To compare two separate groups, such as baskets in two stores, use a two-sample t-test. Its statistic is the difference in means divided by the standard error of the difference, sqrt(s1²/n1 + s2²/n2).
The classic Student version assumes both groups have the same spread. Welch's version doesn't, and costs almost nothing when the spreads are equal, so it is the safe default: stats.ttest_ind(a, b, equal_var=False). SciPy's default is equal_var=True, so pass it every time.
The order matters for the sign: ttest_ind(south, north) tests south minus north.
Report the difference with a confidence interval, not only the p-value. statsmodels gives it: CompareMeans(DescrStatsW(a), DescrStatsW(b)).tconfint_diff(usevar='unequal').