4 GROUP BY Traps, Wrong vs Right
WHERE vs HAVING, nullable grouping keys, bare SELECT columns: 4 GROUP BY traps that run without errors yet fail take-homes, each with the fix.
The killer detail: all four run without errors. The take-home grader doesn't check that your query executed — it checks the output.
Trap 1: WHERE on aggregates
WHERE SUM(x) > 100 filters before grouping, so the condition never sees aggregated values. Move it to HAVING SUM(x) > 100 after GROUP BY.
Trap 2: Bare SELECT columns
Selecting name while grouping by dept works in MySQL and explodes in Postgres — and in interviews. Group by every non-aggregated column or aggregate it.
Trap 3: NULLs in the grouping key
NULLs form a hidden group. Half your rows vanish and the total still "looks right." Fix: COALESCE(key, 'UNKNOWN') plus an explicit null-rate audit.
Trap 4: HAVING without GROUP BY
Valid syntax, nonsense semantics. Move the condition to WHERE or add the grouping. Interviewers plant it to see if you flinch.
Drill all four with wrong-vs-right outputs at Topfolio Practice.
Related: Sql 30 Interview Questions Practice · What Is Sql
Frequently Asked Questions
Why can't I use WHERE on aggregated values?
WHERE filters rows before grouping, so it never sees aggregated values. Filter on aggregates with HAVING, which runs after GROUP BY.
What happens to NULLs in a GROUP BY key?
They form their own silent group. Totals can look right while rows vanish into a bucket nobody asked for — audit with COALESCE plus an explicit null check.
Why does MySQL accept queries Postgres rejects?
MySQL permits bare non-aggregated columns in GROUP BY; Postgres and interviewers treat them as errors. Write portable SQL.
Written by
Founder at Topfolio with 6+ years in data & analytics across JPMC, Ultrahuman, and high-growth startups. Sat on hiring panels, reviewed 500+ resumes, and writes practical SQL & data guides.
Related Articles
5 Window Traps That Lie
Basket-split LAG gaps, ghost orders, nondeterministic ROW_NUMBER, RANK gaps, unframed running totals: 5 window traps with fixes and SQL included.
30 SQL Interview Questions, Free
The 30 SQL patterns analyst interviews repeat — JOINs, windows, GROUP BY traps, NULLs, dates, CTEs — indexed by pattern with live practice links.
5 SQL Date Patterns for Interviews
Month-over-month growth, rolling averages, cohorts, weekday splits, days-between events: the 5 SQL date patterns analyst JDs repeat, with queries.