Seasonality or a store problem? How to tell the difference

Store owner planning the season with a calendar and laptop

On Monday morning the panel shows 41 orders for the whole week. The week before, it was 63. Your first thought is a discount, your second is “probably the season, it was the same last year”, and both can be wrong, because a seasonal dip and a broken payment look exactly the same on a chart. Before you cut a price, check the drop step by step, and only then plan what to do with the quieter weeks.

How to recognise a seasonal sales dip

Fewer orders than last year is not enough for a diagnosis. A drop in the same weeks as always is a hint, but something entirely different can coincide with it, for example a payment gateway that has been rejecting some cards since Wednesday, or a bestseller that has been out of stock for ten days. That is why the season is the last diagnosis, once everything else has been ruled out.

The order that saves the most nerves looks like this:

  • The number of visits, and with a drop also the traffic sources and campaigns, because one may have ended or run out of budget.
  • A full purchase on a phone, all the way to the end of payment, if traffic is similar and orders are fewer.
  • Availability of the products you are currently advertising.
  • Finally the year-on-year comparison, together with the calendar of your promotions and mailings, which may have shifted sales between weeks.

Year-on-year comparison has its traps

Last year's chart shows the sales, but not what you did to generate them. If a promotion landed in the third week of the month last year and you moved it two weeks later this year, a week-to-week comparison will show a dip that does not exist in demand, so keep the calendar of campaigns next to the chart.

Also check that you are comparing like with like. A report for the last 30 days and a report for a calendar month cover different periods, and a change of analytics tool or of the way visits are counted can shift the numbers without any change on the customers' side.

Your season does not have to match anyone else's. If you sell school backpacks, the turn of August and September is your peak, and your quiet period falls somewhere else entirely. Plan the work from your own statistics.

A discount will not tell you what happened

A price cut made on the first reflex has one big downside. It hides the cause. If payment fails for some cards or the bestseller is out of stock, a lower price fixes nothing and only eats your margin on customers who would have bought anyway. The season stays as an explanation only once payment works and the goods are in stock.

What to fix in the store outside the season

The best material is in your inbox. Go through customer questions from recent months and pick a topic that keeps coming back on a product you plan to promote at the peak. Fix that product's description and photos, and confirm any missing data with the supplier before you enter it. The whole catalog can wait.

Write the task so that you can check whether it is finished. “Add a photo of the inside of the bag and the dimensions of the main compartment” can be ticked off in one afternoon, while “improve the descriptions” drags on for weeks and you never know if it is done.

It is also better to launch a bot in a quiet month, when there is time for adjustments. Test it on matters customers really come with, such as whether an accessory fits a given device model. A few good answers are a good sign. You cannot conclude from them that the bot will cope with every December question.

Contacts left over from earlier months

A quieter month tempts you to write to everyone who once left a contact. If someone asked to be told when a product returns, first check which product the request concerned and whether your store system supports such notifications at all. Addresses collected in conversations are not yet a notification list.

The autumn campaign is simpler, if less convenient, because Poland's Electronic Communications Law requires prior consent to receive commercial messages. A customer who left an email to get an answer about a dimension receives the checked dimension and nothing else until they agree to promotions.

A sample two-week plan

In the first week you collect the gaps, in the second you make the fixes. Start with a few product pages for which you already have the materials, and for each task write down who does it and when you will check the result together.

Before you change anything, note the starting point. For a description it will be the question that kept coming back and the number of page views, for the checkout it will be the result of a test order. After the change, repeat the check.

Do not try to calculate in advance how much this work will earn in December, because nobody can do that honestly. Count instead which question stopped coming back. You can pick the first task today, by opening your inbox and looking for a question you received at least three times this month.

Try CartBot in your store

Quote within 24 h. No commitment, no card.