The January problem: filling the quietest month
January is the lowest month of the year for food and drink in Great Britain, and the hole is deeper than the December peak is tall. Plan for it in November.

In December 2025, food and beverage service across Great Britain took £10.34bn. In January 2026 it took £8.20bn. That is a fall of 21% in a single month, and it happens on roughly the same schedule every year.
Most advice about January arrives in January, which is too late to act on. By then the month is largely written. The decisions that change it get made while the shop is still full.
Key takeaways
- January was the lowest-turnover month of the year for food and beverage service in each of the last three years, running 15 to 18% below the annual average.
- It is not only the December hangover. The January hole is deeper than the December peak is tall, and February is the second-worst month, so the trough runs about nine weeks.
- Seasonally adjusted figures make January disappear completely. That is a statistical convention, not a reprieve.
- A quiet Tuesday is an attention problem. A quiet January is a demand problem, and they do not respond to the same things.
- The work that matters happens in November and December: collecting a way to reach people while they are standing in front of you.
How bad January actually is
The Office for National Statistics publishes monthly turnover for food and beverage serving services in Great Britain, at current prices and without seasonal adjustment. It is the closest thing to a public record of what this trade actually banks each month.
Here is 2025, with each month shown against that year’s own average.
January has been the lowest month of the year in nine of the last twelve years. In two of the exceptions, February took the title by a narrow margin. In the third, April 2020, the sector was closed.
For the three most recent complete years the pattern is consistent. January came in at 84.4% of the annual average in 2023, 84.9% in 2024 and 82.3% in 2025.
It is not only the December hangover
The obvious objection is that January only looks bad because December is abnormal. Compare January against a normal month and perhaps there is nothing here.
The figures do not support that. In 2025 January sat 17.7% below the annual average while December sat 14.6% above it. The hole is deeper than the peak is tall, and the same held in 2023 and 2024.
February is the second-lowest month every time. So the shape is not a single bad month bracketed by normal ones. It is a trough of about nine weeks, and a shop that plans only for January has planned for half of it.
Why the official figures say there is no problem
This is worth knowing before somebody shows you a chart that contradicts everything above.
The ONS also publishes a seasonally adjusted index for the same sector. On that measure January 2026 came in at 102.0, marginally ahead of February at 100.8 and July at 99.1. Read that series and January does not look like a problem at all.
Both numbers are correct. Seasonal adjustment exists to strip out the repeating annual pattern so that underlying trends are visible, which is exactly the right thing to do if you are measuring the economy.
It is exactly the wrong thing if you are paying rent. Seasonal adjustment removes the part that hurts you. The unadjusted series is the one that matches your bank account.
A quiet January is not a quiet Tuesday
The quiet Tuesday problem covers moving demand from a saturated hour into an empty one. Some of that transfers here, but not all of it, and the difference decides what works.
A quiet Tuesday is an attention problem. The customer would happily come in, they have the money, and they simply are not thinking about you at three o’clock. A well-timed message solves a surprisingly large part of that.
January is a demand problem. The same customer has less money, has often decided to spend less on purpose, and in some cases has given up the exact thing you sell.
Three things are genuinely different:
The budget is smaller, and the reason is recent. December is expensive and the bill arrives in January. This is not a mood, and no notification changes it.
Some customers have made a rule. Somewhere between a tenth and a third of UK adults tell pollsters in November that they intend to take January off alcohol. Alcohol Change UK, which runs the Dry January campaign, put it at 32% for 2026 on Censuswide fieldwork of two samples of 2,000 adults. YouGov, asking independently in the same period, got 10%.
Those two numbers are measuring stated intention in November, which is the cheapest month of the year to intend something. Registrations with the campaign are much smaller and are not growing: 175,000 in 2023, 215,000 in 2024 and 200,000 in 2025. They undercount heavily, since most people who cut back never register, so treat them as a floor rather than as the answer.
The planning position that survives all three figures: expect a real dent in wet sales, not the disappearance of a third of your trade, and have something worth ordering that is not alcohol.
Your costs do not fall with your takings. Rent and staffing carry on into the quietest weeks of the year, and heating a room in January costs more than heating it in September. That is why January shows up as a cash problem before it shows up as a customer problem.
The work happens in November and December
Here is the part that makes this a September and October article rather than a January one.
The single most valuable thing you can do about January is collect a way to reach people during the weeks when your shop is full of them. December puts strangers in front of you in volume, many of them once a year. In January you need somebody to talk to, and the list is closed by then.
| When | What to do | Why then |
|---|---|---|
| October | Decide the gift voucher offer and get it produced | Print lead times are the part that slips |
| November and December | Sign up every new face at the counter | This is the only month they are standing there |
| Late December | Write the January messages | You will not want to in January |
| Early January | Send to the people who joined in December | They are the warmest names you have |
What to put next to the till covers the counter ask itself, which is the whole of step two.
