Science of reading apps, and how to tell what one actually teaches

“Follows the science of reading” on a store listing is a claim the developer wrote about their own product. No one certifies it, and there is no approved list to check it against.

That does not make the phrase meaningless. It points at a real body of research about how children learn to read, and apps differ in how much of it their design reflects. It just means the checking is yours, and one question comes before all the others: how does the app teach your child to read a word? A word can be sounded out from its letters, memorized whole as a shape, or guessed from a picture or from the rest of the sentence. An app may use more than one route. Check whether sounding out is the main route and whether it is taught at all. The quickest way to see it is to watch when a picture appears next to a word: before your child reads it, or after?

An app can carry the part of reading that needs repetition, which is the letter and sound work. It cannot do the talking, the reading aloud, and the questions that build the rest.

What does “science of reading aligned” actually mean for an app?

“Science of reading aligned” describes a body of research, not a program an app can adopt. So the question to judge an app on is which of the research’s findings its design is built on, and those findings are specific. The National Reading Panel found that systematic, explicit phonics instruction works better than unsystematic or incidental phonics. The Institute of Education Sciences, the research arm of the US Department of Education, publishes a practice guide on foundational reading skills for kindergarten through third grade. It puts two of its four recommendations in the top evidence tier: develop awareness of the sounds in speech and how they link to letters, and teach children to decode words, analyze word parts, and write and recognize words. The second is broader than the phrase usually suggests, because it asks for writing and spelling and not reading alone.

Whole language and systematic phonics are the clearest contrast here. Whole language leans on context and pictures; systematic phonics directly teaches letter-sound relationships in a planned order. A phonics label alone says nothing about how well an app applies that method.

What are the signs that a reading app actually teaches phonics?

We make one of the apps on this page, Wonster Words, and it is held to the same eight checks as everything else, including where it falls short. The checks come from the Institute of Education Sciences practice guide and from reading researchers Sharon Vaughn and Nathan Clemens, whose review addresses parts of the science of reading that short summaries often omit. The practice guide is written for school leaders about teacher instruction rather than about software, so turning it into something a parent can watch for in an app is our own move. We have used all six apps ourselves; where a check reports what an app does rather than what it says, that is from our own use and we say so.

Does it teach sounding out, and is that the main route? Watch what your child is asked to do with a word they do not know: work through the letters, or look at the word and remember it. Some memorizing is expected and correct, and the practice guide asks for common words to be taught so children recognize them quickly and for words that cannot be sounded out to be introduced whole. But an app where most words arrive as shapes to be remembered is teaching something other than decoding, whatever its listing says.

Does it put your child somewhere, or hand them a menu? Open the app cold. Does it start your child at a point and move them on, or is it a grid of activities to choose between? Both the National Reading Panel and the practice guide ask for teaching in a planned order. No particular order is proven best. Reading researcher Timothy Shanahan points out that about 18 different phonics programs were examined in the National Reading Panel’s studies, “each had its own sequence for introducing letters and sounds. And they all worked.” So a company advertising its uniquely research-backed sequence is advertising. A program with a path rather than a pile is the thing to look for. Our guide to what order to teach phonics skills shows what a sequence looks like.

Does the practice go both ways? Seeing a letter and producing its sound is one direction. Hearing a sound and choosing or writing the letter is the other. Vaughn and Clemens argue both matter and that neither has been shown better than the other, even though speech to print is currently marketed as the superior method. An app that only ever runs one direction is doing half the work.

Does your child spell as well as read? This is the one most often missing. Recognizing a word on screen is easier than producing it from nothing, and the practice guide asks for both in the same recommendation. Segmenting a word into its sounds feeds spelling; blending those sounds feeds reading. An app that never asks a child to build a word is teaching one half of a pair.

Are the words real? Made-up words have a narrow use in assessment: they can show whether a child applies a sound pattern instead of recognizing a memorized word. They are not vocabulary and should not be the material an app teaches. Watch whether the practice is built from words your child can connect to meaning.

What is the picture doing? This check applies to one kind of task only: one with a written word your child is meant to read. There, a picture that shows after your child has read the word confirms the meaning, and a picture that shows first can supply it. Where there is no written word, such as listening for the sound a word starts with, rhyming, or pulling a word apart into its sounds, a picture is how a child who cannot read yet is shown a word at all, and picking the right picture is the task working as designed. Vaughn and Clemens are explicit that the issue is not illustration but what the illustration is asked to do.

Does the practice widen as your child gets better? Decodable books, written using only the patterns a program has taught so far, are useful for a beginner. They are also meant to fade. A child kept indefinitely on books built to one pattern misses the vocabulary, the sentence shapes, and the knowledge that other books carry. Decodable always means decodable relative to that program’s own lesson order, and no app knows what your child has been taught at school, at home, or in another app. Decodability is a real and useful property rather than a promise that every word will be one your child can sound out.

Can you point it at what your child needs? If your child is stuck on blending, you want to spend the next month on blending. Look for whether you can choose a skill, a level, or a category, or feed the app your own words, such as this week’s spelling list. Treat “personalized” carefully: it usually means the app was set up around answers you gave at the start, not that it is watching your child and re-sequencing itself.

Which parts of reading an app can teach, and which need a person

An app can carry the repetition. Letter sounds, blending, spelling patterns, and sight words take repeated practice, and an app can provide it. The language half is yours, and your child’s school’s. Vocabulary, background knowledge, the sentence shapes children meet in books, and the habit of asking what happened and why come from talk, from being read to, and from school.

