A claim about Overwater

Spring outsells winter by more than half again. February sags.

In every spring after the first full year, March to May daily revenue is at least 1.6 times the preceding December to February; the first year is the launch ramp and is flatter. February orders run at most 0.8 times the monthly average.

does not hold on the current window

as of moments ago

Monthly sales seasonality

SELECT strftime('%Y-%m', ordered_at)     AS month,
       COUNT(*)                           AS orders,
       ROUND(SUM(total_usd))              AS revenue,
       ROUND(AVG(total_usd), 2)           AS aov
FROM orders
GROUP BY 1
ORDER BY 1;

View data table

month	orders	revenue	aov
2023-06	10	750	75
2023-07	42	3120	74.29
2023-08	45	3757	83.49
2023-09	70	5183	74.04
2023-10	70	5883	84.04
2023-11	159	13148	82.69
2023-12	277	20740	74.87
2024-01	153	12700	83.01
2024-02	106	8553	80.69
2024-03	571	47532	83.24
2024-04	634	49436	77.97
2024-05	789	66869	84.75
2024-06	393	33220	84.53
2024-07	419	34608	82.6
2024-08	342	29382	85.91
2024-09	286	25822	90.29
2024-10	262	23624	90.17
2024-11	329	29957	91.05
2024-12	343	29616	86.34
2025-01	94	8587	91.35
2025-02	61	4363	71.52
2025-03	442	38296	86.64
2025-04	643	59202	92.07
2025-05	782	73082	93.46
2025-06	299	31363	104.89
2025-07	273	25771	94.4
2025-08	261	24799	95.02
2025-09	245	21986	89.74
2025-10	331	34514	104.27
2025-11	577	56346	97.65
2025-12	1100	106944	97.22
2026-01	292	29051	99.49
2026-02	248	27427	110.59
2026-03	1496	148613	99.34
2026-04	1757	176117	100.24
2026-05	1720	169670	98.65
2026-06	711	78230	110.03
2026-07	781	87438	111.96
2026-08	590	62807	106.45
2026-09	563	59694	106.03
2026-10	589	62467	106.06
2026-11	95	10242	107.81

holds on the current window

Spring over winter by year

WITH daily AS (
  SELECT date(ordered_at) AS day,
         CAST(strftime('%Y', ordered_at) AS INTEGER) AS year,
         CAST(strftime('%m', ordered_at) AS INTEGER) AS month,
         SUM(total_usd) AS revenue
  FROM orders GROUP BY 1
)
SELECT year,
       ROUND(AVG(CASE WHEN month IN (3, 4, 5)  THEN revenue END))  AS spring_daily_revenue,
       ROUND(AVG(CASE WHEN month IN (12, 1, 2) THEN revenue END))  AS winter_daily_revenue,
       ROUND(AVG(CASE WHEN month IN (3, 4, 5)  THEN revenue END)
           / AVG(CASE WHEN month IN (12, 1, 2) THEN revenue END), 2) AS spring_over_winter
FROM daily
GROUP BY year
ORDER BY year;

View data table

year	spring_daily_revenue	winter_daily_revenue	spring_over_winter
2023	null	669	null
2024	1781	559	3.19
2025	1854	1394	1.33
2026	5374	957	5.61

does not hold on the current window column "spring_over_winter" year 2025 value 1.33 is below 1.6

House plants are a spring purchase and a gift purchase. Overwater’s acquisition and repeat-visit rates follow a two-peak year. March to May is the main peak. November and December add a smaller gifting bump, and January and February compound into the trough. Plant SKUs spike in spring and gift-friendly accessories in December. The product mix moves with the volume. On top of that the business drifts, so year two is not year one scaled.

What you should see

A monthly series with a visible hump every spring and a dip every February. The spring_over_winter ratio is above 1.6 in every full year after the first. In the launch year the store is still ramping, so its first spring is compared with a winter when it barely existed and the ratio is close to 1. The later peaks are not the same height from year to year. That is the business drifting.

Run it yourself

SQLite one file, ready to query CSV tar.gz, one file per table