A Guide to Rally Reader's Mission and Technology

That's the goal. That's the North Star. From the beginning, that's what Rally Reader was purpose-built to do.

What you might be wondering:

If you're a parent or educator reading this, I understand your skepticism. We're living through a moment when AI gets applied to everything. This guide addresses your important questions: Is this tool helping my child build real skills? What, if anything, is Rally doing with my child's data? And is this just one more screen?

What We Built and Why We Built It

The Problem: Instruction Alone Isn't Enough

Two seemingly contradictory things are true:

• Canadian reading scores are at their lowest point since Canada began measuring against the rest of the world ¹
• Schools are getting really good at teaching kids how to read

Canadian fifteen-year-olds are reading at the lowest level PISA has ever recorded, and nearly one in five now falls below the baseline the OECD considers necessary to take part fully in adult life. ¹ That gap doesn't open at fifteen. It opens in the early grades, in the years when reading practice either happens or it doesn't.

Why Rally Reader? Because instruction is only half the equation. Educators have worked diligently to improve reading instruction: structured literacy, phonics-first approaches, Science of Reading frameworks. What's missing is the other half, the practice that converts that instruction into skill.

The Solution: Instruction + Practice = Skill

Educators provide the instruction. Rally Reader supports the practice.

Think about it this way. A coach can teach a wrist shot in one practice: weight back to front, roll the wrists, follow through to the target. But it doesn't become a shot until a kid has taken a thousand of them at a net in the driveway. A kid can get guitar lessons from an actual rock star, but practice makes the musician. Reading works the same way.

What's Missing

Schools are working hard to provide effective instruction. What's missing is the structured, supported practice that converts that instruction into fluent, joyful reading. Kids who learn to read become kids who love to read. And when kids read more, they expand vocabulary and background knowledge, strengthen comprehension, and build the critical thinking needed to excel in school and beyond.

Rally's Role

In a perfect world, every child would have a supportive adult, available at just the right moment, to listen when they read aloud and gently correct errors.

Rally isn't here to replace parents or teachers. It's here for the moments when a supportive adult can't be, providing valuable error-correction during independent practice. It matters because correcting mistakes is essential for learning.

Rally offers this error correction in the context of an enormous selection of books, so kids get the support they need while reading content that interests them. We will share more about this in the section below.

What Happens When a Child Reads on Rally

Step 1: Choose a book from 50,000 titles (Penguin Random House, Scholastic, HarperCollins, Macmillan, Disney, and others). Scholastic Canada's research is blunt about why that matters: 94% of Canadian kids say their favourite books are the ones they picked out themselves, and 92% say they're most likely to finish a book they chose. When kids choose, they read.

Choice only works if there's something to choose from, and that's where Canadian families are at a disadvantage. The same survey found that Canadian households reported significantly fewer books at home than their US, UK and Australian counterparts, and that Canadian children had more trouble finding books they liked. Nearly half of Canadian kids (47%) said they struggle to find books they enjoy, and among infrequent readers that figure climbs to 69%. A library of 50,000 titles in a child's hand is a direct answer to that. ²

Step 2: Choose a reading mode

• Mode 1: Silent Reading A child reads silently with access to vocabulary support. Progress tracking and real-time error correction are not available.
• Mode 2: Oral Reading with Feedback A child reads aloud with real-time feedback from Rally's digital reading coach. Instant correction on mispronounced or skipped words. This mode tracks progress, including accuracy, Persistence, Sight Word mastery and Trip Words. Nothing is recorded.
• Mode 3: Oral Reading with Feedback and Screen Recording Same as Mode 2, plus screen recording for parent-teacher meetings, student-teacher meetings, and assessment. Screen recording is available on iOS only; there is no recording feature on Android. Recording is off by default, requires an active opt in, and can be turned off at any time. There is no front-facing camera, in any mode, ever. What a recording captures, how it is stored and how long we keep it is in "How recording actually works" below.

