Rain or a real problem? Reading outlet numbers in monsoon season

When a restaurant sales drop lands in monsoon season, the first question is whether the rain caused it or simply arrived at the same time. Compare each outlet with its own normal for that weekday and daypart, then with neighbouring outlets on the same day. If every outlet nearby fell by a similar amount, it was probably the weather. If one outlet fell well beyond its neighbours, something inside that outlet needs a look.

The test matters more than usual this year. The northeast monsoon (musim tengkujuh) began on 27 October and is expected to continue until March 2027, Bernama reported on 1 November. Heavy rain is expected on Kelantan, Terengganu and northern Kedah in November and December, then on Johor and Sarawak in January and February, and afternoon downpours have already caused flash floods in Kuala Lumpur. That is five months in which “it was the rain” is a believable explanation for almost any weak number. Believable explanations are the ones nobody checks.

Where flooding affects staff or premises, safety comes first, and keeping every outlet stocked when the roads flood covers the supply side. This piece is about the ordinary wet days in between, when outlets are open, numbers are down and someone has to decide whether to act.

What is the difference between a weather dip and an outlet problem?

A weather dip is a fall in sales or footfall that hits every outlet under the same rain at about the same time and by about the same amount, and recovers when the rain stops. An outlet problem is a fall that is larger, longer or differently shaped than the neighbours’ fall, which points to a cause inside that outlet: an item sold out, a shift ran short, a fryer was down, or complaints are building.

The two often arrive together. A storm at 6pm gives a short-staffed outlet perfect cover: it is down 35%, the area is down 20%, and the outlet’s own 15 points disappear into a sentence about the weather.

Why a restaurant sales drop needs two comparisons

The outlet against its own normal. Compare yesterday with the same weekday and daypart over the last four to six clean weeks, not with the group average. A mall outlet and a street-front shop react to rain differently, and an average blends them into a number that describes neither. Demand forecasting with the data you already have covers how to build that baseline.

The outlet against its neighbours. Group outlets into weather areas: outlets close enough to get the same rain at the same time. Then compare each outlet’s drop with the typical drop in its area that day. The difference is the part the weather doesn’t explain.

Here is how that reads for six hypothetical outlets in one area on a wet Thursday dinner, where the typical drop was 20%.

Outlet	Dinner sales against its own normal	Outlet minus area	Read
Outlet 01	−19%	+1 point	Area-wide: rain
Outlet 02	−22%	−2 points	Area-wide: rain
Outlet 03	−35%	−15 points	Outlet-specific: check
Outlet 04	−18%	+2 points	Area-wide: rain
Outlet 05	−21%	−1 point	Area-wide: rain
Outlet 06	−20%	0 points	Area-wide: rain

Five of the six fell by about the same amount. That is the rain. Outlet 03 fell 15 points further, and the rain doesn’t explain those.

How do you tell what is behind the gap?

Once an outlet stands out, work through the checks the rain can’t fake.

Dayparts. Did the drop sit in the hours it rained, or start at lunch, before the rain? A drop that starts early or outlasts the rain has another cause.

Dine-in against delivery. If dine-in fell across the area but delivery fell only at Outlet 03, look at that outlet’s delivery side: paused on a platform, items marked unavailable, or prep times that keep riders waiting.

Footfall against transactions. If people still came in but fewer bought, the rain didn’t stop them; the queue, the service or the payment did. Footfall against sales at the front of house covers how to read that gap.

Sold-out times and staffing. A best-seller that sold out at 6.30pm looks exactly like a quiet evening in the sales report. So does a shift that ran two people short because commutes took longer in the rain.

Complaints. A rise in reviews or mentions naming one outlet, especially about waiting times or missing items, often explains what the sales data can’t.

Pattern	Likely cause	What to check	Who owns it
Every outlet in the area down by a similar amount, in the rainy hours	Weather	Rain timing; dine-in against delivery	Area manager notes it; no action
One outlet down well beyond its area	Something inside the outlet	Sold-out times, staffing, equipment, complaints	Outlet manager
Delivery down at one outlet only	Platform or kitchen issue	Platform status, unavailable items, prep times	Outlet manager with the delivery lead
Footfall normal, transactions down	Service or conversion	Queue length, staff on the floor, payment problems	Area manager
One or two items missing from the mix	Stock-out or short delivery	Sold-out times; delivered against ordered	Outlet manager with the central kitchen
The drop continues on dry days	Not the weather	Week-on-week by daypart; recent changes	Operations director

Tag weather days before they distort next month

A heavy-rain evening is real data, but it isn’t normal data. Left untagged, it drags the baseline down, and next month’s ordinary Thursday looks like a strong one.

