Home / Guides / How to find a song’s BPM (step-by-step)
How to find a song’s BPM (step-by-step)
You do not need a DAW, an account, or an upload to find a song’s tempo. You need a clean section of audio, a detector you trust, and one habit: check half-time and double-time before you write the number down.
Beats per minute is a count, not a genre. At 120 BPM, four beats pass every two seconds, so one beat lasts 500 milliseconds. At 90 BPM, a beat lasts about 667 milliseconds. That number is useful because it tells a DAW where to put the grid, tells a DJ how much pitch correction a blend needs, and tells a dancer how fast the pulse is.
It does not tell you how the track feels. A 70 BPM trap record and a 140 BPM house record can share a pulse; the difference is where the snare lands and how the subdivisions are played. This is why two people can disagree about the “real” BPM of the same song and both be counting correctly. Pick a convention, write it down, and stay consistent across your project.
2. Before you analyze: choose the right section
Most bad BPM readings are caused by a bad input, not a bad detector.
Use 20–40 seconds with a steady pulse. A verse, a drop, or a chorus with drums is ideal.
Avoid intros, outros, and breakdowns. Pads, risers, and spoken intros give the detector very little to lock onto.
Use a file you own. MP3, WAV, and FLAC work in modern browsers. Streaming-service files are usually protected, so export or record a clip you have the rights to use.
Do not use a clip shorter than about ten seconds. There are not enough repetitions for a stable estimate.
If the song changes tempo, analyze the section you will actually use. A single BPM for a track with a ritardando or a half-time bridge is a fiction; note the tempo of each section instead.
Advertisement
3. Run the detector and read the confidence
Open the free BPM tool, drop the file, and wait for the result. The tool analyzes the audio in your browser; it does not require an account, and the file is not uploaded to a server for tempo detection.
You will get two numbers: BPM and a confidence score. Treat confidence as a hint, not a grade. A high score on a four-on-the-floor house track is normal. A lower score on live jazz, a rubato piano intro, or a field recording is also normal. The question is not “is the score high?” but “does the number match the pulse I hear?” Play the track and count along for eight bars. If your count and the detector agree, move on. If they do not, do the next step before you touch the project tempo.
The tool reporting 128.0 BPM and full confidence for a generated click track. The confidence bar reflects how tightly the detected beats cluster, not whether the count matches your ear. Screenshot: BPMfinder.ca.
4. Check half-time and double-time before you trust it
Half-time and double-time are the most common source of wrong BPM values, and the easiest to fix. A track that feels like 70 BPM may be counted at 140. A drum-and-bass tune at 174 may feel like 87 if you only follow the kick. A boom-bap loop at 90 may read as 180 if the detector locks to the hi-hats. None of those readings is “broken” — they are different counting levels of the same pulse.
Ask yourself: where do the snare and the main backbeat land? If the snare hits on beats two and four of a slow count, the slow number is usually the useful one. If the energy is carried by fast subdivisions, the double may be the better session tempo. Write down which convention you chose. Changing it later is cheap; discovering it after you have tracked eight bars is not. The half-time and double-time guide goes deeper, and the BPM for DJs guide covers beatmatching.
5. Confirm with tap tempo
Tap tempo is the simplest sanity check ever invented. Tap along with the track for eight bars — not two, not four. Eight bars gives you enough taps to average out a nervous finger. Compare the average with the detector. If they are within a BPM or two, you are done. If they are far apart, go back to a cleaner section and repeat both steps.
Do not tap the hi-hat if you want the song’s main tempo. Tap the pulse you would clap to. That is the number a drummer, a DJ, or a DAW grid will agree with.
6. When automatic detection fails
Some material is genuinely hard, and no detector should pretend otherwise.
Rubato and live drums. Tempo breathes. A single number may be wrong for half the track. Note a range, or analyze a steady section.
Ambient, drone, and spoken word. There may be no stable beat to find. A low confidence score is the honest answer.
Short clips and fades. The detector needs repetition; a five-second intro may not contain enough.
Heavily compressed or clipped audio. Transients get smeared, and the detector may lock onto the wrong pulse. A cleaner export usually fixes it.
Tempo changes. Analyze each section separately. Do not average a 90 BPM verse and a 140 BPM drop into one meaningless number.
If the result still feels wrong, trust your ears and use tap tempo. The tool is a second opinion, not a judge. Our measured accuracy test shows where the detector is strong and where it is not.
7. Record the result where you will use it
A BPM you cannot find again is a BPM you will re-detect. Put it where the work happens:
In the DAW session tempo, before you record MIDI or loop-based audio.
In the filename or the playlist notes, especially for a DJ set.
In the track metadata or a simple session note, with the half/double convention you chose.
In the arrangement notes if the song changes tempo.
For a DJ set, BPM belongs next to key and energy, not in a separate spreadsheet you will forget to open. For a production session, it belongs in the project template so every collaborator starts on the same grid.
Common mistakes
Analyzing a three-second clip with no drums.
Assuming every track has one “true” BPM with no half-time option.
Ignoring swing and human timing on live recordings.
Trusting a single detector pass without listening.
Writing the number down without noting which count you used.
Check half-time and double-time against the snare.
Tap eight bars and compare.
Write the number and the counting convention where you will use it.
That is the whole job. The rest is practice.
FAQ
Why does the tool say 140 when the song feels like 70?
Both can be correct. The detector may be counting double-time subdivisions. Check where the snare and backbeat land, then choose the count that matches how you will use the tempo.
What is a good confidence score?
There is no universal threshold. High confidence on a clear electronic beat is expected; lower confidence on live, rubato, or ambient material is normal. Use the score to decide whether to double-check, not to decide whether the result is “allowed.”
Does the tool upload my audio?
No. The analysis runs in your browser. The privacy policy explains what the hosting and advertising layers may log.
Can I find BPM from a Spotify or Apple Music link?
Not reliably. Streaming files are usually protected. Use a file you own or a short clip you have the rights to use.
What formats work?
MP3, WAV, and FLAC in current browsers. If a file will not decode, export a short 16-bit WAV clip and try again.
What if the song changes tempo?
Analyze each section separately and write down the tempo for each. A single number is only useful if the pulse is stable.
If you want something to test the workflow on, the free player at asiancoastline.com streams the Excavationpro catalog without an account. Pick a track, run the BPM check, then listen for where the snare actually lands.
Justin Helmer publishes BPMfinder.ca and records as
Excavationpro. He writes about tempo, music tools, and the
practical side of making records without a label. About the site.