There is no one answer to which skincare expert you can trust, but there is a number worth knowing about each of them: how often she actually lands on a no. We counted every verdict we hold, one per expert per product, and the 32 experts who have covered enough ground to rank range from 44.0% nos at one end to 0.9% at the other.
That is 17,412 verdicts in total across the 50 experts we track, and the spread inside it is the interesting part. The middle of the ranked group says no 8.9% of the time. Nine of the 32 say no less than 5% of the time, nine say no more than 15%, and the other fourteen sit somewhere between. None of which makes anybody a bad reviewer. It does mean that a yes carries different information depending on who is giving it.
Dr. Vanita Rattan is out on her own at the top. She has covered 414 products in our data and said no to 44.0% of them, which is nearly half. Behind her come CYoung at 25.9% of 305 products, Beyond50Skin at 20.4% of 529, Hyram at 19.5% of 522 and James Welsh at 18.4% of 663.
At the other end, Dr. Jenny Liu has said no 5 times in 578 verdicts, a rate of 0.9%. Allie Glines sits at 1.5% across 716 products, Caroline Hirons at 1.5% across 206, Gothamista at 1.9% across 216 and Dr. Sam Ellis at 2.1% across 533. In the middle of the pack, Sincerely Miss Ash lands on a no 8.9% of the time and Doctorly 8.7%, which is about as close to typical as this group gets.
The busiest reviewer we track is Dr Dray, with 1,308 verdicts, more than twice most of the roster. She says no 16.1% of the time, which puts her in the upper third.
No, and it is worth being clear about that before anybody reads a ranking into this. There are at least two blameless reasons a rate comes out low, and both are about how somebody makes videos rather than how honest she is.
The first is format. A creator whose channel is built on roundups of what she loves, or on routines she actually uses, has very few occasions to pan anything. The products that would have earned a no simply never make the edit. The second is screening. An expert who reads an ingredient list before she buys is filtering out the duds before the camera is on, so her published verdicts are drawn from a pre-sorted shelf. Neither of those is flattery, and neither makes her advice less useful. It does mean her yes is telling you she liked this one, not that she weighed it against the things she rejected.
The reverse holds too. A high rate does not make somebody a better reviewer, only a more talkative critic. Dr. Vanita Rattan's 44.0% comes with a channel that spends a lot of time pulling formulations apart, and if you want that, her no is exactly what you came for.
The honest answer is that we cannot tell you from a tally, and we are not going to guess. What we can rule out is the most boring explanation, which is small samples. Look at the ten experts with the biggest bodies of work in our data and the spread is just as wide: Dr Dray at 16.1% and No BS Beauty at 16.7% sit near the top, AliceintheRabbitHole at 14.3%, HotandFlashy at 15.1% and Cassandra Bankson at 12.1% in the upper middle, and Allie Glines at 1.5% and Dr. Jenny Liu at 0.9% right at the bottom. These are people with between 530 and 1,308 verdicts each. A run of bad luck does not produce that.
There is one thing we have measured that moves a rate, and it is money. Across the whole corpus, verdicts from clearly paid videos end in a yes 89% of the time against 76% for unpaid ones, and the nay rate falls from 12% to 4% when a brand is involved. That is covered properly in our piece on whether sponsored reviews are trustworthy, and the likeliest cause there is cherry-picking rather than dishonesty. It is a different question from this one, though. Most of what sits behind these per-expert rates is unpaid.
Three habits, all of which take a few seconds on any product page here.
Check how many experts are behind the score before you read the score. A Guru Score of 90% from three people and a Guru Score of 90% from thirty are not the same claim, and the second one has survived a lot more chances to go wrong.
Notice who is missing as much as who is there. If a product has been seen only by experts at the gentle end of this range, a clean sweep of yeses is less surprising than it looks. If somebody who does hand out nos has looked at it and still said yes, that is the stronger signal.
And check the Sponsored badge, which sits on every paid take on the site. It does not mean a review is dishonest. It means the relationship existed, and on the numbers above that is worth knowing.