Ask outlet managers to tag the day the same night, with a short code: heavy rain and the hours it fell, early closing, access road flooded, outlet closed. Tag by daypart, because rain at 6pm says nothing about lunch. The tags let you exclude or flag those hours when you build baselines.

By the end of the season, the tags also tell you what a wet evening usually costs each type of outlet, which makes the next unusual dip easier to spot.

A morning routine for the monsoon months

1. Set the weather areas once. List which outlets share the same rain, and revisit the list when outlets open or close.
2. Read yesterday by outlet and daypart against each outlet’s own normal. Start with the dayparts that carry most of the sales.
3. Work out each area’s typical drop. Use the median, not the average, so one bad outlet doesn’t pull the area figure towards itself.
4. Flag outlets that fell further than their area by more than an agreed margin. Start with, say, ten points and adjust once you see how often it fires.
5. Give each flagged outlet one owner and one question. “Outlet 03 was 15 points below the area at dinner. Why?” The owner runs the checks and replies before the lunch rush.
6. Record the cause. Weather, stock, staffing, equipment, complaints or unknown. A month of answers shows which problems the rain keeps hiding.

Most mornings the list is short. The area comparison clears the outlets that only got wet, so attention goes to the ones that need it.

Where the AI-Kitchen Command Center fits

The routine works on a spreadsheet until the checks live in four systems. The AI-Kitchen Command Center brings them into one view. It compares every outlet, product and shift with its own baseline and flags drift, and its live dashboards put sales, stock, footfall, food safety and social signals side by side, from RMS, VisionAI, Adqlo and your POS. The triage checks become a scroll rather than five phone calls.

Its AI agent takes the morning question in plain language: “Which outlets were furthest below their normal at dinner yesterday, and what else changed there?” Answers come from your own records with the underlying rows traceable, and when the data can’t support a conclusion, it says so. It doesn’t hold weather data; the comparison with neighbouring outlets does that work. Alerts can go to each outlet’s owner by WhatsApp, Microsoft Teams, Slack or email. For the wider view, see what a kitchen command center actually shows you.

Questions worth asking this week

• Do we know which of our outlets share the same weather?
• Can we see yesterday by outlet and daypart, against each outlet’s own normal, before 9am?
• When an outlet manager says “it was the rain”, what does anyone check?
• Are heavy-rain hours and closures tagged anywhere we can use later?
• Who owns the call when one outlet falls well beyond its neighbours?

If most of the answers are “not yet”, start with the first two. Five months of monsoon is long enough for the habit to settle in.

Frequently asked questions

Why do restaurant sales drop when it rains?

Restaurant sales can drop when it rains because fewer customers travel to dine in, some outlets become harder to reach and some orders move to delivery. The size of the drop varies by outlet type, location and daypart, so the useful comparison is each outlet against its own normal for that weekday, and against neighbouring outlets on the same day.

How can I tell if a sales drop is caused by the weather?

A sales drop caused by the weather usually appears at every outlet under the same rain, at about the same time and by about the same amount, and it recovers when the rain stops. If one outlet falls well beyond its neighbours, or the drop continues on dry days, check stock-outs, staffing, equipment and complaints at that outlet.

When is the monsoon season in Malaysia?

The northeast monsoon in Malaysia began on 27 October 2026 and is expected to continue until March 2027, Bernama reported. Heavy rain was expected to concentrate on Kelantan, Terengganu and northern Kedah in November and December, then shift to Johor and Sarawak in January and February. Check the Malaysian Meteorological Department (MetMalaysia) for local forecasts.

Should rainy days be excluded from sales baselines?

Rainy days should be tagged, then either excluded from sales baselines or treated separately, because a heavy-rain evening is real data but not normal data. Tag at outlet and daypart level with short codes such as heavy rain, early closing or outlet closed, so the affected hours can be filtered out when baselines and forecasts are built.

What should an outlet manager check when sales fall more than at nearby outlets?

When an outlet's sales fall more than at nearby outlets, the outlet manager should check sold-out times, rostered against actual staff hours, equipment that was down, delivery platform status and unavailable items, and any rise in complaints naming the outlet. Recording the cause each time, even when it is unknown, shows repeat problems within a month.

Can an AI kitchen command center tell a weather dip from an outlet problem?

An AI kitchen command center such as the Codemax AI-Kitchen Command Center compares every outlet, product and shift with its own baseline and flags the ones that drift, with sales, stock, footfall and social signals in one view. It does not hold weather data, so comparing an outlet with its neighbours is what separates the rain from a real problem.

Sources

1. MTL, Super El Niño, Malaysia faces flood and drought risks — Media Selangor (Bernama), 1 Nov 2026