Blog, page 39

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

2026 2025 2024 2023

1,201 posts. , or browse by year and tag below.

Feb 9, 2024

rating by genre: a 16% spread

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

statistics practice Analytics practice

Feb 8, 2024

A essentially none correlation, and what it is not

A measured correlation of 0.14 between two columns, what its square says, and the diagnostic that costs one line.

statistics Statistics

Feb 7, 2024

What churned_at actually contains in saas-subscriptions

2,500 rows, 624 distinct values, and the one fact about churned_at that changes how you query it.

quality practice Data quality

Feb 6, 2024

Cutting a lesson down — Turning a request into a question

"Can you pull the numbers for last quarter" is not a question. Converting it into one is the most under-taught skill in the job, and the one that most separates a senior analyst from a fast one.

practice Analytics practice

Feb 5, 2024

Counting rows is not testing grain

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

sql quality warehousing Analytics engineering

Feb 4, 2024

The top of gdp_per_capita_usd is 1.4× the 99th percentile

A max 1.4× the 99th percentile, and the three different things that could be.

statistics quality Statistics

Feb 3, 2024

What seats_now actually contains in saas-subscriptions

2,500 rows, 294 distinct values, and the one fact about seats_now that changes how you query it.

quality practice Data quality

Feb 2, 2024

A proportion with an interval, in short

A percentage without a denominator is not a finding. The Wilson interval behaves sensibly at small n, where the normal approximation falls apart.

statistics experiments Statistics

Feb 1, 2024

The overall average hides 3 different numbers

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

statistics practice Analytics practice

Jan 31, 2024

Five minutes with posting_id in data-job-postings

3,500 rows, all distinct, and the one fact about posting_id that changes how you query it.

quality practice Data quality

Jan 30, 2024

A weak correlation, and what it is not

A measured correlation of 0.26 between two columns, what its square says, and the diagnostic that costs one line.

statistics Statistics

Jan 29, 2024

Reader question — 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

Jan 28, 2024

Five minutes with device in ab-test-checkout

6,000 rows, 3 distinct values, and the one fact about device that changes how you query it.

quality practice Data quality

Jan 27, 2024

Counting rows is not testing grain

5,354 distinct combinations against 9,000 rows, and what that answers about the table's key.

sql quality warehousing Analytics engineering

Jan 26, 2024

sched_dep_hour by dest: a 8% spread

Real group means for one column across 14 segments, and what the pooled average conceals.

statistics practice Analytics practice

Jan 25, 2024

What temp_c actually contains in sensor-telemetry

8,000 rows, 1,414 distinct values, and the one fact about temp_c that changes how you query it.

quality practice Data quality

Jan 24, 2024

r = -0.25 between temp_c and wind_mw

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

statistics Statistics

Jan 23, 2024

Is account_id + industry the grain of saas-subscriptions?

2,500 distinct combinations against 2,500 rows, and what that answers about the table's key.

sql quality warehousing Analytics engineering

Jan 22, 2024

What country actually contains in data-job-postings

3,500 rows, 7 distinct values, and the one fact about country that changes how you query it.

quality practice Data quality

Jan 21, 2024

r = 0.11 between sched_dep_hour and dep_delay_min

A measured correlation of 0.11 between two columns, what its square says, and the diagnostic that costs one line.

statistics Statistics

Jan 20, 2024

Cutting a lesson down — One metric, one definition

The reason three dashboards show three different revenue numbers is never a bug. It is three people who each wrote their own definition, and all of them were reasonable.

practice Analytics engineering

Jan 19, 2024