Host Metric Studio
Airbnb Bookings Down in 2026? Diagnose It With 6 KPIs

It's 27 October. Last year at this point you had 16 nights booked for December. Today you have 11. Your first thought is to cut the price for the whole month.
Wait before you do. A discount costs you money on every night that would have sold anyway. First find out why you're behind. (Example numbers, invented.)
If your Airbnb bookings are down, check six numbers before you cut your price: pace against last year (nights on the books compared with the same date last year), lead time (how many days ahead guests book), average stay, the price the guest sees since the fee change, local supply and conversion (views to bookings). Across the platform, demand has not collapsed in 2026, so look for the cause of a drop in your own bookings in your listing and your local market first. Each KPI below comes with its formula, a threshold to act on and an invented example, followed by a checklist that tells you what to fix first.
"My bookings are down" is a symptom, not a diagnosis. A common reflex is a blanket discount across the calendar, which costs money on every night that would have sold anyway. Fifteen minutes with your own numbers usually tells you whether the problem is timing, price, calendar rules, competition or the listing itself.
Think of a doctor. A fever is a symptom; nobody treats it before finding the cause. "Fewer bookings" is your fever. The six KPIs are the tests that tell you what is really wrong.
First check: is demand down, or just your bookings?
Short version: across the platform, demand has not collapsed. What changed most is when guests book, not how many book. If you only have a minute, skip to the 6 KPIs.
Start with the market, briefly, so you know what you are comparing against. The most recent public figures available when this was written:
- Airbnb, Q2 2026 shareholder letter (6 Aug 2026): Nights and Seats Booked grew 10% year over year, EMEA grew at a high-single-digit rate, and Airbnb expected "low double-digit growth" in Nights and Seats Booked for Q3 (SEC filing, as of 27 Oct 2026). The metric also counts seats for experiences and services, so it is a platform figure, not a host benchmark.
- AirDNA, U.S. midyear outlook (8 Jul 2026): occupancy is forecast to stay roughly flat at 57.4% for 2026, demand is expected to soften through the rest of 2026, and in July AirDNA saw guests booking later and taking shorter trips. Studios to two-bedroom listings were roughly flat, with occupancy down about 0.7% (AirDNA, as of 27 Oct 2026). AirDNA's October report (below) updates the timing part: U.S. lead times lengthened through 2026 as fewer guests booked at the last minute.
- AirDNA, European review September 2026 (published 8 Oct, updated 22 Oct 2026): demand nights fell 1.2%, and Spain accounted for 77% of the net decline (Spain โ8.1%) as licensing rules shrank its supply, while occupancy held flat at 55.4% and ADR rose 8.9%. The same report shows nights booked in September (adjusted for cancellations) down 13.5% and November nights on the books 3.8% behind last year on 1 October; AirDNA adds that European lead times did not shorten but stretched at the far end (the average rose from 62 to 68 days for September stays), so guests "changed when they booked more than how much they booked" (AirDNA, as of 27 Oct 2026).
The picture is mixed, not a collapse: more nights on the platform, flat occupancy in the big markets, rates up, and a booking curve that has shifted in time. So if your bookings have fallen, measure your own listing before you blame the market. Anecdotes in host forums are useful for knowing what other hosts are asking, not as data.
The 6 KPIs, with formulas
You need a bookings export with one row per stay: booking date, check-in, nights and revenue. The STR spreadsheet template (Excel or Google Sheets) already has these columns. Always compare the same stay month, measured on the same date last year.
KPI 1 โ Pace vs last year. How many nights are on the books for a future month today, compared with the same date last year.
Nights on books (this year) =SUMIFS(Nights,StayMonth,M,BookedOn,"<="&TODAY())
Nights on books (last year) =SUMIFS(Nights,StayMonth,M_last,BookedOn,"<="&EDATE(TODAY(),-12))
Pace vs last year =this_year/last_year-1
KPI 2 โ Lead time. The number of days between booking and check-in. If your guests now book later than last year, a weak pace today may only mean the bookings haven't arrived yet.
