There is a conversation we have every autumn with very different businesses, and it always opens the same way: "we just can't keep up right now."
The "right now" changes with the trade. For anyone selling corporate gifts or branded merchandise, it is November and December, when half the country decides at once what to give clients and staff. For home improvement firms and dental clinics, it is January, the month of resolutions: the house that will finally get sorted, the teeth that will finally get seen to. For renewable installers, the peak follows the campaign rather than the calendar: a grant gets announced, an energy price rise makes the news, and the inbox fills up within a week.
The underlying pattern is the same everywhere: demand in your sector does not arrive as a steady stream. It arrives in waves. And almost no small business staffs for the crest of the wave, because doing so would be absurd — for the rest of the year you would be overstaffed.
The cost of the queue
When the peak arrives, what forms is an invisible queue. Enquiries waiting hours or days for a first reply.
And here is the detail we keep forgetting: the customer does not know it is your high season, and would not care if they did. Their comparison is not "they're replying slower than in March". Their comparison is "this one isn't answering and that one is". The person asking about 200 branded mugs before Christmas has a deadline; they are not going to wait three days for your quote when they can have someone else's this afternoon. The well-known InsideSales research on response times suggested conversion falls by roughly 8× once the first five minutes have passed; you do not need to take the figure as gospel for the direction to be obvious, and it matches what anyone can see in their own pipeline: the customer who gets no answer buys from whoever gives them one.
The cruel part is that the queue eats precisely the weeks that fund the year. You are not losing enquiries on a random Tuesday in May; you are losing them in your best month.
Why the temporary hire never quite adds up
The classic answer is to bring in extra hands for the season. On paper it works; in practice the sums rarely do.
Finding someone for six weeks is hard enough. Training them takes two or three of those weeks — longer in any business with a real catalogue or technical edge cases — so just as they start answering with confidence, the season ends. And the hours where help is needed most — evenings and weekends, when the retail customer actually writes — are exactly the hours nobody wants to cover.
The net result: you pay a spike in cost to cover a spike in demand, and the most valuable part of the spike slips through anyway.
Prepare the channel before the peak
Our way of thinking about this has shifted: the question is not "how many people do I need in December", but "how much of this wave can be handled without people at all".
A surprisingly large share of high-season enquiries is first-conversation work: what options are there, will it arrive in time, roughly what would it cost, what do you need from me to get started. That is work of answering fast, asking the right questions and gathering details — exactly what an AI agent does without breaking stride, whether it is three conversations at once or thirty, ten on a Tuesday morning or eleven on a Sunday night. Whatever genuinely needs human judgement — the negotiation, the odd case, the delicate customer — gets handed over with the full context, so your team spends the peak closing rather than triaging.
There is one condition: the channel has to be built before the wave, with the content loaded and the team used to receiving handovers. Setting up your response capability mid-avalanche is buying the umbrella inside the storm.
And there is a financial wrinkle that high season makes especially visible: if your support tooling charges per conversation, your best month is also your biggest bill — it penalises you exactly when things go well. It is one of the reasons Zatio runs on a flat monthly fee: the December peak costs the same as the August lull. How the structure works is laid out on the pricing page, and how the agent behaves day to day is on the product page.
This year's wave is already forming. The only real decision is whether it finds your channel built — or half-built.