Gift vouchers deserve their own mention here, because they are the one instrument that moves money from December into January directly. Somebody pays in December and the visit happens later, which is the cash flow most shops want. Gift vouchers before Christmas covers what to sell and when to stop taking orders.
One caution on that. There is no published UK evidence that gift instruments move hospitality demand between months in aggregate, so treat this as arithmetic about your own till rather than a proven market effect.

Photo by Jakob Cotton on Unsplash.
What to send, and when to close it
A January message has a harder job than a Tuesday one, because it is asking someone to spend in the month they decided not to.
That argues for giving a reason rather than a discount, and for making the reason time-bound.
- Something genuinely new. A January menu, a winter drink, a quiet-hours service you do not run in December. New beats cheap, because new is a reason rather than a concession.
- A reward they already earned in December. The customer who collected four stamps over Christmas has something sitting on their phone. Telling them so costs you nothing and is not a discount.
- A reason to come in a named week. Not “January offers,” but a date.
On that last point there is useful evidence from an unlikely place. Christoph Thommen and Beat Hintermann ran a field experiment on Swiss rail fares. Moving the deadline for discounted tickets forward, to midnight the day before travel, cut discounted sales by 18 to 30%.
In their estimate, that single change to the deadline was equivalent to raising the price by 26 to 43%.
It is rail, not hospitality, and note which way the effect runs. A tighter deadline did not drive more take-up. It suppressed take-up, by a margin as large as a substantial price rise.
That is still the argument for a named date, just not the romantic one. The deadline moves behavior about as much as the price does, so a short window is how you stop an offer being quietly collected by the regulars who were coming anyway. An offer that runs all January will be taken up more, and much of that extra take-up is margin you did not need to give away.
What does not work
A general January sale. It discounts the customers who were coming anyway and trains everyone else to wait. The case against this is in the quiet Tuesday problem.
Messaging everyone more often because it is quiet. The fortnight ceiling does not relax because your takings did. A quiet month is exactly when an irritated customer deletes the pass.
Waiting until January to start. The month is nine weeks long counting February, and the list you message in it is the list you built in December.
Frequently asked questions
Does the January drop apply to salons and barbers too? Probably, but we cannot show it. The ONS monthly turnover series covers food and beverage service, and there is no equivalent published monthly series for personal care, so any claim about a salon’s January is owner experience rather than evidence. The planning advice is the same either way.
Should I close for part of January instead? It is a legitimate answer, and for some trades a better one than discounting. Reduced hours in the two worst weeks can beat staying open at a loss. Decide it in November so staff have notice, and tell customers before they find a locked door.
Is it worth running a January offer at all if margins are already thin? Only if it moves someone who was not coming. An offer taken up mainly by regulars is a cut to the small amount of trade you still had. Aim it at the people who joined your card in December and have not been back since.
When should the first January message go out? Early, and to the newest names. The people who joined in December have not formed a habit with you yet, so there is nothing holding them beyond the visit that brought them in. That makes them the group most worth spending your first message on.
Where to go next
January is the seasonal version of a problem this blog covers at a smaller scale in the quiet Tuesday problem. The instrument that moves December money into January is covered in gift vouchers before Christmas, and the counter habit that fills the list is what to put next to the till.
For the messages themselves, win-back messages that actually bring people back covers the customers who drifted, and automated loyalty campaigns covers setting them to run without anybody remembering.
Collecting those contact details at the counter in December, and being able to reach them by push, email or SMS in January, is what Passumo is for. The channels page has what each one costs.
Sources: ONS series JQ37, “GBServTO: 56 - Food & beverage serving services TOTAL”, Great Britain, monthly, current prices, not seasonally adjusted; monthly percentages calculated from the published values. Seasonally adjusted comparison from ONS series S2LA, Index of Services, food and beverage service activities. Dry January intention figures from Alcohol Change UK (Censuswide, two samples of 2,000 UK adults, November 2025, measuring stated intention) and YouGov, with registration counts from Alcohol Change UK’s own campaign history. Christoph Thommen and Beat Hintermann, “Price versus Commitment: Managing the demand for off-peak train tickets in a field experiment”, Transportation Research Part A: Policy and Practice 174, 2023, article 103691; Swiss rail, cited for the effect of a deadline rather than as hospitality evidence. No published UK series exists for monthly personal care turnover, and no UK evidence was found that gift instruments shift hospitality demand between months.