Take one hotel booking and work out what private-rate access saved. Depending on what you compare it with, the answer can be 6% or 29%. Both numbers are arithmetically correct, and only one of them will hold up when finance asks how you got it.
The gap comes almost entirely from the baseline: the price you say the company would have paid without the private rate. Pick it first, write it down, and use the same one every quarter. The rest of the calculation is bookkeeping.
Pick the baseline before you count anything
There are four baselines in common use for company-paid stays, plus the vendor's own headline figure, which isn't one.
| Baseline | What the saving measures | Where it goes wrong | Use it for |
|---|---|---|---|
| The hotel's own public price for the same room, dates and terms, captured when the booking is made | What the company avoided paying | Hotel prices change daily, so a price looked up a week later proves nothing | Your headline savings number |
| Your negotiated corporate rate at that hotel | What the private rate adds on top of the deal you already have | Only exists at contracted hotels, and only counts if the negotiated rate was bookable that night | Contracted hotels, reported separately |
| Per diem or policy cap | How far under the cap the booking came in | A cap is a ceiling, so a booking can land well under it without anyone saving anything | Compliance reporting, or cost avoidance at most |
| Last year's average paid in that city | Whether your hotel spend per night went up or down | Moves with the market and with where your people traveled, as well as with the rate | Trend context, next to the headline number |
| A vendor's "up to" percentage | The best case the vendor found | It's a maximum, so the typical booking saves less | Nothing in your own report |
Why a per diem saving changes with the month
Chicago shows the problem. The GSA per diem lodging rate for Chicago in fiscal 2027 is $148 from January to March and $238 from April to June. GSA's FAQ says the rates are set from "average daily rate (ADR) data of local lodging properties" and that the lodging rate is "a maximum amount."
Book a $170 room in May and you're $68 under the cap. Book the same room at the same price in February and you're $22 over. The "saving" moved $90 because the cap did.
GSA hit the same problem buying IT. Its IT category program says its cost avoidance methodology "uses last fiscal year pricing data as a basis of comparison against current pricing instead of using published rates (Master Contract Ceiling)". The same applies to hotels: a ceiling makes a poor baseline.
Match every comparison like for like
A baseline price only counts if it's for the same thing you bought. Check six things before a booking goes into the sample:
- Same hotel and same dates. A nearby hotel or a shifted night is a different price.
- Same room type and occupancy. A king against a twin, or one guest against two, changes the rate at many properties.
- Same cancellation terms. Compare prepaid with prepaid and flexible with flexible, or adjust for the difference (the next section shows how).
- Mandatory fees in. Under the FTC's fees rule, 16 CFR Part 464, in force since May 12, 2025, a hotel's displayed "total price" has to include every mandatory fee. Government charges "may be excluded."
- Taxes on both sides or neither. Some channels show taxes in the price and some add them at checkout. Per diem caps leave them out: GSA's FAQ states, "Lodging taxes are not included in the CONUS per diem rate." So if a per diem is one of your baselines, compare every baseline pre-tax.
- Captured at the time of booking. Screenshot or log the baseline price when the booking is made. You can't reconstruct it afterwards.
If you want the single-quote version of this, with fees and taxes walked through, it's in our guide to comparing a member rate with a public rate.
Subtract what the private rate costs you
Three things can eat into a saving. Each one goes in your calculation or, if it isn't company money, next to it.
Non-refundable terms. Private and wholesale rates are often prepaid. If the baseline you're comparing against is a flexible rate, take off the expected cost of cancellations: the share of these trips that get cancelled after the deadline, times what you lose when they do. Use your own cancellation history for that share. We couldn't find a published industry rate to use instead.
Lost hotel points and elite benefits. Private rates are usually wholesale rates, and some hotel programs exclude those. Marriott's Bonvoy terms list a room "booked at a tour operator, wholesaler, or crew room rate or package" as a non-qualifying rate, which earns no points and no elite benefits. Booked direct, most Marriott brands earn 10 base points per dollar.
Most other major chains credit points at check-in, but that's the hotel's call. The points are the traveler's loss, not the company's, so they don't come off your savings line. Report them next to it anyway, because the traveler who loses status is the one who goes back to booking direct. The arithmetic for valuing them is in how private hotel rates work.
Leakage. Savings only exist on bookings made through the channel. Multiply an average discount by your total hotel spend and you've counted every booking made somewhere else. Sum the saving booking by booking, and report the in-channel share of hotel nights next to it. Our piece on corporate travel dashboards covers how to measure that share.
A worked calculation
The hotel prices below are illustrative round numbers. The per diem is the real fiscal 2027 GSA rate for Chicago in May. Say a traveler books three nights at a private rate of $170 a night before tax, prepaid and non-refundable.
| Baseline (per night, before tax) | Baseline price | Saving per night | Saving as % of baseline |
|---|---|---|---|
| Hotel's own flexible rate, same room and dates | $200 | $30 | 15.0% |
| Hotel's own prepaid, non-refundable rate | $180 | $10 | 5.6% |
| Company negotiated rate (flexible) | $185 | $15 | 8.1% |
| GSA per diem, Chicago, May (FY2027) | $238 | $68 | 28.6% |
The defensible numbers are the first two rows. Against the prepaid rate, the terms match, so the saving is $10 a night: $30 over the stay, or $34.50 with an illustrative 15% tax.
