Our Process

How We Evaluate

We rate apps on two standards: Does it actually teach? And does it respect your data rights? An app must pass both to earn our full recommendation.

Binary Verdicts

We don't use star ratings, percentages, or spectrums. An app either meets the standard for effective educational software, or it doesn't.

This isn't about being harsh—it's about being clear. Parents and educators deserve straightforward answers, not hedged recommendations.

An app that scores 70% on a rubric might sound decent, but if it's missing a critical learning element, it won't actually teach. Our binary system reflects this reality.

MEETS STANDARD
Evidence-Based Effective
DOES NOT MEET
Standard Not Achieved

Two Standards. No Compromises.

A great educational app must both teach effectively and respect your family's data. We evaluate each axis independently because both matter.

Learning Standard

Based on the Invariant framework—a rigorous set of 11 principles from learning science research that distinguish actual learning from mere engagement.

Developed by Carl Hendrick and evidence-based cognitive scientists. These aren't nice-to-haves—they're requirements for learning to occur.

Visit invariant.education

Data Transparency

Your child's learning data should be yours. We check whether apps provide practical access to your data—not just legal compliance.

We evaluate three criteria: Public API, Data Export, and Parent Visibility. Apps that lock away your child's data fail this standard.

Full data transparency criteria

The Learning Standard: 11 Invariants

The Invariant framework, developed by learning science researchers including Carl Hendrick, identifies the non-negotiable elements that must be present for learning to occur. These principles are drawn from decades of cognitive science research.

An app doesn't need to excel at all 11—but it must not violate any of them. A single critical failure means the app doesn't meet our standard.

01

Active Learning

The learner must be mentally engaged, not passively consuming. Swiping, tapping, and watching don't count unless they require thinking. An app that lets you guess randomly until you're right isn't teaching—it's a slot machine.

02

Engagement in the Learning Process

Attention must be focused on the educational content, not distractions. If an app uses candy, pets, or mini-games as rewards, it's training kids to rush through learning to get to the fun part.

03

Meaningful Content

Content must connect to real learning goals and build on existing knowledge. Random trivia or disconnected facts don't constitute education, no matter how many of them you drill.

04

Social Interaction

Learning is inherently social. Apps should facilitate discussion, collaboration, or at minimum, provide hooks for caregivers to engage. Solo drill-and-kill misses how humans actually learn.

05

Iteration Toward a Goal

Learners need clear objectives and the ability to work toward them incrementally. Apps that present random challenges without building toward mastery fail this criterion.

06

Joyful Learning

Learning itself should be rewarding, not just the prizes for completing it. If the joy comes from unlocking stickers rather than understanding concepts, the motivation is misplaced.

07

Appropriate Scaffolding

Support should be provided at the right level—not too much (which prevents learning) and not too little (which causes frustration). Good apps adjust to where the learner actually is.

08

Feedback and Reflection

Learners need to know what they got wrong and why, not just that they got it wrong. 'Try again' without explanation is not feedback—it's a guessing game.

09

Transfer of Learning

Knowledge gained should apply beyond the app itself. If a child can solve problems in the app but not in real life, the learning hasn't transferred.

10

Appropriate Cognitive Load

The mental effort required should match the learning goal. Cluttered interfaces, unnecessary animations, and competing elements overwhelm working memory.

11

Prior Knowledge Integration

New learning must connect to what the learner already knows. Apps that don't assess or build on existing knowledge are shooting in the dark.

What Fails the Learning Standard

These patterns are common in educational apps but actively undermine learning:

Gamification Over Learning

Points, streaks, and leaderboards that make the game the goal instead of the learning.

Guess-and-Check

Multiple choice where you can tap randomly until correct with no consequence.

Extrinsic Rewards

Candy, stickers, or unlockables that train kids to rush through content.

Passive Video

Long videos with no interaction or comprehension checks.

Drill Without Context

Flashcard-style repetition with no connection to meaning or application.

Age-Inappropriate Claims

Apps claiming to teach reading to 2-year-olds or algebra to kindergartners.

How Apps Can Get Evaluated

We source our app evaluations from the Invariant framework. App developers who want their products evaluated against the Learning Standard can submit them directly to Invariant for assessment.

For Data Transparency evaluations, we conduct our own assessments based on publicly available information, testing of the app, and review of privacy policies and data practices.

See the verdicts

Browse our registry of 2,000+ educational apps, each evaluated against these standards.

Browse Directory