CapableAll articles

How We Source Our Claims

Updated on September 4, 2026

Direct answer: every scientific claim on this blog is checked against a research file of primary studies, tagged by evidence level, before publication. If a number is not in that file, or not traceable to a verified primary study, we do not publish it, no matter how often it circulates online.

Habit science is full of numbers repeated until they sound true. This page explains where our facts come from, how we grade them, and what happens when a cited study turns out to be wrong.

Key takeaways

  • Every statistic traces to a primary study named and dated in our internal research file, never a paraphrased blog post.
  • Each claim is labeled by evidence level: what a study measured versus a reasonable inference from it.
  • Widely repeated numbers without a verifiable source, like the "21 days to form a habit" claim, are named as myths, not fact.
  • Every article passes a human science review, separate from a mechanical format check, before it publishes.
  • If a cited study is later invalidated, we correct the article publicly instead of editing it quietly.

Where do our facts come from?

Our factual base is a single research file, compiled by Capable's product lead while designing the app, from academic literature on habit formation and behavior change. It holds primary studies, not summaries, each tagged by what it measured and its sample size. An article starts from that file, not a search engine: a claim either exists there, sourced to a named study, or it does not get written.

Why do you reject some famous statistics?

Because being famous is not the same as being sourced. The claim that habits take exactly 21 days has no primary study behind it; it traces to a 1960 personal observation about patients adjusting to a new appearance, not to research on habit formation. Real research on habit timelines exists and shows a wider, more variable range, and that is what we cite instead. The same rule applies to any figure that spreads without a checkable source: if we cannot verify it ourselves, we leave the number out, even if it would read better in a headline.

What happens if a study you cited turns out to be wrong?

We correct the article publicly. If a study is retracted, fails to replicate, or a flaw comes to light, we add a visible note explaining what changed, and update our internal research file so the error cannot resurface elsewhere. Silent correction would let a bad number reappear later; a visible one closes the loop for anyone who read the original.

FAQ

Do you ever cite statistics without a source link?

No. Every statistic links directly to the primary study it comes from, with its publication year. If we cannot verify a source, we describe the finding qualitatively instead of attaching an unverified figure to it.

Is the "21 days to form a habit" claim ever true?

No. It originated from an unrelated 1960 observation, not from research on habit formation. Studies that do measure this show it varies widely by person and behavior, with no fixed number of days.

What is the difference between a primary study and a source you reject?

A primary study is the original research, published with a method and sample size we can check. We reject aggregator posts that cite "studies show" without naming one, and figures rounded or stripped of context as they spread online.

Who writes and reviews these articles?

Capable's growth team writes the content; the person who compiled the app's research file reviews it for scientific accuracy. See our author pages for background on both.

Sources

Studies referenced in this page are drawn from Capable's internal research file, which compiles and dates the primary sources cited across the blog. Individual studies are linked within each article that cites them.

Authors

Olivier David

Olivier David has spent twenty years building consumer web products used by millions. He co-founded Miwim, the company behind Bons-de-Reduction.com and Poulpeo.com, sold it to RetailMeNot in 2012, then ran RetailMeNot France as Country Manager. He also co-founded DocForYou, a symptom-checker app that reached #2 on the French App Store, and invests in early-stage startups as a business angel. Capable is his answer to a question he could not shake: why does no habit app apply what behavioral science actually shows? He wrote Capable's research file, dozens of primary studies reviewed and tagged by evidence level, and reviews every scientific claim on this blog against it before publication.

Théo David

Théo David leads growth and content for Capable. He was a growth engineer at Finary, one of France's fastest-growing fintech companies, where he built automated content pipelines and the editorial standards behind AI-search-ready articles. Before that, he worked as a data analyst at HelloWork and co-founded Tiro, a creative agency. He studies Big Data, AI and Management. At Capable, he runs topic selection, writes or edits every article, and enforces the format checks each piece must pass before science review.