Against the flexible rate, you have to price the flexibility. Three nights at $30 is $90, or $103.50 with tax. Now assume one in 20 of these trips is cancelled after the deadline and the whole $586.50 prepaid stay is lost. That's an expected loss of $29.33 a booking. A cancelled trip also earns none of the $103.50 saving, so the expected net is 95% of $103.50 minus $29.33: about $69.
If the hotel is a Marriott, a direct booking at $200 would have earned the traveler 6,000 base points. Value a point at an illustrative 0.7 cents (use your own figure) and that's $42 the traveler gives up. The company's $69 stays as it is. The $42 is what your frequent travelers will be weighing.
Sample bookings for a number finance will accept
If your booking tool logs a like-for-like public price at the moment of booking, use every booking. Most don't, which means sampling.
- Choose bookings as they're made. Take, say, every tenth booking reference. Picking from memory skews the sample toward the dramatic trips.
- Price the baseline the day it's booked. Check the hotel's own site for the same room, dates, occupancy and terms, and save the screen with a timestamp.
- Split by city or hotel tier if your spend is concentrated. Weight each group by its share of nights when you combine them.
- Report the range alongside the average. A rough 95% margin of error is twice the standard deviation of the savings divided by the square root of the sample size. With 100 bookings, an average saving of 9% and a standard deviation of 10 points, that's 2 points either side, so 7% to 11%. With 25 bookings it's 4 points, so 5% to 13%.
- Repeat every quarter with the same rules, so a change in the number means a change in the savings and not in the method.
How to report hotel savings to finance
Finance teams usually separate hard savings, spending removed from a line that was going to be paid, from cost avoidance, spending that might have happened and didn't.
Where a saving against the hotel's own price lands is a judgment call. Some finance teams count it as hard savings; others call any saving against a price you never paid cost avoidance. Agree the label with finance before the first report. A saving against a per diem cap is cost avoidance at most, so label it that way or leave it out.
A savings line that survives review names its baseline, its sample size and its range, and keeps the offsets visible. For example: "Against the hotel's own price for the same room, dates and terms, captured at booking, in-channel stays saved 9% (range 7% to 11%, 100 sampled bookings). In-channel share of hotel nights was X. Travelers gave up points on Y stays."
Decide whether you report booked or settled value and stick to it, since trips change after they're booked. If the savings figure is going into a business case for a travel benefit as a whole, it becomes one input to the ratio set out in our employee travel benefits ROI framework.
Where Dyme fits
Dyme's private rates are wholesale, which is where the discount comes from. On the Dyme for Business page we say teams can save "up to 35%," that this "reflects current member rates against published consumer rates," and that "savings vary by hotel, date and inventory." In the terms of this guide, that's a best case against something close to the first baseline in the table. Your own sample is the number to report.
The price on Dyme includes taxes and booking fees, so add taxes to the hotel's price before you compare, or take them off Dyme's if a per diem is one of your baselines. The exception is resort fees some hotels collect at check-in, which Dyme flags when they apply.
Marriott Bonvoy doesn't credit stays booked through third parties like us, while most other chains credit points at check-in at the hotel's discretion. Every booking lands in one place with its receipt, which gives you the booked side of the sample. You still capture the hotel's price yourself.
FAQ: calculating corporate hotel savings
What baseline should we use to measure hotel savings?＋
The hotel's own public price for the same room, dates, occupancy and cancellation terms, captured when the booking is made. It measures what the company avoided paying. Report savings against your negotiated rate separately, for contracted hotels only. Use last year's average paid per night as trend context.
Can we count savings against the per diem rate?＋
Only as cost avoidance, if at all. GSA describes the per diem lodging rate as "a maximum amount" set from local average daily rate data, so a booking can come in under it without anyone negotiating anything. Compare pre-tax prices against it, because GSA's FAQ says lodging taxes are not included in the CONUS rate.
Should lost hotel points come off the savings figure?＋
Not off the company's figure, because the points belong to the traveler. Report them next to it. Marriott's Bonvoy terms exclude rooms booked at a wholesaler rate from points and elite benefits. A traveler who loses status that way may stop booking through the channel, and then the saving disappears.
How many bookings do we need to sample?＋
Enough that the margin of error is small next to the saving; 100 bookings is a sensible start. A rough 95% margin is twice the standard deviation of the savings divided by the square root of the number of bookings. With a standard deviation of 10 points, 100 bookings gives about 2 points either side and 25 bookings gives about 4.
How do non-refundable private rates change the calculation?＋
If the baseline is a flexible rate, subtract the expected cost of cancellations: the share of trips cancelled after the deadline times the prepaid amount lost. Those cancelled trips also earn no saving, so count the saving only on the trips that go ahead. Use your own cancellation history for the share. Or compare against the hotel's own prepaid rate, where the terms match and no adjustment is needed.
