One store takes more. Real, or a good fortnight?

Twelve trading days from each of two stores, same format, same city. The gap is $342 a day and the test says it is real, which turns a hunch about a manager into a number about a location.

Operations Starter Statistics free

After you install, this is the model to open.

Are These Two Stores Really Different?

  1. In your spreadsheet, click the Sortia icon in the strip of icons down the right-hand edge. No strip? Click the arrow at the bottom-right to open it. You can also use Extensions, then Sortia, then Open Sortia.
  2. Click Start from a template and put that name in the search box.
  3. Pick the card with that name and click Load this template. It arrives on a new tab with real numbers already in it.

The answer

The gap is real
p = 0.00004 t = 5.09 on 22 degrees of freedom, two-tail
Kingsway takes
$4,098 a day, against $3,755 at Halden Road
Worth a year
$102,750 the $342.50 daily gap over 300 trading days
It is basket, not traffic
$12.95 vs $11.02 per head; Halden Road actually gets more footfall

Twelve comparable trading days at two same-format stores. Kingsway averages $4,097.50 a day and Halden Road averages $3,755.00, a gap of $342.50. Click Run: t comes back at 5.09 on 22 degrees of freedom with a two-tail p-value of 0.0000428, so two stores trading at the same underlying rate would show a gap this wide about four times in a hundred thousand fortnights.

The difference is real, and over a 300-day trading year it is worth about $102,750. Before you decide what it means, read the two footfall columns, because they say the opposite of what you would guess: Halden Road gets more people through the door, 340.7 a day against Kingsway's 316.5. The gap is not traffic, it is basket, $12.95 against $11.02, and the two basket lines beside the data do that arithmetic for you.

That points at range, layout or the till prompt rather than at the location or the manager, and it is a different meeting from the one the takings alone would have started. The test assumes the two stores have similar day-to-day spread, which is worth checking rather than assuming: run F-Test Two-Sample for Variances on the same two ranges.

It returns F of 1.87 with a one-tail p-value of 0.157, comfortably short of the bar, so the assumption holds and the equal-variances test is the right one here. If your own data fails that check, switch to t-Test: Two-Sample Assuming Unequal Variances, which does not need the assumption; on this data it returns the same t of 5.09 on 20 degrees of freedom rather than 22, which is what a passing F-test looks like from the other side.

Twelve days is enough here only because the gap is large. Put the same spread against a gap of $100 a day and t falls to 1.49 with a p-value of 0.15, which is nowhere, so a small difference needs a quarter of trading rather than a fortnight. What the test cannot tell you is whether the days were comparable, so keep weekends out of a weekday comparison and keep a promotion at one store out of it altogether. To use your own stores, paste one store per column and widen both ranges.

The model

It arrives on a tab called Template: Two Stores, One Question, carrying these columns:

  • Trading day
  • Kingsway ($)
  • Halden Road ($)
  • Footfall, Kingsway (visitors)
  • Footfall, Halden Road (visitors)

with the model computed beside the data:

Average basket, Halden Road ($)11.02

Once it is in your sheet

  1. The model arrives with real numbers in it and runs as it stands, so you can press the button first and understand it second.
  2. Change the numbers to yours. The sheet marks which cells are inputs and which hold formulas, and most labels carry a note explaining the row.
  3. Press the run button at the bottom of the panel. It is labeled for the tool you are in, and the result lands on its own tab, with a written reading of it beside the figures.

Never used Google Sheets? Start here goes the whole way, in seven steps, and assumes nothing.