Track eight numbers, not eighty. In my own portfolio and across the operators I have trained since 2016, the short-term rental businesses that survive a soft season watch occupancy rate, average daily rate (ADR), RevPAR, net operating income (NOI), cost per stay, direct-booking share, inquiry-to-booking conversion, and review average. Everything else is a report nobody reads.
This guide gives you the formula for each one, the failure mode it catches, and how often to look at it. The numbers in the worked examples are illustrative and labelled as such. Your job is to run the same arithmetic on your own bookings.
What a KPI Actually Does for a Rental Business
A key performance indicator is a number tied to a decision. If a metric moves and you still do not know what to change, it is trivia, not a KPI.
Three tests before a number earns a spot on your dashboard:
It is arithmetic you can reproduce. You should be able to rebuild it from your booking export and your bank statement.
It has a threshold. "Occupancy below 55% for two consecutive months" triggers a pricing review. A number without a trigger gets ignored.
It maps to one owner. Cleaning cost per stay belongs to whoever schedules turnovers. Conversion rate belongs to whoever writes the listing.
The most common mistake I see is operators tracking platform dashboards only. Airbnb shows you Airbnb. If you also run Vrbo, a direct site, and mid-term stays, the platform view will tell you a flattering story about a fraction of your business.
The Eight Core Metrics and Their Formulas
Definitions first, because half the arguments in host groups come from two people using the same word differently.
Occupancy rate = nights booked ÷ nights available × 100. Decide once whether owner-blocked nights count as unavailable, then never change it. If you block 10 nights of a 30-night month and sell 15, occupancy is 15 ÷ 20 = 75% on available nights, or 50% on calendar nights. Both are defensible; mixing them month to month is not.
Average daily rate (ADR) = room revenue ÷ nights booked. Exclude cleaning fees and taxes. Including cleaning fees inflates ADR on short stays and makes your pricing look stronger than it is.
RevPAR = room revenue ÷ nights available, or equivalently ADR × occupancy. This is the single number I would keep if forced to pick one, because it punishes both empty calendars and desperate discounting.
Net operating income (NOI) = total revenue − operating expenses, before debt service. Operating expenses include cleaning, supplies, utilities, insurance, property tax, software, and management. It excludes your mortgage principal and interest, which is what makes NOI comparable across properties financed differently.
Cost per stay = total turnover cost ÷ number of stays. Cleaning, laundry, consumables, restocking. This is the metric that decides your minimum-night rule.
Direct-booking share = direct reservations ÷ total reservations × 100. Platform commissions vary by fee structure — a few percent of the host payout under Airbnb's split-fee model, mid-teens under the single host-only fee — so every point of direct share is margin you keep. Check your own payout statements for your exact rate.
Inquiry-to-booking conversion = bookings ÷ inquiries × 100. A falling rate usually means price, photos, or response time, in that order.
Review average and 5-star rate. Track the share of stays that produce a 5-star overall rating, not just the running average. The average moves slowly and hides a bad month.
Here is how those eight land on a calendar. I run the daily items in under five minutes with coffee.
Metric | Formula | Review cadence | Trigger to act |
|---|
Occupancy rate | Nights booked ÷ nights available | Weekly | Below 55% two months running |
ADR | Room revenue ÷ nights booked | Weekly | Down 10% year over year |
RevPAR | ADR × occupancy | Monthly | Down two months in a row |
NOI | Revenue − operating expenses | Monthly | Any negative month |
Cost per stay | Turnover cost ÷ stays | Monthly | Above 20% of ADR |
Direct-booking share | Direct ÷ total bookings | Quarterly | Below 15% |
Inquiry-to-booking | Bookings ÷ inquiries | Weekly | Below 20% |
5-star rate | 5-star stays ÷ total stays | Monthly | Below 80% |
Worked Example: One Unit, One Month
Illustrative numbers, but the arithmetic is the point. Assume a two-bedroom unit, 30 nights available, no owner blocks.
Nights booked: 21. Occupancy = 21 ÷ 30 = 70%.
Room revenue: $4,410. ADR = $4,410 ÷ 21 = $210.
RevPAR = $4,410 ÷ 30 = $147. Same answer as $210 × 0.70.
Stays: 7. Turnover cost at $95 each = $665. Cost per stay = $95, which is 45% of ADR — well over the 20% trigger, because the average stay is only three nights.
Other operating expenses: rent or ownership carry $1,800, utilities $180, insurance $95, software $60, supplies $70 = $2,205. Total operating expenses including turnovers = $2,205 + $665 = $2,870.
NOI = $4,410 − $2,870 = $1,540.
Now change one variable. Raise the minimum stay to four nights. Suppose bookings fall to 19 nights across 5 stays at the same $210 ADR. Room revenue = $3,990, RevPAR = $133, turnover cost = $475, total operating expenses = $2,680, NOI = $1,310. Lower NOI, so the longer minimum did not pay at that ADR.
Run it again with a $20 ADR lift that longer stays often support: 19 nights at $230 = $4,370, NOI = $1,690. That beats the original $1,540. The lesson is that minimum-night rules only work when they come with a rate change, and RevPAR alone would have told you the wrong story in scenario two.
Scenario | Nights sold | ADR | RevPAR | Turnover cost | NOI |
|---|
3-night minimum | 21 | $210 | $147 | $665 | $1,540 |
4-night minimum, flat rate | 19 | $210 | $133 | $475 | $1,310 |
4-night minimum, +$20 rate | 19 | $230 | $146 | $475 | $1,690 |
Notice that scenario three has slightly lower RevPAR than scenario one and higher NOI. Revenue metrics rank your calendar. Profit metrics rank your business. Keep both.
