Nigeria Property Centre
METHODOLOGY

How the Rent, House Price and Short-let Index are made.

What we count, the statistic we use, and how figures are labelled, corrected and cited. Applies from the September 2026 release.

Last updated 11 Oct 2026 · September 2026 release · Method version 2 (from September 2026)

01

What's counted

Each index counts flats and houses with 1 to 5 bedrooms that were live on Nigeria Property Centre on any day of the month, plus those first published during the month.

Duplicates count once: listings with the same advertiser, the same locality, the same bedroom count and the same standard price are treated as one property.

The Rent Index covers listings to rent, the House Price Index covers flats and houses for sale, and the Short-let Index covers short-lets priced by the night.

02

What's left out, and why

These listings are left out, so a figure describes real homes at the prices advertisers ask for them:

  • Removed by staff during the month as invalid, a duplicate or a repost1,562 in September 2026
  • Flagged "Multiple property in listing", or kept offline after Advanced Review, during the monthLeft out
  • Duplicates beyond the first (same advertiser, locality, bedrooms and standard price)Counted once
  • Junk prices: more than 20 times the median for the same state, category and bedroom count (high side only)617 in September 2026

Studios, rooms, 6+ bedroom homes, land and commercial property are not in these series.

03

The price used

  • Rent and sale: the standard price, meaning the rent for a year or the sale price of the whole property. A listing counts at its price on the last day it was live in the month; a later change never alters that month.
  • Listings priced in US dollars are converted at the site's own USD rate on the night each listing was recorded, and that value is then frozen. A later rate never changes a past month. This is not the Central Bank rate.
  • Short-let: nightly prices only. Short-lets priced by the week or the month are left out, not divided down to a night.

USD-priced in September 2026: 1.4% of rent listings, 2.7% of sale listings and 0.2% of short-let listings.

04

The statistic

Each figure is a grouped interpolated median. We sort the listings into fixed price bands, find the band that holds the middle listing, and read the median off inside that band, in proportion to how far into the band the middle listing falls.

The typical range runs from the grouped lower quartile to the grouped upper quartile, read off the same way. It covers the middle half of listings.

Price bands Each power of ten is divided at the round prices advertisers use: 1, 1.2, 1.5, 2, 2.5, 3, 4, 5, 6, 7, 8, 9 times a power of ten. Each round price sits inside its own band, never on an edge: the edges are the geometric midpoints between neighbouring round prices, so the ₦8M band runs from about ₦7.5M to ₦8.5M. The bands are the same for every market, series and month, in the market's own currency.
Example with made-up counts: 2,184 rent listings sorted into price bands, each named by the round price it is built around.
  • Lower quartile ₦9.1M: 60% of the way into the ₦9M band (₦8.5M–₦9.5M).
  • Median ₦11.4M: the middle listing (1,092nd) falls 20% of the way into the ₦12M band (₦11M–₦13.4M).
  • Upper quartile ₦17.3M: at the top of the ₦15M band (₦13.4M–₦17.3M).
05

Thresholds and labels

  • Small sample Published from 20 to 99 listings. Quote with care: these move more month to month.
  • 3-month figure Too few listings in one month, so we pool the last 3 months. Changes on last month and on a year ago read Not available.
  • Not enough listings for <Month> Fewer than 20 listings, even over three months. We show the last published figure with its date.
  • Not available for <Month> Too many listings were last recorded just before a gap in our nightly record, so the month could not be counted reliably. We show the last published figure with its date.
  • Reconstructed Worked out after the month, from listings archived at the time.
  • Corrected <date> Changed after release. The old and new value and the reason are shown with it.

Below 20 listings, even over three months, we don't publish a figure, and a count under 20 is never shown. A missing comparison reads Not available, never 0.0%. The biggest-movers list uses areas with 100 or more listings and single-month figures only.

06

The method change of September 2026

Previous method: median asking price of listings created in the previous six months. From September 2026 each figure is a grouped median of listings live on any day of the month, or first published in it. The two can't be compared, so charts show older history in grey behind a "Method changed" marker, and no change is worked out across it. Changes on a year ago return with the September 2027 release.

07

Reconstructed months

September 2026 is reconstructed: we worked it out after the month ended, from listings archived at the time, so the series has a first month to compare the next one against. It is labelled Reconstructed wherever it appears.

08

Corrections policy

If we find an error after release, we correct the figure, label it Corrected with the date, and show the old value, the new value and the reason next to it. Changes that depend on it are worked out again. Corrections are listed in the release where they happen and in the release email. We never change a figure without saying so.

Releases come out on the 3rd working day of each month at 09:00.

09

Rental yield

On a sale figure page, the House Price Index shows a gross rental yield under the price. It answers one question: what would a home like this rent for, against what it costs to buy?

Gross yield = a year's typical asking rent ÷ typical asking price × 100 Both are the month's medians for the same area, property type and size, from the same release. The result is rounded to one decimal place.
  • A yield is worked out for flats, or for houses, at one size (1 to 5 bedrooms) or all sizes, nationally or in one state or locality. Never for flats and houses together, and never for land, commercial property or short lets.
  • The sale figure and the matching rent figure must both be published single-month figures with at least 100 listings each. If either side falls short, there is no yield.
  • We publish yields of 1–12%, judged before rounding. Outside that range, a mismatch between the sale and rent listings is more likely than a real yield, so we hold it back.
  • It is a gross yield: before service charge, maintenance, agency fees, empty months and tax. Furnished and serviced lets are included in the rent side.
  • Flats and houses rent and sell at different ratios, so a yield for the two together would move whenever the mix of listings changed, even if no price or rent did.
  • Each release stores its yields when it is published, and they never change afterwards. The only exception is a correction, which is dated and shows the old and new yield with the reason.

Changes to how yields are worked out

No changes yet. A change applies only to releases published after it, and will be listed here with its date.

10

How to cite

Quote the series, the month and the link to that month, so readers find the exact figure even after newer releases:

Nigeria Property Centre Rent Index, September 2026 release, Nigeria Property Centre, https://nigeriapropertycentre.com/market-trends/average-prices#release-2026-09

11

Licence

The figures are free to quote and republish under CC BY 4.0, as long as you credit Nigeria Property Centre and link back to the page you took them from. Listing-level data is not part of the licence. See our terms of use.

Get each release by email.

The Rent, House Price and Short-let Index on release day, on the 3rd working day of each month at 09:00. One email a month.

We use your email only to send you the Rent, House Price and Short-let Index and, in it, links to our other market reports and data. See our privacy policy.