Phonics is necessary and it is not sufficient. Decoding is the enabling skill, because a child who cannot work out the words cannot understand the page, but comprehension is the point of reading rather than a module that comes after phonics is finished. The practice guide puts that plainly by making academic language, inferential and narrative language, and vocabulary knowledge its first recommendation, sitting alongside the phonics recommendations rather than beneath them.

Reading aloud above your child’s reading level, talking about what you read, and answering the endless questions are not replaceable by software, and an app that implies otherwise is selling something. A specialized app is narrower, not worse, as long as it states which part of reading it teaches.

Which reading apps actually teach phonics, and what is each one built to do

The six apps in this comparison are grouped by what they are built to do rather than ranked by quality. Every claim below is what each company publishes about its own product, read in October 2026, except where we say we have used it. Cost is not compared here; our guides to the best phonics apps and free and no-subscription options cover price.

App	Built to be	How it teaches a word	What it publishes about method
Wonster Words (ours)	Focused word practice	Sounding out, with a sight-word section alongside it	Phonics, spelling, and sight words. Developed with an educator using the Orton-Gillingham approach, with speech-language pathologist input
Hooked on Phonics	A named, sequenced program	Sounding out, and it leads with that	”The step-by-step phonics method… turns letters into sounds, sounds into words.” Organizes its own material by phonics, phonemic awareness, fluency, vocabulary, comprehension, and spelling
Reading.com	A parent to teach the lessons	Sounding out, taught by you rather than by the app	”Research-backed, phonics-based.” 120 step-by-step lessons for an adult and child together, plus letter-writing and spelling
Khan Academy Kids	Free, broad early learning	Interleaved with math and other early skills, in our use	”Standards-aligned,” built in partnership with the Stanford Graduate School of Education, aligned with Head Start and Common Core
ABCmouse	One subscription, many subjects	Reading is one subject among several	An “early learning curriculum” across reading, math, science, and art. “Backed by research,” linking what it calls ABCmouse-sponsored studies
Epic	A library, not a program	It does not teach reading, and does not say it does	A “digital library for kids 12 and under.” Claims no teaching sequence. Spotlight Words and Dictionary Lookup “teach vocabulary and pronunciation”

Three of the six teach a word from its letters: Wonster Words, Hooked on Phonics, and Reading.com, which differ in who leads the lesson rather than in the method. Khan Academy Kids and ABCmouse place reading inside a broader curriculum, where what your child meets depends on which activities they pick. Epic does not claim to teach reading.

Reading.com and Hooked on Phonics are the only two that publish a decodable claim. Both say their storybooks are written to their own lessons. Wonster Words’ books are leveled by learning level rather than labeled for decodability, so they can carry words a child cannot yet sound out, and Khan Academy Kids’ storybooks are not described as decodable either. Epic’s 40,000 books are leveled for reading rather than built to a phonics sequence, which makes it a library to read alongside your child rather than practice to leave them with.

Hooked on Phonics is the only one that publishes a position on pictures, and it is the right one: its storybooks are read “by decoding, not by guessing from the picture.” Using it, we found its later reading practice often carries no picture at all, and the storybooks we saw read as decodable. The books are still illustrated. The claim is about what the text asks of a child, not about taking the art away. Reading Eggs covers similar ground, a sequenced program with a library attached.

Reading.com is not really an app for your child, and that is the biggest difference on this page. It is an app for you to teach from. It briefs you on the skill, then you deliver it, and using it we found a child cannot work through it alone. Most of the practice runs through sliders your child drags while saying each sound aloud. The slider plays nothing, so nothing in the app can hear whether the sound was right. You can, which is the design rather than a gap in it. It is also the strongest answer in the set on pictures: it hides them in its decodable books, which goes further than asking that a picture arrive after the word rather than before it. Whether this suits you is not a question about reading instruction: it is whether an adult can sit down with your child every session.

ABCmouse describes its own research as sponsored, in those words, and links it. That is more candid than most and it is also the thing to understand about every “backed by research” claim in this category. The 2021 meta-analysis Measures Matter, published in AERA Open, opens by noting that thousands of educational apps are available while research on their effectiveness is limited. Almost no individual app has independent research about itself, ours included. The phrase nearly always describes a teaching method rather than a study of the product.

About Wonster Words. We built it, so we can describe its design: the phonics activities were developed with an educator using the Orton-Gillingham approach, with input from speech-language pathologists. That is not a certification, and Orton-Gillingham practitioner certification applies to people rather than software.

Phonics practice is the primary route, with a separate sight-word section alongside it. Practice runs in both directions, from letter to sound and from sound to letter. Your child builds words from their sounds as well as reading them. Every word is a real word, not made-up. Pictures typically appear after a word is read; in the earliest levels they intentionally represent the sounds, for children who are not yet reading.

You can target practice by skill, level, or category, or add your own words through custom and smart lists. What it will not do is respond to a particular mistake: your child retries each item until they answer correctly, and you see their accuracy afterwards rather than a diagnosis of what went wrong.

What to watch for away from the app

No report screen tells you whether your child can read. Watch instead for your child reading a word they have never seen before, out loud, and getting it right by working through the letters rather than recognizing the shape. That can happen on a cereal box or a street sign, and it is what every check on this page is ultimately for.

If it is not happening yet, that is information rather than failure. Our guide to what to do when a child knows letter sounds but cannot blend them is the next thing to read.