Skip to main content

Bottled water vs filtered tap water

Is bottled water worse than filtering my own?

The verdict

Filtered tap, by a wide margin and on measured evidence. Bottled water is the single most-measured microplastic exposure there is — around 240,000 particles per litre in the 2024 PNAS work. A reverse-osmosis membrane is the only home mechanism that acts below a micron.

66/ 100Aquafina

Lightweighted PET (#1) bottle with a polypropylene cap

ElevatedGrade A

88% confidence

Lower exposure20/ 100Under-sink reverse osmosis system

Polyamide membrane in a plastic housing, usually with a bladder storage tank

LowGrade C

62% confidence

These carry different evidence grades. Aquafina is grade A and Under-sink reverse osmosis system is grade C. One component of our score is measured release, which only earns points where a laboratory has actually counted particles — so a product nobody has tested scores lower than an identical one that has been. Part of the gap you are looking at is a difference in how much anyone has looked. Why we do it this way →

Where they actually differ

Bottled water vs filtered tap water compared across four dimensions
 AquafinaUnder-sink reverse osmosis system
Measured particle load~240,000/L (PNAS 2024); 93% of bottles contaminated (Mason 2018)Unmeasured, but the membrane pore size is below the particle size
Where the plastic comes fromThe bottle wall and cap, over monthsHousings and tank bladder, over seconds of flow
Evidence gradeA — brand was in a tested cohortC — membrane physics, no particle study
Cost per litreRoughly a thousand times tapFilter changes only

Where this comparison stops being reliable

This is a comparison across evidence grades, which needs care: bottled water scores high partly because it has been measured, and RO scores low partly because nobody has counted particles in its output. The direction is not in doubt — the mechanism is a bottle versus no bottle — but the size of the gap is less certain than the numbers imply.

The full picture on each

Neither of these is what you own

A comparison is between two product types. The item in your cupboard has its own packaging, its own age and its own condition. Scan it and get the same five-component score for the actual object.

Scan what I actually have

Frequently Asked Questions

Is bottled water worse than filtering my own?

Filtered tap, by a wide margin and on measured evidence. Bottled water is the single most-measured microplastic exposure there is — around 240,000 particles per litre in the 2024 PNAS work. A reverse-osmosis membrane is the only home mechanism that acts below a micron.

Aquafina or Under-sink reverse osmosis system — which is lower?

Under-sink reverse osmosis system, at 20 out of 100 against 66. They carry different evidence grades (A and C), so part of that gap reflects how much each has been measured rather than a physical difference.

Where does this comparison stop being reliable?

This is a comparison across evidence grades, which needs care: bottled water scores high partly because it has been measured, and RO scores low partly because nobody has counted particles in its output. The direction is not in doubt — the mechanism is a bottle versus no bottle — but the size of the gap is less certain than the numbers imply.

Does either brand pay to win this comparison?

No. Scores are computed from five published components and no brand can pay to change one, appear here, or be ordered differently. Where a link earns us a commission we disclose it, and several of the options we recommend most earn us nothing.

← All comparisons · Product database · Methodology