Skip to content
Sign inGet started free

The record · 10 September 2026

Six Gulf markets carry 621 live openings and not one of them can print a median advertised salary.Vienna does it on three job boards.

Fourteen markets, one day, one question a job seeker asks first. This is not a statistic about pay. It is a count of how many adverts state one, and the answer in six markets is fewer than five out of hundreds. Everything below is checkable against the live market.

0 / 6of the six Gulf markets can publish a median advertised salary, including Dubai, which reads twelve job boards
22 / 65of the 65 markets served worldwide do publish one, among them Vienna on three sources and Johannesburg on five
621live openings held across the six Gulf markets on the day of reading, none of which produced five disclosed annual figures in one currency

Twenty-two markets out of sixty-five

Read on 10 September 2026 across every market this platform serves: 22 of the 65 publish a median advertised salary, and 43 do not. That is the ordinary state of a job market seen through its adverts. Most listings, in most places, do not name a figure.

The six markets of the Gulf are not part of the 22. Dubai, Abu Dhabi, Riyadh, Doha, Kuwait City and Muscat carry 621 live openings between them and produce no median at all. Not one of the six reaches five adverts quoting an annual sum in the same currency, out of eighty to a hundred and twenty openings each.

The number that makes this hard to dismiss is Dubai's. It reads twelve job boards and holds 120 live openings, which is more breadth than Vienna, Johannesburg or Tokyo, and all three of those print a figure. Riyadh and Abu Dhabi read nine each, more than Johannesburg's five. Breadth is not what is missing.

Provenance: Source
Live openings, the number of job boards read, and the median advertised salary where the market can produce one, read on 10 September 2026. The six Gulf markets, then the eight markets elsewhere that publish a figure on the fewest sources. The median is of what employers chose to print, never of what anyone is paid.
MarketLive openingsJob boardsMedian advertised
Dubai12012None
Abu Dhabi809None
Riyadh1109None
Doha1183None
Kuwait City1073None
Muscat862None
Vienna1203EUR 54,500
Johannesburg1205ZAR 350,500
Tokyo12011JPY 6,000,000
Warsaw12013PLN 80,000
Mumbai12014INR 750,000
Melbourne12015AUD 152,500
Bangalore12015INR 1,000,000
Barcelona12016EUR 23,000

Vienna, on three boards

The obvious objection to all of this is that the silence is ours: we read fewer boards in the Gulf, so of course we see fewer salaries. It is the right objection, and the table answers it directly.

Vienna holds 120 live openings from three sources and prints EUR 54,500. Doha holds 118 openings from three sources and prints nothing. Same number of boards, same order of listings, opposite outcome. Kuwait City, also on three sources with 107 openings, prints nothing either. Johannesburg manages a figure on five sources; Muscat, on two, does not, and Muscat is the one market here thin enough that the objection still stands against it.

So the silence is not a property of our reach. It is a property of the adverts. Somewhere between the employer and the board, the number is not being written down, and it is not being written down consistently enough across six markets and 621 listings for a single one of them to reach five.

Provenance: SourceWhat the figure asks of a market
  1. Does the advert state a figure, and state it as an annual one? A monthly or hourly figure is refused rather than multiplied, because an advert that says "4,000" without a period does not tell us which.
  2. Is the figure plausible as an annual salary? Anything under a thousand is dropped: a feed that mislabels an hourly rate as yearly would otherwise drag a market's median to the floor.
  3. Do at least five of them quote the same currency? The largest currency group in a market is the one used, and below five values there is no median at all. Five is a low bar, and it is the bar the Gulf does not clear.

Nothing in those three questions asks how much anyone earns. They ask how many employers decided to say. A market that prints no figure has not told us it pays badly, or well, or anything: it has told us that fewer than five of its live adverts named a yearly sum.

A market page prints a median advertised salary only when the adverts can carry one. Three questions decide it, and they are questions about the listings, not about the jobs or the employers.

What the silence does not say

It does not say the Gulf pays badly, or well. No advert on this page reports a wage anyone actually receives; every figure in the table is a median of what employers chose to print, in the markets where enough of them printed anything. A market with no figure has told us nothing about its pay, which is precisely the point being made about it.

