EWG or Pura: a hazard score or your own profile

EWG is the most consequential organisation in this whole category. Its Skin Deep database made ingredient transparency a mainstream expectation, and a lot of people reading this page first learned to turn a bottle over because of them. That should be said before anything else.

It is also the most argued-about, and any comparison that skips the argument is not much use.

The core difference: a database versus a photograph

EWG’s app scans a barcode and looks the product up in a database built over many years. Pura photographs the ingredient list and reads it.

Barcode lookup is fast and consistent, and when the product is there, the depth behind it is substantial. When it is not there, you get nothing. That is the ordinary experience for local brands, imports, pharmacy own-label ranges and anything reformulated recently.

Reading the label works on whatever is in front of you. It costs you a good photograph, and a curled or glossy label will sometimes defeat it.

The argument about hazard scores

This is the part worth understanding properly, because it is not a marketing dispute. It is a real methodological disagreement, and it happens to be the reason Pura has no score at all.

EWG assigns ingredients and products a hazard rating on a 1 to 10 scale. The critique, made for years by cosmetic chemists and toxicologists, has two main parts.

Hazard is not risk. A hazard score describes what a substance can do in principle. Risk depends on exposure: how much, by what route, how often. Something can be hazardous at a laboratory dose and irrelevant at the concentration used in a face cream. Rating the substance rather than the exposure produces alarm that the evidence does not always support. If you read one thing about this, make it the difference between hazard and exposure, because it is the single idea that defuses most ingredient scares.

Missing data is penalised. Where public data is thin, an ingredient can score worse for that reason alone. That is defensible as caution and awkward in practice, because it can penalise newer, well-characterised molecules while long-standing ones with more literature score better.

EWG’s position, set out in its own methodology, is that a precautionary approach is appropriate when the alternative is waiting for certainty about things people put on their skin daily. That is a coherent argument, not a foolish one, and reasonable people land on both sides of it. We have linked the critique and EWG’s own explanation above so you can read both rather than take our word for it.

Where we come out: we would rather tell you what an ingredient is, what the evidence actually shows including where it is thin, and whether it matters for you specifically. That means no number. It also means our answers are longer and less shareable, which is a real cost.

The second difference: whose answer is it

An EWG score is the same for everyone. Pura’s answer depends on the profile you set up, because the useful question is usually not “is this bad” but “is this a problem for me.”

That matters most in exactly the situations people care about: an allergy, a condition, a pregnancy, a child. A universal rating cannot see any of that.

Where EWG is better

More products, more years, free, nonprofit, and a published methodology you can actually argue with. As a reference library it is deeper than anything we will have for a long time.

Where Pura is better

The product no database contains. Food and supplements alongside cosmetics. Anything where the answer turns on you rather than on the product. And a plain-language follow-up after the scan.

What neither of us can do

Neither reads anything a label does not say. Cross-contamination, storage, adulteration and actual concentrations are all outside what either app can see. If a reaction is persistent, the page on skin that reacts to everything is more useful than either of us, and a dermatologist with a patch test is more useful still.