Lead time (per booking) =CheckIn-BookedOn
Median for a stay month =MEDIAN(FILTER(LeadTime,StayMonth=M))
KPI 3 โ Average length of stay. Shorter trips mean fewer nights per booking and more turnovers. If your minimum stay is longer than the stays guests now want, you are filtered out of their searches.
Average stay =SUMIFS(Nights,StayMonth,M)/COUNTIFS(StayMonth,M)
KPI 4 โ The price the guest sees, before vs after the fee change. Under the old split fee, the guest saw your price plus a guest fee of 14.1%โ16.5% (Airbnb Help 1857, as of 27 Oct 2026). Under the single fee, the guest sees your price. Compare like with like, before taxes:
Guest price before =old_price*(1+old_guest_fee)
Guest price now =new_price
Change =guest_now/guest_before-1
A host who raised prices by more than the fee required is now more expensive than last year; one who didn't raise them is cheaper but earns less. The 15.5% repricing guide has the exact formula.
KPI 5 โ Local supply. The number of comparable listings available for the same dates. Search Airbnb with your guest count, bedrooms and key filters for a date in the month you're diagnosing and note how many results come up. If you never recorded it last year, start a monthly log today; a market-data tool can give you the history.
Supply change =comps_now/comps_last_year-1
KPI 6 โ Conversion (views to bookings). Airbnb's Insights tab shows views and booking conversion for hosts who use professional hosting tools, split into search impressions, search-to-listing and listing-to-booking (Airbnb Help 2714, as of 27 Oct 2026). A simple version you can track yourself:
Conversion =bookings/views (same 30 days, this year vs last year)
A worked example: one listing, December
๐งฎ Example only โ invented numbers
An invented one-bedroom listing, measured on 27 October for December stays. Last year the host charged $100 and the guest paid a 14.5% guest fee on top; this year the host moved to the single fee and set $125 "to be safe".
| KPI | Last year (same date) | This year | Change |
|---|---|---|---|
| 1. Nights on books for December | 16 | 11 | โ31% |
| 2. Median lead time (December stays) | 38 days | 24 days | โ14 days |
| 3. Average stay | 3.2 nights | 2.4 nights | โ25% |
| 4. Price the guest sees per night | $114.50 | $125.00 | +9.2% |
| 5. Comparable listings available | 42 | 49 | +16.7% |
| 6. Views (last 30 days) | 1,250 | 1,180 | โ5.6% |
| 6. Bookings (last 30 days) | 9 | 5 | โ |
| 6. Conversion | 0.72% | 0.42% | โ41% |
Reading it KPI by KPI:
- Pace looks alarming (โ31%), but it's partly timing. Last year 16 of the final 24 December nights (66.7%) were booked by 27 October. With the median lead time down from 38 to 24 days, and December only 35 days away, more of this year's bookings are still to come. At last year's curve, 11 nights would project to about 16.5; with later bookings it could finish higher.
- Stays are shorter (3.2 โ 2.4 nights), so the same demand produces fewer nights per booking. Check that the minimum stay isn't blocking 2-night searches.
- The guest price is the real problem. The host needed $114.79 to keep the old $97 payout ($100 ร 0.97 รท 0.845). At $125 the guest pays 9.2% more than last year, in a market with 16.7% more comparable listings.
- Views are stable (โ5.6%) but conversion fell 41%. Guests still find the listing; they just book something else once they see it. That points to price or the listing page, not to search ranking.
Views down means guests aren't finding you. Views stable and conversion down means they find you and pick someone else.
Diagnosis for the example: an over-adjusted price in a busier market, plus a booking curve that has moved later. The fix is to bring the price back to about $115 (which keeps the $97 payout) and re-measure in two weeks. If 2-night searches are still blocked by your minimum stay (question 3), relax it for December weekdays next. Don't discount the whole calendar.
Decision checklist: price, calendar, listing or market?
Work through all six questions; don't stop at the first "yes". Timing (question 1) tells you when to judge the gap, not whether something else is wrong. Fix every "yes" you can act on now, one change at a time, starting from the top. The thresholds are my own rules of thumb for a single listing, not industry standards; adjust them to your volume.