Step 3: Motivation and Progress Tracking

In Mode 2 and Mode 3, children's effort becomes visible in rewarding ways. For example:

• Trip Words Every child stumbles on certain words. Those words automatically become personalized flashcards. Instead of generic review, a child practises the specific words they struggled with. Targeted practice accelerates progress.
• Sight Words 1,000 high-frequency words make up 75% of all K-8 books. Building automaticity with these words is essential for fluency. Rally turns that work into a game called Sight Words. When a child reads these words correctly, they sparkle. This visual celebration of hard work is immediate and rewarding. Once a word is read aloud correctly three times, it is "won." Win 125 words to complete a level; win eight levels to complete the game and master the 1,000-word foundation.
• Persistence We track accuracy (the percent of time a child reads a word correctly on the first attempt). And we developed our own metric to measure something more important: whether a child sticks with it. If a child reads a word correctly after multiple tries, they get 100% on Persistence. We celebrate the willingness to work through challenges, not just instant perfection.

What This Creates

Rally makes the invisible work of reading visible. Kids feel seen. Their effort is recognized. And that clarity and celebration changes how they see themselves as readers.

Does Rally Have Proven Results?

YES! Rally works because it's built on practices that four decades of literacy research have proven. We will share more about that research at the end of this guide, but before we go any further, we want to provide some clear data points about Rally Reader's efficacy.

We recently completed a study of 1,651 students in Grades 2-6 across the 2024-25 school year, measured with i-Ready and district assessment scores.

A note on where that study was done. It was conducted in a California district, because that's where Rally has been in schools longest. The reading science it rests on is not country-specific. Decoding, fluency and the value of corrected oral practice work the same way in Moncton as in Claremont.

We'd rather have more evidence than less. Rally is looking for Canadian research partners, including districts, boards and ministries. If you are interested in doing research with us, please write to us at Hello@RallyReader.com and put Research in the subject line.

Grade 3 reading was a key focus of our study. Grade 3 marks the transition from learning to read to reading to learn. It is the point where a reading gap starts to compound.

The Canadian evidence on what that gap costs is unusually good. Statistics Canada followed 30,000 fifteen-year-olds into adulthood and found that reading proficiency was the strongest single predictor of whether they finished high school. Only 75% of youth at the lowest reading level had graduated by age 21, compared with 99% of youth at the highest level. ³

Grade 3 students who used Rally as designed grew 132% faster than their peers who did not, a gain of 11.4 i-Ready points (p=0.002).

That is a key headline, not the whole study. Grade 5 students who used Rally as designed gained 21.6 points on the CAASPP state summative assessment (p=0.048) and 20.0 i-Ready points (p=0.010). Across every grade and every user, including students who used Rally far less than recommended, the gain was 8.3 i-Ready points (p<0.001). More detail on all of it is below.

What we learned that surprised us:

These results came at a remarkably low dose. Students in the study logged just 3.5 total hours of supported oral reading with Rally across the entire school year, roughly six and a half minutes per week, and the gains were still impressive.

Although Rally users show those gains with low-dose usage, we continue to recommend more reading time. Our recommendation is fifteen minutes a day, consistent with the daily independent reading practice most literacy research supports. The Claremont results provide a baseline. Treat them as the least you should expect, not the most.

Quality instruction + supported practice = measurable change, whether or not a child faces learning challenges, and whether or not there are books at home. ⁴

What Rally Does Not Do

Every tool should tell you where it stops. Here is where we stop.

Rally does not teach a child how to read. Phonemic awareness, phonics and decoding are the foundation, and building that foundation is skilled instructional work that takes training and judgment. This is best done by educators. The field has gotten markedly better at it over the past decade, and the shift towards structured literacy is one of the real success stories in education. That work is not ours, and we would not do it better.

Rally comes next. Passing the written driver's test is not the same as being a confident driver. That takes hours behind the wheel. Once a child can decode, what they need is time on real pages. That is the part we are built for. It is also the part that all too often goes missing, because it requires a patient adult listening to one child at a time. Everything Rally does rests on instruction an educator delivered first.