Operational Metrics Worth Keeping
Revenue metrics tell you what happened. Operational metrics tell you why.
Median first-response time. Measure in minutes, not hours, and use the median so one overnight message does not distort the month. Under an hour during waking hours is the bar I hold my teams to.
Turnover completion rate. Share of cleans marked done before the next check-in window opens. Anything under 100% is a future one-star review.
Maintenance tickets per 100 nights. In my portfolio, rising counts have tended to show up a quarter or two before review scores slip.
Cancellation rate by channel. Split host cancellations from guest cancellations. Host cancellations carry platform penalties, and Airbnb publishes the specifics in its host cancellation policy.
Payout timing. Know when money actually lands. Airbnb's payout schedule determines your working-capital cycle, and operators who ignore it end up short on the week cleaners invoice.
The Monthly Review Process
This is the routine I teach. Ninety minutes, once a month, same day each month.
Export the prior month's reservations from every channel into one sheet. One row per reservation, with channel, nights, payout, and fees.
Pull the bank and card transactions for the same period. Categorize into cleaning, supplies, utilities, insurance, tax, software, management.
Calculate the eight core metrics per property. Do not average across the portfolio yet; a strong unit hides a failing one.
Compare each metric to the same month last year, not to last month. Seasonality will lie to you otherwise.
Flag every metric that crossed a trigger in the table above. Write the trigger down before you look at the number so you are not rationalizing.
Choose at most two changes for the coming month. One pricing change, one operations change. More than that and you cannot attribute the result.
Set a review date for those two changes 30 days out and put it in the calendar.
Two changes a month compounds to twenty-four experiments a year. That is more deliberate testing than most operators do in five years.
Pricing, Occupancy, and the Discount Trap
Demand-based pricing tools adjust your nightly rate against market signals, and they are worth using. They are not worth trusting blindly. The tool optimizes for bookings; you optimize for NOI.
Check the trade every time you consider a discount. Dropping ADR from $210 to $180 is a 14% rate cut. To hold the same room revenue you need occupancy to rise from 70% to about 82%, because $210 × 0.70 = $147 and you would need 147 ÷ 180 = 0.817. Those extra nights also add turnovers, so your cost per stay line grows while revenue stays flat. That is how operators end up busier and poorer.
The reverse test is cleaner. If a $15 rate increase costs you two nights out of 21, you go from $4,410 to 19 × $225 = $4,275 in revenue but save one turnover's worth of wear and roughly $95 in variable cost. Close to a wash on cash and easier on your team.
Tax treatment also belongs in this conversation. Average stay length affects how the IRS classifies your rental activity and what you can deduct, and many jurisdictions apply hotel or lodging tax to stays under a set number of nights. Read the rules for your state before you change your minimum-night setting.
If your numbers are telling you something you cannot interpret, bring them to a complimentary one-on-one diagnostic call at cashflowdiary.com/diagnostic. We will read the actual figures together. If it makes sense to go further, the $997 action plan builds the next 90 days around what the numbers said.
Building a Dashboard You Will Actually Open
Start in a spreadsheet. One tab of raw reservations, one tab of expenses, one summary tab with the eight metrics per property and a year-over-year column. Most operators do not need software until they pass roughly five units.
Rules that keep a dashboard alive:
One screen. If you scroll, you stop looking.
Absolute numbers next to percentages. "62% occupancy" means nothing without "18 of 29 nights."
Color only for triggers. Red means a threshold was crossed, not "this number is smaller than that one."
Archive monthly snapshots. Year-over-year comparison is impossible if you overwrite the same cells.
When you outgrow the sheet, a property management system with real reporting will pull channel data automatically. Test any tool by rebuilding one month by hand and checking whether the software agrees with you. Where they disagree, the software is usually counting fees or blocked nights differently than you are.
Frequently Asked Questions
Which KPIs should a new operator start with?
Start with four: occupancy rate, ADR, RevPAR, and NOI. Those four tell you whether the calendar is full, whether the price is right, whether the combination is working, and whether any of it is profitable. Add cost per stay and direct-booking share once you have three months of clean data.
How is RevPAR different from ADR?
ADR only counts the nights you sold, so a single expensive booking in an empty month can make ADR look excellent. RevPAR divides revenue by all available nights, so an empty calendar drags it down. Use ADR to judge pricing and RevPAR to judge overall calendar performance.
Should cleaning fees be included in ADR?
No. Keep cleaning fees out of ADR and track them separately as turnover cost. Folding them in inflates the apparent rate on short stays and hides the real cost of high turnover. Report room revenue and fee revenue on separate lines.
How often should I review these numbers?
Look at occupancy, ADR, and inquiry conversion weekly, since pricing and listing fixes are time-sensitive. Run the full eight-metric review monthly with expenses reconciled. Compare against the same month last year rather than the prior month so seasonality does not distort the read.
What occupancy rate should I target?
There is no universal target, because it depends on your market's seasonality and your cost structure. A 60% occupancy at a high ADR often beats 85% at a discounted rate once turnover costs are subtracted. Set your target by working backward from the NOI you need, then test it against three months of actuals.
Do I need software to track vacation rental KPIs?
Not below about five units. A spreadsheet fed by monthly channel exports handles the arithmetic fine and forces you to look at the raw data. Move to a property management system when manual consolidation takes more than 90 minutes a month or when multiple people need the same view.
Sources
Related Reading