- 1Is the gap outside your normal booking window? If the month is further away than your median lead time, part of a weak pace is timing: don't cut prices across the month for that reason, and re-check weekly. Timing never rules out the other five questions, so keep going.
- 2Is the guest-visible price more than about 5% above last year? Compare KPI 4 with last year and with 3โ5 comparable listings. If you over-adjusted for the fee, correct it with the repricing formula.
- 3Are stays shorter than your minimum stay allows? If the average stay fell and your minimum is above it, test a shorter minimum on weekdays or gaps.
- 4Did comparable supply grow more than about 10%? Then the market got more crowded. Compete on what guests compare first: price per night for your size, the cover photo, reviews and flexibility.
- 5Are views down by more than about 20%? That's visibility: check your calendar is open, the availability window is long enough, the price isn't outside common search filters and nothing is blocking dates.
- 6Views stable but conversion down? Then it's the listing page and price at the moment of decision: photos, title, the first lines of the description, reviews, cancellation policy and total price. Change one thing at a time and watch occupancy against a sensible benchmark.
In the example, question 1 is a "yes" (December is 35 days away, the median lead time is 24 days), so part of the gap is timing. But questions 2 (guest price +9.2%), 4 (supply +16.7%) and 6 (conversion โ41%) are "yes" too, and together they point to an over-adjusted price in a busier market. Question 3 depends on the host's minimum stay, which the example doesn't give: stays fell to 2.4 nights, so a minimum above 2 nights would make it a "yes" too. That's why the fix is the price first, then the minimum stay, not waiting.
Diagnosing from one month. Small listings have small numbers: 5 bookings instead of 9 can be noise. Look at two or three stay months together, and treat any single KPI as a hint until a second one points the same way.
Save this check as a tab in your bookings sheet and run it on the same day each month. The first run tells you little; by the third you will see which KPI moves first in your market.
3 ideas to keep, and your next step
- 1Fewer bookings is a symptom, not a diagnosis. Measure before you discount.
- 2Compare like with like. Same stay month, measured on the same date last year. Check timing (lead time), but don't let it hide a price problem.
- 3Run the whole checklist. A "yes" on timing doesn't end it. Fix what you can act on now, one change at a time, and re-measure.
Export your bookings today and fill in the six KPIs for your next weak month. If that month is further away than your median lead time, re-check it weekly, but run the price check now: if the price guests see is more than about 5% above last year, fix that first.
FAQ
Are Airbnb bookings down in 2026?
Not across the platform, according to the latest public figures. Airbnb reported 10% growth in Nights and Seats Booked in Q2 2026, and AirDNA forecast roughly flat U.S. occupancy at 57.4% for 2026. Some markets are softer: AirDNA reported European demand nights down 1.2% in September 2026, mostly because supply shrank in Spain. Check your own KPIs before assuming the market is the cause.
Why am I not getting bookings on Airbnb?
Check all six things: whether the month is still outside your normal booking window, the price guests see compared with last year and with similar listings, whether your minimum stay is longer than the stays guests want, how much local supply has grown, whether your views dropped (visibility) and whether views turn into bookings (conversion).
Should I lower my price if bookings are slow?
Only after you know why. If the month is still far away, a discount mostly lowers revenue on nights that would have sold. If your guest-visible price rose after the fee change, correcting it fixes the problem without a blanket discount.
Did Airbnb's 15.5% fee change what guests pay?
With the single fee, guests see your price instead of your price plus a guest service fee. If you repriced with the correct formula, guests pay roughly what they paid before; if you raised prices by more than needed, they now pay more.
How long should I wait before changing something?
Give each change about two weeks, or enough views to compare, and change one thing at a time. If you change price, minimum stay and photos on the same day, you won't know which one worked.
To get two of these KPIs, average booking window (lead time) and average stay, from your bookings every month, along with revenue on the books against your own target, the STR Income & Expense Tracker does it for up to 10 properties โ $19 at https://emprendemos.gumroad.com/l/str-income-expense-tracker.
Not affiliated with or endorsed by Airbnb, Vrbo or Booking.com. Example figures are invented; Airbnb and AirDNA figures are cited with their source. This article is not tax or financial advice.