Your comp set is probably wrong — and East Africa's construction pipeline is about to prove it

Most properties define their competitive set by who is nearby and similar. Buyers define it by what they substitute you against. With four in five of Kenya's pipeline rooms already under construction, the gap between those two definitions is now a datable, forecastable risk.

Ask an owner who they compete with and you usually get a list of properties that look like theirs, priced like theirs, within a few kilometres. It is a reasonable answer and it is the wrong one. Your competitive set is not decided by you, by proximity, or by star rating. It is decided by whoever is choosing — and what they put next to you on the screen or in the proposal.

That distinction is about to get expensive, because East Africa's new supply is unusually predictable.

The number that changes the exercise

W Hospitality Group's 2026 Hotel Chain Development Pipelines in Africa report, published 10 March 2026, put the continental pipeline at a record 123,846 rooms across 675 properties. The headline ranking is not the interesting part — Kenya sits fourth on volume, with 6,190 rooms across 35 properties, well behind Egypt's 45,984.

The interesting part is the construction ratio. Of Kenya's 6,190 pipeline rooms, 4,922 are already under construction: 79.5%. Ethiopia is at 79.9%, Tanzania at 77.5%. Egypt, which owns more than a third of Africa's entire pipeline, is at 51.4%. Nigeria is at 39.2%. Cape Verde is at 8.6%.

Read that as a risk table rather than a confidence table and it inverts. A signed-but-unbuilt pipeline is optional — it can slip, be renegotiated, or quietly die when demand disappoints. A pipeline in the ground is a delivery schedule. Egypt can defer roughly half its supply. Kenya, Ethiopia and Tanzania largely cannot.

W Hospitality adds its own caution, and it is worth holding onto: historical actualisation rates suggest delivery falls short of projections. That is a statement about timing, not cancellation. The rooms are coming. Only the date is negotiable — which is precisely why the date is the thing to research.

Why proximity is a bad proxy

Three failure modes recur.

The buyer substitutes across categories you don't consider. A corporate booker filling a Nairobi programme is not choosing between two upper-upscale hotels in the same suburb. They are choosing between your room and a serviced apartment, an extended-stay product, or a different suburb entirely with a shorter airport run. Michael Pownall of Valor Hospitality made a version of this point in June: oversupply becomes a problem mainly when developers build identical products for the same customer, while genuinely new categories — branded residences, lifestyle aparthotels, extended-stay — can expand the market instead (Tourism Update, 24 June 2026).

The buyer isn't the guest. If a tour operator, DMC or corporate programme controls the booking, your comp set is whatever else sits inside their agreement. You can be the best property on the street and lose the room night because a competitor is contracted and you are not.

The comp set is a moving object. Most owners set theirs once and inherit it for years. Roughly 2,000 branded rooms opened in Nairobi between 2023 and 2025, and Trevor Ward of W Hospitality described that supply as contributing to reports of occupancy declines of nearly ten percentage points, alongside pressure on rates. Around 20 branded hotels comprising some 3,650 rooms sat in the city's pipeline in the first quarter of 2026, roughly 80% under construction, with about 10 hotels and 1,500 rooms expected to open during 2026 (Tourism Update, 24 June 2026). A comp set defined in 2023 is describing a different city.

A method that survives new supply

1. Start from lost business, not from a map. Pull your last two quarters of declined, cancelled and lost enquiries and record where they went, wherever you can establish it. Your real comp set is in that list. It will contain at least one property or category you would not have named.

2. Segment it, because you have more than one. A city hotel typically has three: a corporate/RFP comp set, a leisure/OTA comp set, and a groups-and-meetings comp set. They overlap far less than owners assume, and a rate move that defends one can concede another. Bush and beach properties usually have two — the direct-booking set and the tour-operator-allocation set — and the second is often invisible from the property.

3. Add the pipeline, with dates. This is the step the construction ratio makes possible. For each competitor arriving in your set, establish brand, room count, positioning and expected opening quarter. Between the W Hospitality data and brand press releases, most of this is public. A high construction ratio is unwelcome news that comes with an unusual consolation: it is legible. You can put a date on it.

4. Ask the operator, and ask early. If you are branded, your operator knows which of its own openings hit your set and when. If you are independent, the construction site is visible from the road and the brand has usually announced.

5. Re-run it annually, and after every opening. Put it on the same cycle as the contracting round.

What a new entrant actually does

Underrated, and it decides how you should respond.

A new branded hotel opens with debt service, no guest history and no legacy rate to protect. It has one thing to buy in year one: occupancy. It will buy it with aggressive corporate and consortia rates, because those fill a base fast and are less publicly visible than a discounted transient rate.

Two consequences follow. The damage lands first in contracted rates, not in your published ADR — so a property watching only its rate card will see the problem late. And because contracting for a given year happens the year before, a hotel opening in 2027 is competing for 2027 corporate business during the round running now. Africa's chain pipeline expects 33,381 rooms to open in 2027 across 177 properties (W Hospitality Group, 10 March 2026). Some fraction of those are already quoting.

The response is not to match the rate. A new entrant can lose money for a year; an established property with a book of repeat business usually cannot, and matching trains your existing accounts to expect the lower number permanently. Defend instead on terms you control — length of stay, cancellation terms, meeting space, F&B inclusions, loyalty recognition — and on being present in the agreements where the substitution actually happens.

The short version

• Your comp set is defined by the buyer's substitution, not by geography or star rating.
• Build it from lost business, segment it by booking channel, and attach opening dates to it.
• East Africa's unusually high construction ratios mean the next three years of new supply are largely committed and therefore forecastable. Use that.
• Expect a new entrant to compete on contracted rates before published ones, which means the damage shows up in the contracting round rather than the rate card.
• Check what you are contracted into before the Q4 round. That is the defensible ground; the discount is not.

Supply you can see coming is not a threat in the way an advisory downgrade or an outbreak is. It is a scheduling problem. The properties that struggle will mostly be the ones that treated a delivery date as a forecast.