Airbnb or Booking.com for a Nairobi short stay: which one actually pays more?

Owners ask which platform is better, and expect an answer naming one of them. For a furnished unit in Nairobi the honest answer is almost always both, which sounds like a dodge until you look at what actually separates them. It is not the commission, because that gap has just closed. It is that the two platforms deliver different guests, pay you on different terms, and will cheerfully sell the same night twice if nobody has connected the calendars.

The fees are now nearly identical, and that is new

Most of what is written about this comparison is out of date, because Airbnb changed how it charges hosts and a lot of advice still describes the old arrangement.

• Airbnb now charges a single host-only service fee of 15.5 percent, taken from the booking subtotal. Note what the subtotal includes: your nightly rate plus any cleaning fee, extra guest fee or pet fee you have set. Raise your cleaning fee and Airbnb's cut of it rises too. The guest sees no separate service fee at all.
• Booking.com charges a commission set when you register, which is typically around 15 percent, within a published range of roughly 10 to 25 percent depending on your market, property type and cancellation policy. The Preferred Partner programme, which buys better placement, adds about 2 to 3 points on top. If you use Booking.com to process the guest's card, that is roughly 1 to 3 points more again.

On a 100 dollar booking with no add-ons, Airbnb leaves you about 84.50 and Booking.com leaves you somewhere between 82 and 85. For practical purposes they are the same number. If you read somewhere that Airbnb is far cheaper for hosts, you were reading about the old split fee, where the host paid roughly 3 percent and the guest was charged the rest at checkout. That model is being retired.

The fee question is the one owners ask first and now the least interesting difference between the two.

What genuinely differs for a Nairobi unit

The guest is a different person

This is the difference that changes your revenue, and it barely gets mentioned. Booking.com's audience skews towards business travel: older, international, booking later, staying fewer nights, and far more likely to be travelling on someone else's budget. Airbnb skews leisure and longer stays, and in Nairobi it carries a particular segment nothing else reaches, which is diaspora Kenyans visiting family and wanting a whole apartment rather than a hotel room.

In Nairobi specifically that split is worth money, because the organisational travel around Gigiri, the UN complex and the embassies is a real and steady market that mostly does not book on Airbnb. The shape of that demand is set out in the diplomatic tenant market, and if it turns out to be most of your demand you may be running the wrong product altogether, which is the argument in Airbnb versus serviced apartment.

The practical upshot is unglamorous and useful: for most Nairobi units Airbnb fills weekends and long stays, and Booking.com fills the midweek gaps that were going to sit empty. Those are not competing bookings. That is the whole case for running both.

The money arrives on different terms

Airbnb collects from the guest and pays you roughly a day after check-in. You do nothing, and the amount that lands is the amount you were promised.

Booking.com's default is close to the opposite. The guest pays the property rather than the platform, and Booking.com invoices you its commission monthly. That means you need a way to actually take the money, and you need to settle an invoice on time every month or the listing gets suspended. For an owner who lives in Manchester and visits Nairobi twice a year, this is the single most common way a Booking.com listing quietly dies: not bad reviews, an unpaid commission invoice.

Cancellations, and the gap between booked and stayed

A large share of Booking.com's demand arrives on free cancellation, because that is what the platform's audience expects and what its search results reward. Bookings therefore look better than they settle, and a calendar that appears full in October can thin out considerably by the time October arrives. Airbnb's cancellation policies are chosen by you and enforced by the platform, so a confirmed booking is closer to money.

This has a direct consequence for pricing. If you set rates off gross bookings on Booking.com you will overestimate demand and underprice. How to think about that is in pricing strategy for Nairobi.

Guest verification, and who carries the risk

Airbnb verifies guest identity and shows you a review history from other hosts. Booking.com is lighter on both, which shifts risk onto you. In Nairobi the concrete version of this is the party let: a one-night booking on a Friday for two guests that arrives as fifteen. It is a solvable problem, with minimum stays, no same-day bookings and house rules that are actually enforced, but it is a bigger problem on the platform that tells you less about who is coming.

Reviews do not travel between platforms

You will build two reputations from zero, and neither counts on the other. Splitting your nights across two listings means both accumulate reviews at roughly half the rate, and the first three months on any platform are the ones where reviews matter most for placement. This is the one genuine argument for not launching on both at once: get an Airbnb listing ranking first, then add Booking.com to fill the midweek.

The real risk of running both is selling the same night twice

Two listings, two calendars, one bedroom. Nothing in Airbnb knows what Booking.com has just sold. The first time it happens you will discover that the cost is not the refund. Cancelling a confirmed guest carries a platform penalty, it damages your placement in search, and on Airbnb a host cancellation can leave a public note on the listing that every future guest reads.

The free fix is an iCal link between the two calendars, and it is worth knowing exactly how much protection that buys. iCal is a file each platform fetches from the other on a polling interval measured in hours, not seconds. For a quiet cottage that books three weeks out it is genuinely fine. For a Kilimani one-bed in December, where two people can book the same Saturday forty minutes apart, it is not, and the owners who get burned are almost always the ones who were told iCal counted as syncing.

A channel manager is the version that actually works, because both listings read from one calendar rather than copying each other on a delay. If you are running the unit yourself, this is the piece of software to buy before you add a second platform.

What we do with a unit, and what you see

We list on both, plus Vrbo where the unit suits it, with all of them reading from a single calendar through a channel manager. Airbnb generally carries the leisure and long stays, Booking.com takes corporate and midweek, and quotes for organisational bookings go out against an invoice directly, which neither platform charges commission on.

The part worth knowing is what lands on your statement. Every booking records which channel it came from and what that channel charged, kept separate from our own fee. So a line reads: gross, the platform's commission, cleaning, our 20 percent, your net. You can see what Airbnb took and what Booking.com took, per booking, rather than a single blended number you have to trust.

One thing we should be straight about, because it gets oversold elsewhere. We do not get a cheaper commission rate from either platform than you would get yourself. Nobody at this scale does. What the fee buys is the synced calendar, the pricing decisions, the person who reconciles the Booking.com invoice every month, and somebody in Nairobi when a guest is locked out at eleven at night. If a manager tells you they have negotiated a better platform rate for you, ask to see it on a statement.

Related reading: short-term rental management in Nairobi, which platforms your unit should be on, including Expedia, what occupancy to actually expect, how short-let income is taxed in Kenya and what short-let management costs.