There is a certain kind of business that exists because a specific person got tired of doing a specific chore, and NoRedInk is one of them. The chore was grading. The person was Jeff Scheur, a high school English teacher in Chicago, and by his own accounting he graded something like 15,000 papers before he decided this was not a sustainable use of a human weekend.
Here is the thing about grading essays that anyone who has done it will tell you, and that everyone who hasn't will underestimate: it is not that it's hard, exactly. It's that it's the same corrections, over and over, written in the margins of student after student, in ink that the student will glance at once and never read again. Scheur, teaching at Whitney Young Magnet High School, noticed that his students often couldn't actually decode the mistakes they'd made. The feedback was there. It just wasn't landing. This is the central problem of writing instruction, and it is less a pedagogy problem than a logistics problem, which is a useful thing to realize because logistics problems can be solved with software.
So Scheur did the two things that, in retrospect, made NoRedInk work. First, he built a taxonomy - years of accumulated data on the specific, recurring errors real students make. This is the unglamorous asset at the bottom of the whole company: not an algorithm, but a well-organized list of how writing goes wrong. Second, in his eighth year of teaching, he posted an ad on Craigslist looking for an engineer. His students voted on what to call the thing. They chose NoRedInk, which is both a name and a thesis statement.
NoRedInk's mission is to unlock every writer's potential.
The first version went out in February 2012 at an Illinois teachers' conference. It reached 15,000 users in two months, which is the sort of number that in consumer tech would be a rounding error and in the world of English teachers sharing a link with other English teachers is a genuine signal. In September 2012 it won the Citi Innovation Challenge, an NBC-hosted contest, for $75,000. In January 2013 it raised $2 million, with Google Ventures among the investors. The startup machinery, in other words, showed up - but it showed up to a product that teachers were already using, which is the correct order of operations and a rarer one than you'd think.
What it actually does
The clever part of NoRedInk, the part worth pausing on, is how it makes grammar practice not boring. It does this by letting students generate practice sentences about things they actually care about - their favorite athletes, celebrities, anime characters, or, in the platform's slightly chaotic best case, their own classmates by name. A subject-verb agreement drill hits differently when the subject is someone you sit next to. This sounds like a gimmick and is instead a strategy: engagement is not a marketing problem to be solved after the fact, it's a design input at the start.
It's worth being precise about why this works, because "personalization" is one of the most abused words in education software and usually means nothing. Here it means something specific and modest: the grammar rule being drilled stays fixed, but the sentence wrapped around it bends toward the student. The pedagogy is old-fashioned - deliberate practice on a discrete skill until it sticks - and the technology's only job is to remove the friction that makes practice feel like punishment. That's a smaller claim than most edtech makes, and smaller claims are the ones that survive contact with an actual classroom of fourteen-year-olds.
On top of that runs the adaptive machinery you'd expect - exercises that meet students where they are and offer scaffolding when they get stuck - and the part schools pay for: a full writing curriculum. NoRedInk Premium gives schools guided composition tools, more than 1,000 skills exercises, hundreds of texts and prompts, guided practice across 10-plus writing genres, and standards-aligned diagnostics and benchmark assessments. Practice can be filtered by alignment to state and national standards, which is the language districts speak. It plugs into Canvas and Schoology so that assignments, rosters, and grades sync where teachers already work, and into Clever for sign-on. There is also a Language Support Suite - text-to-speech and translation tools - so that English Learners can focus on ideas rather than fighting the interface.
The AI part, handled soberly
In 2024 NoRedInk did the thing everyone in education technology was under pressure to do, which is add AI, but it added it to the exact spot where the company started: grading. The AI Grading Assistant reads a student essay, evaluates it against the rubric items a teacher has chosen, and suggests scores plus specific, written, supportive feedback. The company expanded it to all Premium teachers for the 2024-2025 school year, starting with argumentative and literary-analysis essays and rolling out more genres over time.
What's notable is the framing. The pitch is not that the AI replaces the teacher's judgment; it's that it removes the reason teachers were rationing feedback in the first place. In beta, 95% of teachers said the tool increased how much feedback they gave, half said it greatly increased it, and grading time fell by roughly half. The line that should give any edtech skeptic pause: students whose teachers used the beta were over five times more likely to receive feedback on their writing than students whose teachers didn't. The bottleneck, it turns out, was never the students. It was the weekend.
Students whose teachers used the beta Grading Assistant were over five times more likely to receive feedback on their work.
The business, and a small surprise
NoRedInk runs the freemium playbook that fits schools unusually well. Individual teachers adopt the free version - personalized practice, a dashboard, quiz activities - and that free usage becomes the wedge for district-level Premium contracts. It's a slow motion to money, but it's a durable one, because the buying decision is downstream of teachers who already trust it. The company reports use in over 10,000 schools and roughly 1 in 2 US school districts, with students having completed billions of exercises. Third parties peg annual revenue around $17.5 million; the company raised a $50 million Series B in 2021, led by Susquehanna Growth Equity with True Ventures, about six years after its Series A. Total funding sits near $58 million.
Now the surprise, which is the sort of detail that tells you something about a company's character: NoRedInk is one of the larger open-source Elm shops in existence, with 200-plus public repositories on GitHub. Elm is a niche, rigorously typed functional language for building web interfaces, beloved by the kind of engineer who cares a lot about not shipping bugs. A K-12 writing company did not have to become a functional-programming reference. That it did suggests the same instinct that produced the error taxonomy - a belief that the boring, structural stuff is where the quality actually comes from.
Who's in the room
NoRedInk is not alone in the writing-and-grammar corner of edtech, and its neighbors are instructive. Quill.org is a free nonprofit doing interactive grammar and writing; Grammarly is the consumer-and-enterprise writing assistant everybody's already heard of; IXL offers a broad, standards-aligned language-arts curriculum; and Khan Academy sits underneath all of it as the free default for foundational practice. What distinguishes NoRedInk is that it is built around the classroom rather than the individual writer. Grammarly makes your sentence better right now; NoRedInk is trying to make you the kind of writer who fixes the sentence yourself next time, which is a slower and more teacherly ambition and also the one that districts sign purchase orders for.
That orientation shows up in who uses it and how they arrived. NoRedInk's customers are teachers, students, schools, and districts across grades 3-12, and the growth pattern is the classic education one: a single teacher tries the free tier, it works, they tell the department, the department tells the district, and eventually there's a Premium contract with admin dashboards attached. It is roughly the opposite of a viral consumer launch - slower, quieter, harder to fake, and much harder for a competitor to dislodge once it's embedded in how a district's English department runs its year.
The company itself is small for its reach - around 80 people - and headquartered in San Francisco, with a distributed, remote-friendly culture. The through-line from the classroom origin is visible in the product decisions and, more surprisingly, in the engineering, where a genuine craft culture produced that sprawling open-source Elm footprint. Jeff Scheur still runs the company as CEO, which matters here: the person who graded the 15,000 papers is still the person deciding what the software should do about it.
That's the whole company, really, followed to the end of its own logic: writing is a skill, skills need practice, practice needs feedback, feedback needs time teachers don't have - so build the software that gives the time back. Most companies stop somewhere in the middle of that sentence. NoRedInk's entire value is in finishing it.