Rally is not a diagnostic tool. We do not screen for dyslexia or any other learning difference, we do not identify a reading disability, and nothing on our dashboards should be read as a diagnosis. What we provide is continuous evidence of what a child does when they read: which words trip them, how often, whether they push through. That is useful information to bring to a reading specialist. It is not a substitute for one.

Rally does not assess comprehension. What we measure is oral reading: accuracy, words per minute, Persistence, words read. Fluency built through Rally supports comprehension, and the Claremont study showed strong gains in both literature and informational comprehension on i-Ready. But the assessment of whether a child understood the chapter belongs to an educator, and we are careful not to imply otherwise.

Rally is not an intervention program. A child who cannot yet decode needs targeted instruction from a person, not practice from an app. Rally will not close that gap, and we will say so plainly to any school that asks us to.

Rally does not replace a person. It is what happens in the hours when a supportive adult is not available.

Let's Talk About Speech Recognition, AI, and Data

Parents ask us two questions about AI and data:

• Is this just AI doing the thinking for my kid?
• Are you harvesting my child's data to train your models?

Our answers: no, and no.

Rally Reader is NOT generative AI

No new content is created. Rally Reader does not generate text, predict what comes next, or do the thinking for your child.

Rally Reader is Speech Recognition and Pattern-Matching

Rally's pattern-matching has three steps: listen, compare, flag.

Here's how it works. When a child reads aloud, Rally listens in real time to what they say. It compares their speech to the words on the page. When there's a mismatch, Rally flags it so the child can try again. Listen, compare, flag. That's the complete scope of what the AI does.

The thinking remains entirely with the child. Rally simply tells them when they've misread or skipped a word.

Rally Reader + Data

Here is our core promise to you:

The rest of what we never do is set out in full below.

Before reaching a child's device, our speech recognition models are trained on public and acquired data. This data is intended specifically for speech model training. We choose sources that help us understand the broadest range of voices, including accents and speech impediments. Your child's data is never incorporated into our models.

Once your child reads on Rally, their data stays secure and is used to help them see and accelerate reading progress. We also look at how Rally is used across all our readers to make the product better, and we do that in aggregated, de-identified form. That means patterns, not people: which features help a child stick with a book, where readers commonly get stuck, what a productive practice session looks like. De-identified means the information no longer identifies anyone and cannot reasonably be used to. Our goal is to support today's students and to keep improving Rally for tomorrow's students.

What every account has

In Rally's system all children are connected to a school and/or family. Before a child reads, Rally needs to know who they are and which classroom or family they belong to.

Through a school, that means a name, a username, a school-assigned email or student ID, a grade level, a school name, and the class group an educator sets up. At home, it means a name, a parent's email or phone, and a grade level. Either way, an account also holds a password and the child's age, plus two optional things a family or educator can add if they want to: a Lexile score, and a profile picture or avatar. We don't collect a child's phone number, home address, date of birth or gender.

We also collect the technical information any app collects: device type, operating system, app version, IP address, and when a session started and ended.

One thing worth noting about school accounts. Sometimes a teacher names a class group after a program, like an EAL or ELL group or a reading intervention. Rally stores that group name exactly as the teacher wrote it. We never ask about a child's EAL status or special education status, and we do not collect them anywhere else.

What we collect when a child reads

• Mode 1: Silent Reading. Minutes read, session start and end times, reading streak, badges, books requested, previewed, read, downloaded and returned, and where a child is in each book.
• Mode 2: Oral Reading with Feedback. Everything in Mode 1, plus: misread and omitted words, words read correctly, Sight Words mastered, accuracy, words per minute, words read, and Persistence.
• Mode 3: Oral Reading with Feedback and Screen Recording. Everything in Mode 2, plus screen recordings, which capture the page, the child's voice and what's on screen. Off by default. iOS only.

Why we collect it: To help each child see and accelerate their reading progress.

What Rally never does

These commitments hold whether a child uses Rally through school or at home.

