Something happens to Saudi customer reviews during Ramadan, and it is not what the raw numbers first suggest.
Averages across a whole corpus are easy to produce and easy to get wrong. If the businesses being reviewed in February are not the same businesses being reviewed in March, any difference between those months may be telling you which businesses were measured rather than how customers felt. So we did this the harder way: we took 152 Saudi businesses that each had at least ten written Google reviews in all three periods, and compared every business against its own earlier self.
Measured that way, ratings fall by 0.19 stars during Ramadan.
The three windows
Ramadan 1447 ran from 18 February to 19 March 2026 by the Umm al-Qura calendar, with Eid al-Fitr on 20 March. We compared it against the 30 days immediately before and the 30 days immediately after.
| Window | Dates | Reviews | Average rating |
|---|---|---|---|
| Before | 19 Jan - 18 Feb | 12,052 | 4.19 |
| Ramadan | 18 Feb - 20 Mar | 9,802 | 4.05 |
| After | 20 Mar - 19 Apr | 17,279 | 4.15 |
Those pooled figures show a dip of 0.14 stars. The within-business comparison - every business against itself - shows 0.19. The effect is slightly larger once you remove the noise of which businesses happened to be measured when, which is the opposite of what you would expect if the pattern were an artifact.
Eid week is its own weather
The recovery looked weak at first: just 0.05 stars back. Then we noticed the problem. Our "after" window began on Eid al-Fitr itself.
Eid is not a return to normal. It is a distinct period with its own crowds, its own pressures and, it turns out, its own ratings. Excluding the first seven days after Eid, the post-Ramadan average rises from 4.15 to 4.20, and the within-business recovery doubles to 0.09 stars.
So the shape is not a dip and a bounce. It is a dip, then a second softer dip across Eid week, then a recovery.
Where it happens
The national figure hides real variation between cities. Sample sizes differ, and we show them, because a difference measured across 50 businesses deserves more confidence than one measured across 23.
| City | Ramadan change | Businesses | After Ramadan |
|---|---|---|---|
| Jeddah | -0.31 | 50 | recovers strongly, +0.19 |
| Mecca | -0.21 | 25 | does not recover (provisional) |
| Riyadh | -0.15 | 54 | recovers, +0.12 excluding Eid week |
| Dammam | -0.04 | 23 | no detectable change |
Jeddah shows the sharpest swing and the cleanest recovery: a hard dip during the month, then a return above where it started. Riyadh follows the same pattern more gently.
Dammam barely moves. With 23 businesses we would not detect a small effect even if one existed, so the honest statement is that we found no change in Dammam, not that there is none.
Mecca looks different, and we are saying so carefully
Mecca's businesses dip during Ramadan like everywhere else. Unlike everywhere else, they do not come back - the measurement keeps drifting downward after Eid.
That is a plausible story. Ramadan in Mecca runs into the heaviest Umrah season of the year, and the pressure on restaurants, hotels and cafés does not end when the month does. But our Mecca sample is 25 businesses, and the continued decline after Eid is small relative to what 25 businesses can reliably show. We are reporting it because it is interesting and because the dip itself is solid. We are flagging the sample size because the non-recovery is not yet a finding - it is a direction worth watching.
We will revisit it with next year's data.
What this means if you run a Saudi business
Your rating is likely to fall during Ramadan, and that is not necessarily a verdict on your service. Your competitors' ratings fall too. A 4.3 in Ramadan may represent the same quality as a 4.5 in January.
Three practical consequences:
Do not panic in week two. If your rating slips in Ramadan, check whether it slipped further than the city's pattern before concluding something is wrong.
Eid week is not the recovery. It is a second pressure period. Ratings in the week after Eid run below the weeks that follow it. Plan staffing accordingly.
Compare yourself to yourself. The single most useful thing in this analysis was comparing each business to its own earlier record. A rating is only meaningful against a baseline, and the best baseline is you, last season.
Why might this happen?
We can measure the dip. We cannot, from review text alone, prove what causes it. What follows is reasoning, clearly labelled as such, not a finding.
The most likely explanation is that Ramadan compresses demand rather than reducing it. A restaurant that served customers across eight hours now serves most of them in the ninety minutes around iftar. The same kitchen, the same staff, a far sharper peak. Waits get longer, orders arrive slower, and the experience degrades even where nothing about the business has changed.
Opening hours shift too. Many businesses close during fasting hours and reopen late, which narrows the window in which anything can go right and concentrates the window in which things can go wrong. Staffing changes during the month in many sectors, and customers themselves arrive hungrier, more tired and less patient than in January.
Two observations fit this reading. The dip is sharpest in Jeddah, the densest of the four cities in our data. And the week after Eid — when families travel, businesses run on holiday staffing and crowds have not yet dispersed — rates lower than the weeks that follow it, which is what you would expect if pressure, not quality, were driving the number.
None of this is tested here. We note it because the alternative reading — that service quality genuinely collapses nationwide for thirty days each year and then recovers — seems less likely than the simpler one, that the same business under heavier load produces a worse hour for the customer who happens to walk in during it.
Method
This analysis covers 362,938 written Google reviews across Jeddah, Riyadh, Dammam and Mecca. For the three windows examined here we pulled every review in the period - 44,597 rows, verified at 100% coverage against the head-count, using keyset pagination rather than a capped query.
The primary design is within-business: 152 businesses with at least ten reviews in each of the three windows, each compared against itself. Of the roughly 1,400 businesses reviewed in any one window, 1,101 appear in all three, so these are largely the same businesses across time.
Ramadan dates follow the Umm al-Qura calendar used officially in Saudi Arabia.
Two things we deliberately did not claim. We make no claim about how many reviews people write during Ramadan: our corpus grew substantially across 2026 as more businesses were added, so monthly review counts reflect when a business entered our data as much as when customers wrote. And we make no claim about which business category is most affected - we can see that restaurants rate lower than gyms during Ramadan, but restaurants rate lower than gyms generally, and we have not yet isolated the Ramadan-specific change by category.