About this site
Who made it
I am a beginner at the piano. At work, I specialize in research using AI agents. I decided to apply that knowledge and experience to choosing my own learning material, and to share what I found. Because I am a piano beginner, I think I can write from the viewpoint of people with the same worries.
I am not a piano teacher. I do not guarantee learning results, and I do not give advice that replaces lessons.
How it was researched
- The intended readers are adults who have never played the piano and can read English-language materials.
- Six materials were compared: four English-language method books for adults (Faber, Hal Leonard, Alfred, Bastien), one public-domain classic (Beyer), and one free PDF coursebook (SightBasics).
- There are 12 criteria (see Criteria & averages). I found no study comparing materials for adult beginners; most criteria are organized from research on children and from teachers' practice reports. They are not proof that "the more a material meets these criteria, the faster you improve."
- Every item carries a source, a check date, and a verification level.
Verification levels
- Official Official (publisher, developer, IMSLP, etc.). Note that a developer's own site is self-reported
- Original Original (the material itself; the table of contents, text, or PDF was read directly)
- Retail What a retail site or store displays (editions and figures may be inconsistent)
- Catalog Library catalogs and bibliographic databases
- Teacher Reviews by teachers and experts
- User Reviews by purchasers and users
- Research Research papers (those whose full text was read)
- Secondary Secondary information (search-result summaries, second-hand quotations, etc.)
- Inference The author's inference (not fact)
- Unconfirmed Could not be confirmed (this does not mean "absent")
How AI was used
AI was used in the research. AI can miss sources, misread them, and overstate. So I do the following:
- Attach a verification level, and distinguish opinions of teachers and users from official information and the original material.
- A person opens the sources and checks them (what could not be checked is stated as such).
- Check the consistency of the data by program (blanks, missing sources, mismatches between scores and averages, scores based only on "could not be confirmed").
- Fix errors and record them in the change log. Many were in fact fixed during the research.
Conflicts of interest
There is no affiliate income and no advertising on this site. I have no relationship with any publisher or developer.
Limits (stated up front)
- The amount of information I could read differs by material. I read the whole SightBasics PDF, but for the others the evidence is mainly official information, retail sites, and teachers' reviews. I could read Alfred's table of contents, but there is no material to check the contents of Bastien.
- User voices are skewed toward affiliate articles. I could hardly read Reddit, purchaser reviews on Amazon, or forums.
- The track record (C12) is structurally unfavorable to new materials.
- The criteria were written out and adjusted after examining the materials. They were not fixed in advance.
- There is no research on which material adults stick with best.
The scope of this research and what is not known are summarized in Scope & limits.