• We never sell personal information. We do not sell, rent, trade or otherwise disclose it for value. Not a child's, not yours, not anyone's.
• We never use a child's information for advertising or marketing. Rally carries no third-party advertising. We do not let advertisers or data brokers collect anything about how a child uses Rally.
• We never build profiles of children beyond what it takes to run Rally. In school use, that means the educational purposes the school has authorized. At home, it means the reading experience the parent set up. Nothing beyond that.
• We never email or text a child.
• We never use personal information to train AI models. Nothing collected through Rally is used to train, refine, fine-tune, evaluate or otherwise develop any artificial intelligence or machine learning model. Not ours, not anyone else's. That holds for information from a child, a family member, an educator or a school. We do not disclose it to any third party for that purpose, and we do not permit any service provider to use it that way.
• A child's account cannot make purchases in the app.

If you would like more information, we encourage you to read our Canadian Privacy Statement.

How recording actually works

Screen recording is off by default. It turns on only when a parent, educator or student actively opts in, and it can be turned off at any time. Screen recording is available on iOS only, so if your child uses Rally on an Android device, there is no recording to turn on.

A recording captures the page the child is reading, their voice, and what's happening on screen as they read. There is no front facing camera, in any mode, ever. Recordings exist for three purposes:

• So that an educator or parent can listen to a read-aloud they could not be present for
• So that a child's oral reading can be documented over time for parent-teacher or student-teacher conferencing, and for assessment
• So that a child can hear their own reading. Sometimes that is a teaching tool for educators working on prosody, the rhythm and expression of reading aloud, the part that makes a sentence sound like natural speech rather than a list of words. And sometimes it is simply a chance for a child to hear how far they have come, how much more fluent they are now than they were at the start of the school year.

One clarification, because precision matters here. Rally Reader's speech recognition happens locally, on your child's device. Their voice is not sent to the cloud or to a data centre to be understood. Dashboard metrics and any opt-in screen recordings are a separate matter: those we do sync to the cloud, so an educator or parent can open them on their own device. Where the speech recognition happens and where the record lives are two different things.

If you are in Quebec, there is something you should know. Rally Reader meets or exceeds the standards set by PIPEDA, the federal privacy law. But we do not yet meet one requirement of Quebec's Law 25, the province's modernized private-sector privacy law. Law 25 requires that for a child under 14, consent to collect personal information, including a recording of their voice, comes from the parent, guardian or tutor. Two things stand between us and that standard today. A child can turn recording on themselves from inside the app. And the app does not ask a child's age or province before allowing it. This affects iOS only. Rally's Android product has no screen recording feature, so a child using Rally on Android in Quebec cannot enable it. If you are a Quebec parent or guardian and you don't want your child under 14 recorded, please leave recording off and don't let your child turn it on. If you are a Quebec school, please get parental or tutor consent before you enable recording for a student under 14. We will tell Quebec families and schools as soon as the control is ready.

Retention and deletion. Recordings are encrypted in transit and when stored. Who can see them depends on how the account was set up. In a school, a recording is visible to the student, their teacher and authorized school staff, and a teacher or administrator can extend that access to a reading specialist or to a family member they invite. School administrators may have access across their school's Rally Reader organization. At home, a recording is visible to the child, the parent or guardian who holds the account, and anyone that parent chooses to invite. In both cases our customer service team can reach recordings where support requires it, though this is rarely needed and rarely done.

You can request deletion of your child's data at any time. Where the request goes depends on how the account was set up.

If you signed up for Rally yourself and use it at home, write to us at Hello@RallyReader.com and put Delete in the subject line. We act on those requests promptly. The only exception is a legal requirement to keep something.

If your child uses Rally through their school, the school holds the student record and decides what happens to it, so start with them. The school can close the account or ask us to delete the information, and we act on that request just as promptly. If you're not sure who to ask, write to us anyway and we'll point you to the right person.

To be clear, deleted information is removed from our live systems right away, but copies can remain in backups until those expire on their normal cycle: 7 days for recovery logs, 30 days for daily snapshots, 98 days for weekly snapshots. Backups cannot be edited after they are written, so a copy inside one is deleted when the backup expires. Backups are held in Canada and used only to restore the service.

