Editorial standard

Methodology

How LearningPick turns product research, occupational evidence, and editorial judgment into a learning decision.

What the methodology is designed to do

LearningPick is a decision aid, not a provider catalog. We separate maintained product facts from editorial recommendations, keep unknowns visible, and use the same evidence boundaries across reviews, comparisons, best-pick pages, guides, and alternatives.

A provider claim can establish what a curriculum says it covers. It cannot by itself establish independent job readiness, employer preference, hiring probability, salary outcomes, or what every learner experiences.

Research

Collect primary product facts, curriculum details, pricing context, project evidence, assessment structure, and relevant learner evidence.

Verify

Check claims against accessible sources and record verification dates so time-sensitive facts can be revisited.

Evaluate

Evaluate the evidence for career relevance, skill coverage, projects, credential context, beginner fit, learning depth, value, and freshness.

Compare

Use the same product evidence across decision pages so a product does not change roles simply because the page type changes.

Listen

Keep learner evidence separate from product facts and from the Score. Exact sentiment percentages are published only when the evidence gate is met.

Human Audit

AI may assist research and organization, but public recommendations receive human editorial review and accountability.

Affiliate Firewall

Affiliate relationships never determine a Score, recommendation role, comparison result, or editorial conclusion.

How the LearningPick Score is calculated

Career-context learning products receive eight editorial evidence ratings from 0 to 100. Every dimension must be present before the product Score is published. The weighted total is divided by 10 and shown on a 0.0–10.0 scale. Platforms do not receive this product Score.

DimensionWeightWhat the rating is intended to reflect
Career relevance20%How directly the researched learning evidence maps to the named career context and decision being evaluated.
Skill coverage15%Breadth and decision-relevant coverage of the skills documented by the current research.
Project evidence15%What the program establishes about hands-on work, artifacts, capstones, and the limits of project independence.
Credential evidence15%The researched credential context, issuer, assessment structure, and limits of recognition claims.
Beginner fit10%How well the entry requirements, sequencing, and expected starting knowledge fit a true beginner in this context.
Learning depth10%The depth supported by curriculum and task evidence; a topic name alone does not prove mastery.
Value10%The trade-off between current price/time commitment and the learning evidence relevant to the decision.
Evidence freshness5%How current and maintainable the underlying evidence is for time-sensitive facts.

Important: A dimension such as “Credential evidence 88/100” or “Career relevance 92/100” is a LearningPick editorial evidence rating. It is not an employer-recognition survey, placement rate, probability of getting hired, completion rate, or market-share statistic.

Worked example: why Google can show 92/100 for career relevance and 88/100 for credential evidence

For the current Google Data Analytics record, the stored research rationale assigns career relevance 92 because the official program explicitly targets entry-level analyst work and documents an analyst workflow. Credential evidence is 88 because the researched product leads to a Google Professional Certificate and the issuer/provider pages establish the credential, while LearningPick does not treat that credential as proof of employer preference or hiring outcomes.

The same rule applies to every dimension: the number summarizes the strength of the evidence for that dimension in the named career context. It does not convert a provider claim into an outcome statistic. Research notes store the rationale used for each published dimension, and unknown or contradictory facts remain visible rather than being silently scored as proven.

How skill depth is represented

Depth uses a separate 0–5 evidence scale: 0 = Not covered, 1 = Exposure, 2 = Basic, 3 = Working proficiency, 4 = Strong, 5 = Advanced. A null value means there is not enough evidence. “Not covered” and “not enough evidence” are deliberately different states.

How learner evidence is used

Learner comments, forums, reviews, and community discussions can surface recurring friction, useful scenarios, or contradictions. They remain separate from official product facts. Small qualitative samples are labeled as anecdotal and are not converted into percentages or employment claims.

When the evidence is too sparse or inconsistent, Learner Sentiment remains “Insufficient Evidence.” That is preferable to inventing precision.

How career and job evidence is used

Career-outcome pages may use bounded occupational or job-posting evidence to identify recurring tasks and tools. The population, geography, time window, and source must be stated. Occupational evidence is used to compare learning with work requirements; it is not used to promise employment.

How LearningPick Works

Research
We review official course pages, provider documentation, curriculum, pricing, and credential information.

Verify
We record sources, check important facts, and keep verification dates visible.

Evaluate
We use the published LearningPick framework when a product has enough structured evidence.

Listen
Learner feedback is kept separate from the editorial score and is only quantified when the evidence contract is satisfied.

Human Audit
AI assists research, extraction, and comparison. Human review verifies facts, sources, and editorial conclusions.

Affiliate Firewall
Affiliate relationships do not influence LearningPick Scores or rankings.