Methodology

How we decide what makes the cut.

SteepGuide is built around a simple promise: useful tea guidance first, commerce second. We would rather recommend nothing than recommend a mediocre object because it pays.

Who edits SteepGuide

SteepGuide is maintained by Dylan Harrington and Shali Harrington under the SteepGuide editorial name. Pages are research-led and reviewed against the sourcing rules below.

When we have not personally tested a product, we say that through our evidence model: official specifications, specialist sources, community signal, and tradeoffs. We do not use hands-on language unless Dylan or Shali has actually used the item.

Editorial principles

  • We recommend the best object we can defend, even when there is no affiliate program.
  • We separate tea education from commerce so a brewing guide can be useful without buying anything.
  • We score evidence from specialist credibility, community consensus, owner complaints, durability, and design quality.
  • We do not publish star ratings, fake hands-on claims, or scraped prices.
  • Every recommendation should include who should skip it, not just who should buy it.

Evidence rubric

Specialist credibility

Tea shops, producers, educators, and manuals carry more weight than generic listicles.

Community consensus

We look for repeated recommendations and repeated complaints across tea communities, not isolated hype.

Use-case fit

A kettle for sencha, a gaiwan for oolong, and a matcha whisk solve different problems.

Tradeoffs

Every pick must say who should skip it.

Design quality

The object should deserve counter space, not just function.

Affiliate independence

No affiliate program is better than a compromised recommendation.

Source hierarchy

  • Specialist tea educators and producers for brewing, processing, and category context.
  • Official product pages for specifications, materials, and availability-sensitive details.
  • Independent editorial reviews and durable community consensus for owner sentiment and recurring complaints.
  • Merchant copy only when it is the primary source for a product claim, never as the only reason to recommend something.

Corrections

If a source breaks, a product changes materially, or a claim is too broad, we update the page, keep the recommendation language conservative, and advance the reviewed date when the review is complete.

What we will not do

  • No fake hands-on claims.
  • No scraped prices or star ratings presented as live truth.
  • No huge product dumps.
  • No medical promises for herbal tea.
  • No “best” claims without visible reasons.