If you don't ask, here is what happens by default. We keep your child's data for as long as the account stays active, so their progress and history are there whenever they need them. If an account becomes inactive, we hold it for at least 90 days, so nothing is lost if your child comes back, and our practice is to delete it at 12 months. That practice is not a contractual commitment, and it can vary where a school agreement or the law requires something different.

A note to my fellow Canadians

Canadian student data stays in Canada. Rally Reader hosts data with Amazon Web Services and Google Cloud Platform, and stores each customer's data in the country that customer requires. Data belonging to Canadian customers is housed in Montréal, Quebec, in AWS's Canada (Central) region (ca-central-1) and/or Google Cloud's Montréal region (northamerica-northeast1). Reading progress, misread and omitted words, Sight Word records and any opt-in recordings for a Canadian child are stored on Canadian infrastructure, encrypted in transit and at rest.

A word about French. The Rally Reader interface is available in English only. There is no French-language version of the product at this time. The Rally Reader library does include a few French-language titles, which a child can read in French, without oral reading support, within an English-language interface.

We'd also like to say something about customer service, because it relates to data. Customer service matters to us. We don't outsource it, and we don't ask customers to talk to AI agents. Every inbound email and text is answered by an actual person who works for Rally Reader. Right now that customer service team is based in Canada, the United States and Europe, and sometimes it includes me. Because part of the team sits outside Canada, there are times when they need to temporarily access a Canadian account in order to resolve an issue or answer a question.

That access is limited, logged, and governed by the data protection commitments we make in writing to every school and family we work with. Your child's reading data lives in Canada.

If you are in Nova Scotia, there is something you should know. Nova Scotia's Personal Information International Disclosure Protection Act is stricter than PIPEDA. Unless an exception applies, it bars public bodies and their service providers from storing personal information outside Canada or letting it be accessed from outside Canada. We store Canadian information in Canada, but our support staff in the United States and Europe can temporarily access it, so we do not currently comply. One exception is available: the head of a public body can permit this where they determine it meets the operating requirements of the organization. If you are a Nova Scotia school considering Rally Reader, talk to us first and we will put the right agreement in place. This binds public bodies, so it does not affect a family using Rally at home.

Where we sit on compliance

• PIPEDA Rally Reader meets or exceeds all PIPEDA standards. We work to its ten fair information principles and meet its breach reporting requirements. Storage and support access are described above and in our Canadian Privacy Statement.
• FERPA Rally Reader meets or exceeds all FERPA standards. When a school adopts Rally, we operate as a school official with a legitimate educational interest, under the district's direct control. The data belongs to the district and the family, not to us.
• COPPA Rally Reader meets or exceeds all COPPA standards. In school settings, consent flows through the school, as COPPA permits. For family accounts, we obtain verifiable parental consent at signup and collect only what the product needs to function.
• Where we fall short Two standards we do not currently meet, both described earlier in this guide. Quebec's Law 25 requires that consent for a child under 14 come from a parent, guardian or tutor, and on iOS our app lets a child turn recording on without checking age or province. Nova Scotia's Personal Information International Disclosure Protection Act bars public bodies and their service providers from storing personal information outside Canada or letting it be accessed from outside Canada unless an exception applies. We store in Canada, but our support staff outside Canada can temporarily access Canadian accounts. We list these clearly, in the spirit of honesty and transparency.

A Note on Screen Time

Parents ask us two questions about AI and data. There is a third one that usually comes first: my kid already has too many screens. Why would I add another?

Fair question. Here is what is actually on the screen.

A page of a book. No ads. No autoplay. No infinite scroll. No recommendation feed engineered to keep a child watching. No comments, no followers, no strangers. There is no version of our business that improves when a child stays on a screen longer than the reading requires.

We do use streaks, badges and daily goals, and we will be direct about why. They exist to get a child to a finish line, not to move the finish line further away. The daily goal is typically fifteen minutes. When a child reaches it, Rally tells them they are done. That may make Rally one of the only apps on the device that celebrates stopping.

A child on Rally is reading real books, nothing AI-generated. It is not another thing competing with reading. It is reading.

A Note on Paper Books

We love paper books.

