Drafted with AI assistance and reviewed, fact-checked and edited by hand before publishing - see our AI transparency page for how we use AI across the app.
At Spelling.Live, we're continually improving the platform based on feedback from parents, tutors and schools.
One area we've recently enhanced is assessment history and data quality controls.
Most spelling attempts are marked accurately, but no assessment system is perfect. A child may be distracted, handwriting recognition may occasionally misread a letter, or a photographed practice sheet could be submitted accidentally. When that happens, the result may not reflect what the child can genuinely spell. Until now, there was no way to stop that single attempt influencing every accuracy figure and practice recommendation going forward.
That's why we've introduced Practice history: the ability to review every individual spelling attempt and exclude specific results from progress reporting, while keeping the original record fully visible.
How it works
Open a child's Reports screen and you'll find a new History tab. It lists every attempt they've made - written, typed, spoken, chosen, or photographed from paper - most recent first, each with a checkbox next to it. Checked (the default) means the attempt counts. Unchecking it excludes that one attempt from the accuracy report, the tricky-word ranking, and the adaptive-difficulty and spelling-pattern signals that drive what your child practises next - and it says so plainly on the row: "Excluded - not counted in reports or plans."
There's also a Writing view inside the same screen, for any attempt that has a handwriting or photo capture attached to it - you can see exactly what your child wrote, replayed stroke by stroke or as the photographed page, next to the same exclude checkbox.
Two things were important to us in how this was built:
- Nothing is ever changed or deleted. We don't believe in quietly editing history. The original attempt - what was written, what mark it received, when it happened - stays exactly as recorded. Excluding it only changes whether it counts, not what it was.
- It's an adult decision, not a child one. Only the account owner, or someone they've explicitly given edit access to a child (a co-parent, a tutor, a teacher), can exclude an attempt. A child's own login can see their own history, but can't edit what counts in it.
One honest limitation, stated plainly rather than glossed over: excluding an attempt keeps it out of the accuracy report and the pattern signals from that point forward, but it doesn't reach back and rewrite the revision schedule that attempt already nudged. We chose not to unwind that retroactively, because doing so safely after the fact isn't something we could guarantee - and a schedule quietly re-derived from scratch is a worse kind of opacity than the one we're trying to fix.
Why we built it this way
We believe transparency is important. Rather than changing or deleting historical results, Spelling.Live lets you decide whether a particular attempt should contribute to progress calculations - and leaves a plain record, on the row itself, of that decision.
During our research, we found that while many spelling platforms provide progress tracking and reporting, tools for reviewing and managing historical assessment data appear far less common. Once a result has been recorded, there is often limited ability to decide whether it should continue influencing future reports and recommendations.
For parents, tutors and schools, this means greater confidence in the reports you actually use to decide what to practise next. For learners, it helps ensure that trends and achievements reflect genuine performance rather than the occasional distracted session or misread word.
As we continue to develop Spelling.Live, our goal remains the same: combine the benefits of AI with the oversight and judgement of the parents, tutors and teachers who know a child best. AI marks the attempt; a person decides whether it's a fair one.
Because when it comes to learning, accurate data matters.