What the 2020 Ecommerce Data Showed

In March 2020, as the first lockdowns began, a Czech analytics company started publishing a daily tracking study of how the pandemic was affecting online shops in Central and Eastern Europe. The study followed more than 1,200 stores and compared each period with the same period a year earlier. Its category figures were widely repeated at the time, and they are still useful today, provided the reader knows what such a study can and cannot measure.

What the study was

The tracking study was run by MonkeyData, an ecommerce analytics firm, and published on a public platform from late March 2020. It gathered sales data from over 1,200 online shops, most of them small and medium-sized businesses that had been trading for more than two years. The figures were updated every day, which was unusual at a moment when most published statistics arrived months late. The launch and the first findings were reported by Ecommerce News Europe and picked up by other trade outlets.

The method was a year-on-year comparison. For each shop, the study compared a recent period with the same weeks in the previous year, then aggregated the change across the panel. That approach has a real strength and a real weakness. Its strength is that it compares like with like, so a seasonal lull does not look like a collapse. Its weakness is that a disrupted year makes the base itself unusual, and a percentage change against an unusual base is easy to misread.

The categories that grew fastest

The first published findings showed growth concentrated in a small number of categories. Health led by a wide margin, with the study reporting growth of around 206 percent year on year. Building materials followed at roughly 119 percent, children's goods at about 103 percent and chemist's goods at about 101 percent. Each of these makes sense in hindsight. People were buying medicine and supplements, preparing for time at home, keeping children occupied, and stocking the household.

It is worth noticing what the four leaders have in common. They are all categories where the physical shop was either closed, crowded or best avoided, and where the product itself is easy to describe on a screen. A box of vitamins or a bag of building sand does not need to be tried on. Categories like these cross the online threshold easily once a shopper has a reason, which is part of the wider pattern described in how COVID-19 changed online shopping.

The categories that fell

The same study reported declines in garden, clothing and fashion, and in the office, workshop and warehouse segment. The office category is the easiest to explain: if offices and workshops were closed, the businesses that stocked them stopped ordering. Clothing and fashion fell for a different reason. These are categories where fit, feel and the pleasure of browsing matter, and a shopper who cannot try a garment on is slower to commit.

Garden is the most interesting of the three. A garden centre is an outdoor shop, and in several countries it reopened earlier than indoor retail, so the online share of garden spending had less room to grow. The decline is therefore partly a story about where the demand went rather than whether it existed. That is exactly the kind of distinction a percentage on its own cannot make.

What the numbers cannot tell you

A tracking study of this kind sees only the shops in its panel. Sales that moved to a marketplace outside the panel, or to a shop that had no analytics installed, are invisible. A category can therefore look weak in the study while the underlying demand was simply moving somewhere the study could not see. This is not a flaw unique to 2020; it applies to most private panel data.

The second limit is that a year-on-year figure blends two things: a change in demand and a change in where people chose to buy. A shop can grow because more people want its product, or because its competitors closed, and the two call for completely different decisions about stock and staffing. The third limit is the base effect. A category that grew 200 percent in one quarter is not on a path to grow 200 percent again; the first jump is usually a one-time reallocation.

How to read a pandemic percentage today

A useful habit is to ask three questions of any 2020 figure. Was the base normal? The previous year was, so a comparison against it is fairer than a comparison against 2021. Was the category structurally suited to online buying? If yes, the growth is more likely to persist. And did the panel include the places where demand actually moved? If not, the figure describes a subset, not a market.

Applied to the four leaders, the answers suggest that health and chemist's goods were the most durable, because the products are simple to ship and the reason to buy them did not disappear. Building materials and children's goods were partly tied to a period at home and gave back some of the gain later. That sorting is the same exercise carried out more broadly in which pandemic shopping habits stayed.

Why this study still matters

Most of what was published in 2020 arrived too late to guide anyone. This study arrived weekly, which is why it was quoted so often by trade press and by shops trying to decide whether to buy more stock. Its value now is different. It is a rare, dated record of what a large panel of small European shops actually experienced during the disruption, category by category.

Read as a record, it is more useful than read as a forecast. The figures mark a direction: health up sharply, clothing down, offices down, gardens split by reopening rules. The wider economic frame, including the closures and the recovery, is documented by the OECD in its work on COVID-19 and the economy. The magazine returns to the shop floor from here, starting with the shopper whose basket produced every one of those percentages.