Trusted bearing data for catalogs, applications, and AI.

Find structured specifications, reconcile supplier records, and maintain a consistent catalog with traceable data.

Explore dataImprove my catalog
  • Catalog checker available
  • Record schema draft v0.1
  • Datasets & API on request

Start with your task

Data, developer access
or catalog operations

One shared data foundation, three ways in. Pick the job you are doing today — no sign-up wall, no persona quiz.

Worked example

One row, four honest states

A supplier says “Width B = 12”. Is that ring width, total width or height? It depends on the bearing family and the source. MyBearings keeps the original, records the rule it applied, and leaves what it cannot prove in plain sight.

Read the width guide
Supplier row → normalized record → evidence → open questionsIllustrative demo · synthetic data

  1. Raw supplier row

    As received

    Supplier SKU
    SYN-88213
    Description
    "DEMO 101 brg open"
    Width B
    12
    Origin
    "see listing"

  2. Normalized

    Mapped to the model

    Manufacturer
    Synthetic Maker A
    Designation
    DEMO-101-OPEN
    Family
    synthetic-radial
    ring_width
    12 mm

  3. Evidence

    Why we believe it

    Source
    Illustrative datasheet, row 4
    Raw label
    "B"
    Rule
    family → B = ring width
    Review
    synthetic_only

  4. Still open

    Kept visible

    total_width
    unknown
    Origin
    lot evidence needed
    Closure
    open vs. sealed — check
    Action
    review queue

Use it today

Useful before you
talk to anyone

Everything below works on this site right now, without an account. Items that are not built yet say so.

Resources

Learn the hard parts
of bearing data

Practical methods for bearing catalog data, written around worked examples rather than keywords.

All resources

Quality commitments

Trust comes from
traceability, not adjectives

No “largest database” claims. Authority has to be earned by showing the work.

Methodology
  • 01

    Every value carries its evidence

    Source, location in the source, original label and unit, extraction method and review state travel with each attribute.

  • 02

    Unknown is not the same as not applicable

    Blank cells get a status: populated, unknown, not applicable, conflicting or not researched. Completeness is measured against applicable fields only.

  • 03

    Identity before similarity

    An exact product is manufacturer plus full designation. Variants and cross-brand candidates stay separate records with typed relationships.

  • 04

    Numbers only when measured

    Coverage, accuracy and match precision will be published with definitions, denominators and dates — not before they are measured.

How a catalog evaluation works

Sample, assess, review,
then decide

A paid evaluation starts small and ends with a decision, not a lock-in. Scope, timing and price are agreed per catalog.

  1. Share a sample

    One category export and the sources you already trust. You can run the local checker first to see structural gaps.

  2. Assessment

    We map fields to a family model, mark applicability, and list identity, unit and evidence problems with examples.

  3. Review together

    You see proposed changes with sources and unresolved cases — nothing is silently overwritten or merged.

  4. Scoped next step

    A delivery file, a recurring maintenance scope, or a clear “not worth it”. Your data stays private to you.

Assessment status: On request — scoped by conversation, no self-serve upload.

Bring one messy category.
Leave with a clear next step.

Share what your catalog looks like today. We will tell you what a scoped assessment would cover, what it would not, and what we would need from you.

Discuss your catalogRun the free checker