Are filler words always bad? A balanced look at the evidence
Updated 13 October 2026 · Speaking Coach guides · 11 sources
A practice tool, not therapy. These guides and Speaking Coach help with everyday speaking habits. App numbers are estimates from automatic transcription of English speech and can be wrong. If you stammer or stutter, or have voice problems, a speech-language professional is the right person to see.
- In experiments, uh helped listeners recognise the next word faster, and um did neither harm nor good. 1
- Listeners were more likely to remember words that came after "er" or after a silent pause. 2,3
- Long delays before an answer can make listeners doubt that you know it. 4
- The practical goal is fewer fillers in prepared speech, not none at all.
Public speaking advice usually treats filler words as faults to eliminate. Research paints a more mixed picture. Some fillers seem to help listeners follow speech, some are neutral, and frequent ones in the wrong setting can still distract or make you sound less sure. Here is what the evidence supports, and what it does not.
The case for fillers
- They carry information. Clark and Fox Tree argue that uh and um signal a short or long upcoming delay, which tells listeners something about what is happening in the speaker's head 5.
- Uh can help comprehension. Listeners in English and Dutch experiments recognised words faster after uh; um had no measurable effect either way 1.
- Hesitation can make words stick. Words preceded by "er" were more likely to be remembered in a later test 2, and the same was true of words after a silent pause 3.
- They point to new information. Listeners hearing a disfluent instruction looked towards things not yet mentioned, because disfluency usually comes before new references 6.
- Linguists see them as normal language. Tottie argues for a more positive view of fillers as planning signals 7, and D'Arcy shows that most popular beliefs about "like" are false 8.
None of this means fillers are good for you as a speaker. It means they are part of how spontaneous speech works, and that listeners are not simply put off by every one.
The case against frequent fillers
- They can signal uncertainty. Listeners judge how sure a speaker is from cues such as delay and intonation. In one experiment, answers that came after a longer delay were rated as less likely to be known by the speaker 4.
- People notice and judge them. Surveys show laypeople think of um and uh as signs of production trouble 9, and broadcasters routinely edit them out 1.
- Fluency ratings follow pauses and speed. When untrained listeners rated how fluent speakers sounded, pause and speed measures predicted their ratings best 10. Lots of hesitation, filled or silent, lowers perceived fluency.
- Habits get in the way of content. A "like" in every sentence or "you know?" after every point pulls attention from what you are saying.
So when should you cut them?
| Situation | Advice |
|---|---|
| Casual conversation | Leave them alone; they are normal and help turn-taking. |
| Job interviews | Reduce at the start of answers; structure helps most. See interview answers. |
| Presentations and pitches | Cut them in the opening and key lines, which you can prepare. See presentation openings. |
| Recorded video or podcasts | Listeners may notice repeated fillers more on replay; practise the key segments. |
| Q&A and thinking aloud | A short "um" before a hard answer is fine; long silences or chains of fillers are what to avoid. |
A sensible target
There is no research-backed "acceptable" filler rate. Rates vary with topic, role and person 11. A useful personal target is relative: fewer fillers than your own baseline on the same kind of task, and none in the sentences you prepared. The progress-tracking guide explains how to compare fairly.
Replacing fillers with brief silent pauses keeps the planning time that fillers buy you. Silent pauses also have effects on listeners 3, so the aim is short, well-placed pauses rather than long gaps. See how to pause.
Measuring without obsessing
A count is a tool for noticing habits, not a score of your worth as a speaker. If you use Speaking Coach, its filler count and filler rate are estimates from automatic transcription of an English practice take, and a few fillers per minute is not a failure. Look at the trend over several similar takes rather than any single number.
Common questions
Do filler words make you sound less confident?
They can. Listeners use delays and similar cues to judge how sure a speaker is 4, and laypeople associate um and uh with production trouble. 9 A few are unlikely to matter; a steady stream can.
Is a silent pause better than um?
For prepared speech, usually yes: it gives you the same planning time without an extra sound. Both affect listeners similarly in memory experiments 3, so keep pauses short and well placed.
What filler rate is acceptable?
There is no agreed standard; rates vary with topic and situation. 11 Compare with your own earlier recordings of similar tasks instead.
Sources and further reading
- Fox Tree JE (2001). Listeners’ uses of um and uh in speech comprehension. Memory & Cognition 29(2), 320–326. doi:10.3758/BF03194926 doi.org
- Corley M, MacGregor LJ, Donaldson DI (2007). It’s the way that you, er, say it: hesitations in speech affect language comprehension. Cognition 105(3), 658–668. doi:10.1016/j.cognition.2006.10.010 doi.org
- MacGregor LJ, Corley M, Donaldson DI (2010). Listening to the sound of silence: disfluent silent pauses in speech have consequences for listeners. Neuropsychologia 48(14), 3982–3992. doi:10.1016/j.neuropsychologia.2010.09.024 doi.org
- Brennan SE, Williams M (1995). The feeling of another’s knowing: prosody and filled pauses as cues to listeners about the metacognitive states of speakers. Journal of Memory and Language 34(3), 383–398. doi:10.1006/jmla.1995.1017 doi.org
- Clark HH, Fox Tree JE (2002). Using uh and um in spontaneous speaking. Cognition 84(1), 73–111. doi:10.1016/S0010-0277(02)00017-3 doi.org
- Arnold JE, Fagnano M, Tanenhaus MK (2003). Disfluencies signal theee, um, new information. Journal of Psycholinguistic Research 32(1), 25–36. doi:10.1023/A:1021980931292 doi.org
- Tottie G (2011). Uh and um as sociolinguistic markers in British English. International Journal of Corpus Linguistics 16(2), 173–197. doi:10.1075/ijcl.16.2.02tot doi.org
- D’Arcy A (2007). Like and language ideology: disentangling fact from fiction. American Speech 82(4), 386–419. doi:10.1215/00031283-2007-025 doi.org
- Fox Tree JE (2007). Folk notions of um and uh, you know, and like. Text & Talk 27(3), 297–314. doi:10.1515/TEXT.2007.012 doi.org
- Bosker HR, Pinget A-F, Quené H, Sanders T, de Jong NH (2013). What makes speech sound fluent? The contributions of pauses, speed and repairs. Language Testing 30(2), 159–175. doi:10.1177/0265532212455394 doi.org
- Bortfeld H, Leon SD, Bloom JE, Schober MF, Brennan SE (2001). Disfluency rates in conversation: effects of age, relationship, topic, role, and gender. Language and Speech 44(2), 123–147. doi:10.1177/00238309010440020101 doi.org