It does not name a cause. There are several plausible ones, and this page has measured none of them: the desk reports that the number is absent from the adverts, not why. Anyone offering an explanation here would be adding it, not reading it.

And it is one day. This is a reading of a live supply that changes continuously, not a series with a trend behind it. A market that reaches five disclosed figures tomorrow will print a median tomorrow, and the honest way to check any of this is to open the pages and look.

What the data establishes

  • Read on 10 September 2026, 22 of the 65 markets served publish a median advertised salary and 43 do not.
  • None of the six Gulf markets does. Dubai, Abu Dhabi, Riyadh, Doha, Kuwait City and Muscat hold 621 live openings between them, from 80 to 120 each, and no market among them reaches the five annual figures in one currency a median requires.
  • Dubai reads twelve job boards, Riyadh and Abu Dhabi nine each. Vienna prints EUR 54,500 on three boards, Johannesburg ZAR 350,500 on five and Tokyo JPY 6,000,000 on eleven.
  • Doha holds 118 live openings from three job boards, the same breadth as Vienna and a comparable number of listings, and prints no figure.

What it does not

  • That pay in the Gulf is low, high, or anything at all. Every figure here is a median of what employers chose to advertise; a market with no figure has reported nothing about its pay.
  • That the silence is our coverage. Vienna and Johannesburg print a median on three and five sources, and Dubai prints none on twelve. Muscat, on two sources, is the single market here where the objection still stands, and it is named rather than hidden.
  • Any cause. Several are plausible and none is measured on this page. The desk reports that the figure is absent from the adverts, not why it is absent.
  • That the 22 markets which print a figure are transparent, or that their employers disclose more often than five in a hundred. Five disclosed adverts out of 120 is enough to produce a median, and it is a low bar in every market that clears it.
  • Anything about wages actually paid, anywhere. An advert reports what an employer offered to write down, which is not a payroll.
  • Any trend. This is one reading of a supply that changes continuously, taken on one day, with no earlier reading behind it.

Method

  1. One source, one day, and the source is us. Read on 10 September 2026 from the served market pages at babzituna.com/jobs and the public market index at babzituna.com/api/markets/cities. Live openings, the number of job boards read and the median advertised salary are the figures each page already publishes about itself, in its own structured data, with the caveats attached.
  2. WHY NO OFFICIAL STATISTIC. No labour statistics office publishes 2026 figures for these markets. The ILO's latest readings are 2025 for the United Arab Emirates and Saudi Arabia and 2024 for Qatar; Eurostat's annual series for 2025 was published in June 2026. Labour statistics lag by six to eighteen months everywhere, which is a fact about official statistics and not a failing of any office. What is being advertised today is the only thing about these markets that can be measured today, and it is what this page measures.
  3. HOW A MEDIAN IS COMPUTED, and when it is refused. Only adverts stating an annual figure count; monthly and hourly figures are refused rather than converted. Values below 1,000 are dropped as mislabelled hourly rates. The values are grouped by currency, the largest group is used, and below five values the page prints nothing at all. That threshold is the whole test the Gulf does not pass.
  4. THE OBJECTION WAS TESTED BEFORE THE PIECE WAS WRITTEN. A first draft of this edition was going to be about remote work: Abu Dhabi advertises 34% of its stated listings as remote-eligible, and Doha, Kuwait City and Muscat 0%. It was dropped, because those three are exactly the three markets where we read the fewest boards, and publishing that would have printed our own coverage gap as a fact about those countries. The salary finding survives the same test only because Vienna and Johannesburg print a figure on three and five sources.
  5. The fourteen markets are the six of the Gulf, then the eight elsewhere that publish a median on the fewest job boards. That selection is a rule applied to all 65, chosen because it is the hardest possible comparison for the finding: if breadth were the explanation, the markets with the least of it would be the ones printing nothing.

BabZituna scores every job against your profile on six real dimensions, and publishes what it can and cannot measure. See the live market.

The markets in this piece, live

Earlier editions