Digikala’s Baskets Have Shrunk for Four Years. Its Report Says They Grew.

Digikala's annual report for the year to 20 March 2026 says, in words, that items per order has risen significantly, from small baskets to multi-item baskets. The chart printed on the same page shows items per order peaking at 5.0 and falling in each of the last four years, to 3.8. The report's own headline growth figures point the same way as the chart: value of goods sold up 56 per cent against average price up 74 per cent implies roughly 10 per cent fewer items sold.

Digikala’s annual report for the year to 20 March 2026 contains a sentence and a chart that disagree. The sentence says the average Digikala basket has been getting bigger. The chart, printed two pages later in the same chapter, shows it getting smaller in every one of the last four years.

Both are the company’s own. Nothing in this note needs a source Tehran Index does not hold, and nothing in it is our opinion set against the company’s. It is the company’s chart set against the company’s sentence.

What the report says in words

Page 38 is headed “cheaper goods, higher price”. It says that the average value of an order has trended down over twelve years, that the average price per item fell and then stabilised, and then this:

تعداد کالا در هر سفارش افزایش معناداری را تجربه کرده و از سبدهای کوچک به سبدهای چندقلمی‌تر حرکت کرده است

Items per order has experienced a significant increase, moving from small baskets to multi-item baskets.

Digikala Group annual report, year to 20 March 2026, page 38

Page 40 closes the argument. The report dates the break to the currency shock of the fiscal year ending March 2019, says that shock changed how people shop for good, and says that from then on the value of each order fell while the number of items rose. Its last line:

مردم کمتر نخریدند، بلکه ارزان‌تر خریدند.

People did not buy less; they bought cheaper.

Digikala Group annual report, year to 20 March 2026, page 40

What the chart on the same page shows

Directly above that sentence, page 40 prints the series it is describing: average items per order, one figure per year, thirteen years. It rises from 1.5 to a peak of 5.0, holds at 5.0 for two years, and then falls. 4.6, 4.5, 4.2, 3.8.

The report’s sentences are an accurate description of the chart up to the peak, in the years to March 2021 and March 2022. They are not a description of the last four, and the report does not mention the last four at all.

Tehran Index derivation: the latest figure, 3.8 items per order in the year to 20 March 2026, is 1.2 items below the peak of 5.0, which on a base of 5.0 is 24 per cent below the peak. The report prints the two figures; it does not print the fall.

A second line of evidence, from the headline figures

The chart is not the only thing in the report pointing this way. Page 16 prints growth in the value of goods sold in the latest year at 56 per cent. Page 17 prints growth in the average price of goods sold across the whole of Digikala, including the supermarket, at 74 per cent. Both are nominal, in toman, against the year before.

Tehran Index derivation: average price is total value divided by the number of items, so the number of items moved by 1.56 divided by 1.74, which is 0.897. That is roughly 10 per cent fewer items sold in the year to 20 March 2026 than in the year before. The report prints no unit count and makes no such statement. The arithmetic holds only if “average price of goods sold” means value over units, which is the ordinary meaning and the only one on which the two figures can be compared.

We lead on the chart rather than on this, deliberately. A derivation invites an argument about definitions. A company’s own chart invites none. But two independent things in one report point the same way: the basket got smaller, and the number of items sold got smaller.

Why this matters to a reader outside Iran

English-language coverage of Iranian e-commerce is almost entirely nominal growth in toman, and nominal toman flatters everything. A reader who sees 56 per cent growth in the value of goods sold reads expansion. The same report, read to the chart, describes a platform selling fewer things at much higher prices, to customers putting fewer items in each order than they did four years ago.

For scale, the same report puts Digikala at 1.8 per cent of all Iranian retail, on page 15. That is a share of total retail, not of online retail. Growth in a platform of that size says as much about prices as it does about the platform.

None of this says the company is doing badly. It says the growth figure and the volume figure are different measurements, and that the report’s narrative describes the decade rather than the year.

What we do not know

The report does not say why baskets shrank, and neither do we. It does not define “items”: whether the count is units or distinct products, and whether supermarket orders, which may carry more items each, are in the series on the same basis every year. A change in that mix, or in how orders are split for delivery, would move the series without any change in how people shop, and nothing in the report lets us rule that out.

The chart’s heading says twelve years and it prints thirteen points; we publish the thirteen as printed. The data labels were read from the PDF by position, and the digits in this document run left to right; read the other way, one label comes out as 6.4 where the chart shows 4.6. Our reading is checked against the heights of the bars and shows no conflict across all thirteen labels, but it is a reading of a chart, not a table the company published.

The 10 per cent figure assumes the 56 and 74 per cent growth rates cover the same scope. The report also prints average price growth of 72 per cent excluding the supermarket and 55 per cent for supermarket goods alone; the 56 per cent carries no qualifier, so we paired it with the whole-platform 74. The report’s order-value and item-price series on pages 38 and 39 are denominated in dollars and are not republished here, nor any exchange rate implied by them. Every figure in this note is a count or a ratio of two toman growth rates, and no exchange rate enters it.

The company’s sourced figures, with their periods and evidence, sit on the Digikala record.

Cite as: Tehran Index, “Digikala’s Baskets Have Shrunk for Four Years. Its Report Says They Grew.”, 24 September 2026. Items per order, value of goods sold, average price of goods sold and share of retail from the Digikala Group annual report for the fiscal year to 20 March 2026, pages 15, 16, 17, 38 and 40, read directly. The fall from the peak and the implied change in items sold are Tehran Index derivations. tehranindex.com