Why We Cite Our Sources: The Drop Method

The short answer

Most skincare advice — including from credentialed sources — doesn't tell you which study a claim comes from. Drop does, on every flag, in every routine generation, in every compatibility check. The point isn't 'we have better evidence.' The point is 'we tell you where the evidence is, what its strength is, and where the consensus stops.'

What 'cited' looks like in practice

When Drop flags an interaction, the flag carries a footnote like ¹ that resolves to a specific named source — usually a peer-reviewed paper with authors, year, journal, and a DOI or PubMed ID, and sometimes a regulatory document or clinical reference where that's the best evidence there is. Either way the tier is labeled, and a one-sentence summary of what the citation specifically supports about the claim. Tap the footnote, see the paper.

The citation format works the same way across surfaces:

• Routine generation explanations
• Conflict detector flags
• Compatibility checker results
• Routine ratings
• Ingredient encyclopedia entries
• Long-form articles

Every flag with a footnote like that points to a specific named source, with its evidence tier on the label. If we don't have a citation, Drop gives a third answer instead of a confident one — 'worth checking': 'no peer-reviewed research on this combination yet, dermatology consensus is X.' We never fabricate citations or paper over uncertainty with confident-sounding language.

Why this matters

Three reasons.

First, it lets you check. A claim with a real citation is a claim you can audit. The footnote takes you to the paper, and you can decide for yourself whether the paper actually supports the claim. Most skincare claims fail this audit because they don't have a paper to point at — the claim either oversimplifies actual research or comes from marketing rather than evidence.

Second, it makes evidence-strength visible. Not all citations are equal. Drop tags each citation with an evidence-strength rating (RCT, systematic review, expert consensus, mechanistic, observational, debated). When a flag carries an 'expert consensus' citation rather than 'RCT,' that tells you something about how confident to be.

Third, it forces honest language. When you have to point at the paper that supports a claim, you can't say 'clinically validated to reverse photoaging signs' if the paper only measured wrinkle depth at 12 weeks. The citation requirement is a check on the language. Drop's voice guide bans phrases that overstate evidence; the citation infrastructure enforces it.

What we don't do

Drop does not:

• Make claims we can't cite
• Use marketing-grade descriptors when the evidence is mechanistic-only
• Promise outcomes the literature doesn't support
• Hide the evidence-strength of supporting citations
• Fabricate claims or citations. Every entry is written against the abstract of the paper it names, and before a batch of entries ships we pull 10% of the papers by hand and check the claim against the source. Automation is only allowed to make a claim more cautious, never less; anything that would loosen a warning goes to a person — rather than being generated free-hand by a model
• Use AI-generated 'studies show' filler

What we DO do

• Lead every flag with the strongest citation we have
• Say 'worth checking' when we have no source to cite
• Tag evidence strength visibly
• Update citations when newer or stronger studies replace older ones
• Test the citation database every quarter against the questions people actually ask on skincare forums, to find coverage gaps
• Show every source without an account — each ingredient's citations have a public page, and the methodology page explains how we grade them. A single browsable index of every citation is not built yet; we'd rather say so than imply one

The honesty discipline

The hardest part isn't writing the citations. It's saying 'we don't know' when the evidence isn't there. Skincare marketing has trained users to expect confident answers about everything; many users are surprised the first time they see Drop surface 'no peer-reviewed research yet — dermatology consensus is X.'

That surprise is the point. Confident answers without evidence behind them are the default in skincare marketing. We're trying to make 'I don't know yet' a credible response, with the curated dermatology consensus as the honest fallback when evidence hasn't landed.

What about the studies we miss

The Drop citation database is curated and won't cover every paper ever published. We focus on the claims that matter most — the ones that drive routine decisions — and test the coverage against the questions people actually ask. When you ask something we can't ground in a citation or consensus entry, Drop says so rather than improvising.

Bottom line

The Drop method is citation-first. Every flag, every routine explanation, every interaction check carries the paper trail that supports it. When evidence is thin, we say so rather than papering over the gap. The point isn't 'we have unique evidence' — it's 'we tell you where the evidence is, and we don't pretend to have it when we don't.'