the accepted offer reaches the front desk. Ask for the definition behind any incremental revenue figure, run a pilot on two offers, and compare against what your team already sells manually at check in.
Every vendor quotes an incremental revenue per room. Ask how it is calculated. If it counts every upgrade sold through the tool, it is also counting the upgrades your reception would have sold anyway, and the true increment is smaller. A vendor who can explain the counterfactual is worth listening to.
Then ask for a range across hotels of your type and size, and what the low end looks like. The average in a case study is usually the best performing segment. What you need is the number a property like yours should plan for.
An upsell that the hotel cannot deliver is a complaint and a refund. Check that the platform reads live availability from your PMS before offering an upgrade, and what it does when a room type sells out between the offer and the acceptance.
The same applies to non room items. Early check in, late check out and parking depend on the shape of the day, not just on a stock count. Ask whether the tool can be restricted by date, occupancy level or day of week, so the offers stop themselves when operations cannot cope.
Skip the research, talk to someone who has done it 45 minutes with an independent specialist. Free for hotels, no pitch, no commissions. Book a free session →Pre arrival messaging reaches a guest who is looking forward to the trip and has budget in mind. Check in reaches a guest who is tired and wants a key. The same offer performs very differently at those two moments, and the platform should let you control which offers appear when.
Frequency is a reputational cost. Three upsell emails before a two night stay reads as pressure. Decide the ceiling before launch and check the tool respects it across channels, including any messages sent by your CRM.
When a guest buys an upgrade at eleven at night, someone has to know. Ask how the accepted offer lands: does it update the reservation in the PMS, does it create a task, does it print on the arrivals list, or does it send an email that nobody reads at seven in the morning.
Payment matters here too. Confirm whether the guest is charged at acceptance or at check out, and how a refund works when the hotel cannot deliver. That path will be used, so it should be tested before launch rather than during.
Relevant, well timed offers generally do not. Repetition does. The common failure is not the first message but the third, especially when it repeats an offer the guest already declined. Cap the frequency and suppress declined offers and the complaints largely disappear.
Start with the two that need no operational change: a room upgrade and late check out. Both use inventory you already control and neither requires a new supplier. Once the reporting is trusted, add items that involve other departments.
At booking the guest is comparing prices and is sensitive to the total. After booking the decision is already made and the reference point has shifted, which is why the same upgrade converts better pre arrival. Comparing both timings on your own data is worth a season of testing.
Yes, and the tool should support them rather than replace them. A front desk that can see which offers a guest already declined avoids the awkward repeat, and a guest who says yes in person is the highest converting moment you have.
Start from the rate difference between room types on that date and offer a supplement below it, since the guest has already paid the base rate. Fixed supplements are simpler to explain and easier for reception to honour when a guest asks at the desk.
Offers sent, acceptance rate by offer and by date, revenue accepted, and how much of it was delivered rather than refunded. Without the last number the reporting overstates performance every time a hotel cannot honour what was sold.