What is Smoke vs Sanity Testing?

Smoke testing is a broad, shallow check that a new build's core functions work well enough to justify further testing, while sanity testing is a narrow, deeper check that a specific function or fix behaves correctly after a change. Both are quick gates, but with different scope and intent.

Start a Free Mock Interview

Free to start · 7-day trial on paid plans

IN DEPTH

In depth.

These two are constantly confused, so interviewers love the distinction. Smoke testing (build verification testing) runs first on a fresh build: it touches the most important features broadly, can the app launch, log in, load key pages, to decide whether the build is stable enough to test at all. If smoke fails, you reject the build immediately rather than wasting effort.

Sanity testing is narrower and usually unscripted: after a small change or bug fix, you sanity-check that the specific area works and the fix did what it should, without running the full suite. It is a focused, "does this particular thing make sense now?" check.

A simple way to remember it: smoke is wide and shallow (many features, surface level) and runs on every new build; sanity is narrow and deep (one feature, more thoroughly) and runs after targeted changes. Both are quick, both gate further effort, but smoke asks "is the build worth testing?" and sanity asks "did this specific change work without obvious breakage?"

WHY IT MATTERS

Why interviewers ask about this.

Smoke vs sanity is one of the most frequently asked QA definition questions. A crisp, correct answer, smoke is broad/shallow build verification, sanity is narrow/deep change verification, instantly signals solid fundamentals, while confusing them is a common red flag.

EXAMPLE

Example scenario.

A new build arrives: smoke testing confirms the app launches, login works, and key pages load, so it is worth deeper testing. Later, a developer fixes the coupon bug; the tester sanity-checks just the coupon flow to confirm the fix works and nearby behavior is not obviously broken, without rerunning the entire suite.

TIP

Interview tip.

Say smoke testing is broad and shallow (verifies a new build's core functions to decide if it is stable enough to test) and sanity testing is narrow and deep (verifies a specific function or fix after a change). Smoke runs on new builds; sanity runs after targeted changes. The wide-shallow vs narrow-deep contrast lands cleanly.

FAQ

Frequently asked questions.

Is sanity testing a subset of regression testing?

Sanity testing is often considered a narrow, focused subset of regression testing: it checks that a specific change or fix works and has not obviously broken nearby functionality, without the breadth of a full regression pass. Full regression is broader and more systematic.

Which runs first, smoke or sanity?

Smoke testing runs first, on a new build, to decide whether it is stable enough to test at all. Sanity testing comes later, after specific changes or fixes, to verify those particular areas. They serve different points in the cycle.