**The most important metric is `petal_length`.** Importance here is a judgement call. The dataset has no revenue-style KPI, and its purpose is to separate the three species (setosa, versicolor, virginica). `petal_length` does that best. **Why petal_length** - Setosa ranges 1.0–1.9 (median 1.5), versicolor 3.0–5.1 (median 4.35) and virginica 4.5–6.9 (median 5.55). - Setosa is completely separated from the other two species, with a gap of more than a full unit between 1.9 and 3.0. - Versicolor and virginica overlap only in a narrow band, roughly 4.5–5.1. - The species averages are far apart: 1.46, 4.26 and 5.55. - By comparison, `sepal_width` ranges overlap heavily: setosa 2.3–4.4, versicolor 2.0–3.4 and virginica 2.2–3.8. It is a much weaker separator. - `petal_width` separates almost as well (setosa 0.1–0.6, versicolor 1.0–1.8, virginica 1.4–2.5), so it is a close second. **Highest and lowest records by petal_length** (from the query on `raw.iris`, shown in the rendered result) - **Highest:** a virginica specimen with petal_length 6.9, petal_width 2.3, sepal_length 7.7 and sepal_width 2.6. - **Lowest:** a setosa specimen with petal_length 1.0, petal_width 0.2, sepal_length 4.6 and sepal_width 3.6. The two extremes come from opposite species, which matches the per-species ranges above. The smallest flower in the dataset is a setosa and the largest-petalled is a virginica. Each species has 50 rows, so the comparison is balanced.