Blog, page 44

Field notes — one a day since May 2023. Things we shipped, things we broke, and numbers we went and checked.

2026 2025 2024 2023

1,245 posts. , or browse by year and tag below.

Nov 25, 2023

r = 0.25 between year and gdp_per_capita_usd

A measured correlation of 0.25 between two columns, what its square says, and the diagnostic that costs one line.

statistics Statistics

Nov 24, 2023

What week12_score actually contains in clinical-trial

900 rows, 434 distinct values, and the one fact about week12_score that changes how you query it.

quality practice Data quality

Nov 23, 2023

Cutting a lesson down — Communicating a result to someone who will not read your notebook

The analysis is not finished when the number is correct. It is finished when the right person understands it well enough to act.

practice Visualization

Nov 22, 2023

Counting rows is not testing grain

4,000 distinct combinations against 4,000 rows, and what that answers about the table's key.

sql quality warehousing Analytics engineering

Nov 21, 2023

What converted actually contains in ab-test-checkout

6,000 rows, 2 distinct values, and the one fact about converted that changes how you query it.

quality practice Data quality

Nov 20, 2023

mrr_usd by industry: a 49% spread

Real group means for one column across 8 segments, and what the pooled average conceals.

statistics practice Analytics practice

Nov 19, 2023

Is trip_id + pickup_at the grain of ride-hail-trips?

8,000 distinct combinations against 8,000 rows, and what that answers about the table's key.

sql quality warehousing Analytics engineering

Nov 18, 2023

What priority actually contains in support-tickets

4,000 rows, 4 distinct values, and the one fact about priority that changes how you query it.

quality practice Data quality

Nov 17, 2023

A essentially none correlation, and what it is not

A measured correlation of -0.01 between two columns, what its square says, and the diagnostic that costs one line.

statistics Statistics

Nov 16, 2023

Reader question — Uncertainty, sampling, and how much to trust a number

Every number computed from a sample is wrong. The useful question is by how much, and the machinery for answering it is smaller than you think.

practice Statistics

Nov 15, 2023

Five minutes with sessions in ab-test-checkout

6,000 rows, 20 distinct values, and the one fact about sessions that changes how you query it.

quality practice Data quality

Nov 14, 2023

The overall average hides 5 different numbers

Real group means for one column across 5 segments, and what the pooled average conceals.

statistics practice Analytics practice

Nov 13, 2023

r = 0.98 between pm25 and pm10

A measured correlation of 0.98 between two columns, what its square says, and the diagnostic that costs one line.

statistics Statistics

Nov 12, 2023

Reading returned before trusting it in retail-orders

7,000 rows, 2 distinct values, and the one fact about returned that changes how you query it.

quality practice Data quality

Nov 11, 2023

humidity_pct by site: a 0% spread

Real group means for one column across 3 segments, and what the pooled average conceals.

statistics practice Analytics practice

Nov 10, 2023

Cutting a lesson down — Frame the problem, then beat a stupid baseline

Most model projects fail at framing, not at modelling. The second most common failure is not knowing what "good" would have been without a model at all.

practice Machine learning

Nov 9, 2023

Five minutes with revenue_usd in ab-test-checkout

6,000 rows, 603 distinct values, and the one fact about revenue_usd that changes how you query it.

quality practice Data quality

Nov 8, 2023

Counting rows is not testing grain

9,000 distinct combinations against 9,000 rows, and what that answers about the table's key.

sql quality warehousing Analytics engineering

Nov 7, 2023

temp_c by status: a 26% spread

Real group means for one column across 3 segments, and what the pooled average conceals.

statistics practice Analytics practice

Nov 6, 2023

Five minutes with posted_on in data-job-postings

3,500 rows, 966 distinct values, and the one fact about posted_on that changes how you query it.

quality practice Data quality

Nov 5, 2023

r = 0.94 between gdp_per_capita_usd and co2_tonnes_per_capita

A measured correlation of 0.94 between two columns, what its square says, and the diagnostic that costs one line.

statistics Statistics

Nov 4, 2023

Cutting a lesson down — When not to build a model

The most valuable modelling skill is recognising the problems that do not need one, which is a large majority of the problems people bring you.

practice Machine learning

Nov 3, 2023

Five minutes with status in sensor-telemetry

8,000 rows, 3 distinct values, and the one fact about status that changes how you query it.

quality practice Data quality

Nov 2, 2023

Is order_id + channel the grain of retail-orders?

4,169 distinct combinations against 7,000 rows, and what that answers about the table's key.

sql quality warehousing Analytics engineering