How We Pick

Start with what this does not include

Most hair care sites describe their method in terms of what they did. It is more useful to start with what we did not, because that is the part that usually goes unsaid.

• No first-hand testing. We have not used the products we recommend. Not some of them — none of them. Every roundup says so on the page.
• No laboratory. Nothing here has been measured, weighed or tested under controlled conditions by us.
• No panel trials. There is no group of testers.
• No original photography of products. Imagery on this site is licensed stock, used for atmosphere, and it never depicts a specific recommended product as though we photographed it.
• No clinical credentials. Janet H. holds no cosmetology, trichology or dermatology qualification, and nothing on the site implies one.

Plenty of competing sites imply some or all of the above. If that is what you want from a recommendation, this is not the site — and knowing that quickly is worth more to you than a page of reassurance.

The six steps

1. Identify the problem, and the queries people actually use for it. Pages start from a real question — "why are my roots greasy by day two", "what do I do about brassy blonde" — not from a product category we want to sell into.

2. Build a candidate set from products currently sold. Real, currently available products, checked against the manufacturer's own site. A recommendation for something discontinued is worse than no recommendation.

3. Assess each one against the specific problem. What the active ingredients are and what they are documented to do; what the formulation type is; whether it is sulfate-free or silicone-free and what that implies; which hair type, strand thickness and porosity it suits; and the known trade-offs.

4. Verify availability. Every product links to the manufacturer's own product page as its source, so you can check both that it exists and what the maker claims about it.

5. Assign a role, not a score. Each pick gets a specific editorial slot: the drugstore option, the one for high porosity, the strongest pigment, the gentlest of the set. That is a claim we can defend from the formulation. A number out of ten is not.

6. Re-verify on a schedule. Roundups are reviewed quarterly, comparisons and guides less often, and the visible date on a page changes only when something substantive changed. Bumping a date to look fresh is a spam signal and we do not do it.

What counts as a source

Ranked, and we use the highest available for any given claim.

1. Peer-reviewed dermatology and cosmetic science. Used for claims about what happens to the hair fiber — how bleaching changes the cuticle, what porosity means physically.
2. Professional bodies, chiefly the American Academy of Dermatology. Used for guidance on washing frequency, scalp conditions and hair loss.
3. Manufacturer documentation. Used for what a product contains and what the maker claims it does — and always labeled as the manufacturer's claim rather than presented as established fact.
4. Named professionals quoted in reputable publications, attributed to both the person and the publication.

What does not count, ever:

• Other affiliate roundups. Circular, and the most common source in this category.
• Forum and social consensus as fact. Occasionally useful as "many users report", clearly framed as that and nothing more.
• Anything unattributed, including AI-generated summaries.

Where the evidence is thin, we say so

This is the part we think is worth most. Some of the questions people search for do not have good answers, and the honest response is to say that rather than pick a side.

Rosemary oil is the clearest example: one small randomized trial, routinely cited far beyond what it supports. Our page on it explains what the trial found, names its four significant limitations, and still concludes it is worth trying — because it is cheap and it has the best evidence in its category. That is a more useful page than either the hype or a flat dismissal.

The same applies to growth serums, where we spend more space teaching you to tell a randomized trial from a consumer perception survey than recommending anything.

Corrections

If a page is wrong, tell us at the contact page. Corrections to factual claims are made and noted on the page with a date rather than quietly edited, and a product that is discontinued or has changed formulation is replaced rather than left as a dead recommendation.

Commercial interest, stated

Links to products are affiliate links and we earn a commission if you buy through them, at no extra cost to you. Nothing about that changes step five: the role assigned to a product comes from its formulation, and we name drugstore products as the best answer on plenty of pages where a salon brand would pay more. The full disclosure is here.