Tool usage
A session starts when someone presses play on the water eject tone and completes when that tone runs its full length, typically 30 to 120 seconds. The gap between the two is the most interesting number here: people frequently stop early, which is consistent with a wet speaker often sounding clearer within the first few seconds.
| Measure | Value |
|---|
| Tone sessions started | 809 |
| People who started at least one | 271 |
| Sessions run to full completion | 315 |
| People completing at least one full session | 154 |
| Average sessions per person | 3.0 |
| Total visitors in period | 475 |
Almost nobody does this on a computer
99% of sessions run on a phone. That is the clearest finding in the dataset, and it has a practical consequence: advice assuming a desktop reader, or a fix requiring a download to a computer, is written for a person who does not exist here. The user is holding the wet device.
| Device | Sessions | Share |
|---|
| Mobile | 801 | 99.0% |
| Desktop | 5 | 0.6% |
| Tablet | 3 | 0.4% |
Where wet phones happen
81 countries appear in the data. The spread is wider than English-language content would predict, which suggests the problem travels even when the guidance does not.
| Country | Visitors |
|---|
| United States | 104 |
| India | 83 |
| Iran | 19 |
| Singapore | 17 |
| Egypt | 15 |
| Brazil | 14 |
| Germany | 13 |
| Canada | 11 |
| Turkiye | 11 |
| United Kingdom | 11 |
What the data does not say
It does not say the tool works. Usage data cannot observe whether water left a speaker, and we will not imply otherwise: completion means a tone finished playing, nothing more. It also cannot separate someone whose phone was genuinely soaked from someone curious about what the sound does.
The sample is self-selected. Everyone counted here found a water eject tool and chose to use it, so the figures describe that population rather than everyone who gets a phone wet. Analytics also under-counts: visitors using ad or tracker blockers are invisible, so true totals are higher by an unknown margin.
Methodology
Figures cover 22 July - 6 September 2026 and come from aggregate Google Analytics data on this site. Two custom events are recorded: one when a tone session starts, one when it completes. Device category and country are standard analytics dimensions derived from the browser and IP address. No audio is captured, no microphone permission is requested by any tool on this site, and no individual is identified or tracked across sites.
Percentages are rounded to whole numbers except device share. The page is updated as the dataset grows, and the period above always states what the current figures cover.
Citing this report
These figures are free to use with attribution, including for journalism and research. Suggested citation:
The Wet Phone Report, Speaker Water Eject, 22 July - 6 September 2026. https://speakerwatereject.com/wet-phone-report
If you need a figure that is not published here, or a breakdown for a specific story, email [email protected] and it will usually be possible.
Questions
Can I cite these statistics?+
Yes, freely, with attribution to Speaker Water Eject and a link to this page. The figures are updated as the dataset grows, so citing the period shown alongside the number keeps your reference accurate.
Why does 99% mobile matter?+
Because it tells you who is actually in this situation: someone holding the wet phone. Guidance written for a desktop reader, or fixes that require a computer, miss almost everyone. It also means phone speaker limitations, rather than headphone or desktop speaker behaviour, are what the advice has to account for.
Does a completed session mean the phone was fixed?+
No, and we deliberately do not claim that. Completion means the tone played for its full duration. Whether water actually cleared depends on where it sat and how much of it there was, and usage data cannot observe that. Treat completion as engagement, not as an outcome.
How is this data collected?+
Through standard aggregate web analytics on our own tool pages: one event when a tone session starts and another when it runs to completion. No audio is recorded, no microphone is used, and no individual is identified. The methodology section states the limitations plainly.
Why publish this at all?+
Because nobody else does. Every guide about wet phones, including ours, rests on physics and repair experience rather than data about what people actually do. Publishing the usage data we have seemed more useful than keeping it private, and it gives other writers something concrete to cite.