Rally Reader is run by book people. We love the beauty and tactile nature of physical books. There is strong research showing that paper books foster deep reading. ⁵ That experience matters and is wonderful. They don't require batteries.

Rally's role: We're a bridge.

Rally isn't here to replace paper books. Rally is a bridge to more reading.

How it works in practice:

• A child builds fluency on Rally and finds that reading has stopped being hard, so they pick up more of it. They discover an author on Rally and ask for the rest of the series in paperback
• A teacher uses Rally for daily oral reading practice, then sends kids home with paper books
• A parent uses Rally to help build their child's confidence enough that they want to read paper books without fear

The best literacy-rich homes and classrooms have both: digital practice when a supportive adult isn't available, and paper books for screen-free reading. The combination works.

What We Stand For

We named the company Rally because every kid deserves a Rally Squad: people who see their effort, celebrate their growth, and believe in their potential.

When a child sees their own progress, something shifts. They stop thinking of themselves as a struggling reader and start to own their learning. That is the whole point.

Skepticism about technology in education is warranted. Not all ed tech is created equal. So look at the research, and think about whether fifteen minutes a day of supported oral reading practice, in the context of 50,000 books, makes sense for your child. We're asking for your scrutiny because your questions help us do better.

This isn't about Rally Reader. It's about whether we'll give this generation the literacy skills and confidence they need to think critically, engage civically, and build the lives they want.

Thank you for reading. We'd love to hear from you at Hello@RallyReader.com. If you want your email to come straight to me, please put Andrea in the subject line.

Andrea JohnsonFounder and CEO, Rally Reader

Appendix: The Research and the Study

The two sections that follow are for readers who want the evidence in full: the literacy research Rally is built on, and the details of the independent efficacy study behind the results cited above.

The Research: Four Research-Backed Elements and Why Rally Reader Works

Rally offers a scalable, measurable, affordable way to follow established best practices. Here are the four that matter most.

1. Volume of Reading Matters. A 2024 meta-analysis of 47 studies ⁶ found that in-school independent reading produces measurable gains in word recognition and in how students feel about reading. Rally drives volume through 50,000+ popular books, daily goals, and reading streaks. Rally's student and teacher dashboards measure actual reading, not minutes in-app.

2. Oral Reading with Feedback. The National Reading Panel concluded that guided oral reading with real-time feedback produces significant gains across all grade levels. Rally delivers individualized feedback on every read-aloud session, at a frequency no single teacher could match.

3. Choice and Authentic Books. A meta-analysis of 49 studies, led by a researcher at McGill, ⁷ found that motivational reading interventions improve both reading achievement and reading motivation. Rally puts an enormous assortment of high-quality full books and authentic choices in the student's hands.

4. Low-Anxiety Practice. Research documents that reading anxiety disrupts reading. Rally's feedback is private, errors are corrected without an audience, and growth is celebrated.

The Study: Claremont Unified School District

We recently completed a third-party-validated efficacy study (ESSA Tier 3 criteria), and the results were truly excellent. Here are a few highlights.

• Who: 1,651 students in Grades 2-6
• When: 2024-25
• Measurement: i-Ready adaptive assessment (the assessment used by that district) and CAASPP, California's annual state summative assessment

The Headline Results

Grade 3 students who used Rally as designed grew 132% faster than their peers who did not. Results by Grade:

• Grade 3 (protocol-adherent students): +11.4 i-Ready points = 132% greater growth (p=0.002)
• Grade 5 (protocol-adherent students): +21.6 points on the state CAASPP summative assessment (p=0.048) and +20.0 i-Ready points (p=0.010)
• All grades (all users): +8.3 i-Ready points (p<0.001)

Across-the-Board Improvements:

• 100% of Rally classrooms outperformed non-Rally classrooms
• All 5 i-Ready metrics improved: literature comprehension, informational comprehension, vocabulary, word analysis, and phonics

Why This Matters

The graduation gap that follows a child out of Grade 3 is not a life sentence. The literacy gap is real but not inevitable, and a year of extra growth in the grade that matters most is exactly the kind of thing that changes a trajectory.