A Cathoven alternative, compared honestly
Cathoven is the most expensive of the tools people compare us with, and also one of the most transparent about what its scores are not. Its help centre says plainly that it is not an official test centre and that its scores are predictions. If you are deciding between us, the two questions that matter are how much you want to pay to start, and what you want published about how the marking was checked.
Cathoven facts last checked 14 September 2026 · every claim links to its source
Side by side
| Cathoven | BandNine | |
|---|---|---|
| Modules marked | Four modules available. Its help centre says AI band scoring covers Writing and Speaking | Writing, Speaking, Reading and Listening |
| Entry price | $3.00 a day with a three-day minimum, or $21.00 a week | Free tier, or $12 once for 15 days that does not renew |
| See a score without paying | No. The IELTS product is paid after onboarding | Yes. Three writing and three speaking scores on the free tier |
| Account needed to start | Yes. Google or email registration, no guest access | No card and no password. An email for the report |
| Published accuracy measurement | 98 per cent, with no sample size, method or dataset published | Re-measured weekly, published with the sample size and the limits |
| Named, credentialed people | No individual named on the About page | Two DELTA-qualified teachers, named, with profiles |
What Cathoven does well
This section is here because a comparison written by one side is worth nothing without it. Everything below is something we think they genuinely do well, and in two places better than we do.
- Detailed sample reports you can read in full without logging in. You can see exactly what you would be buying before you pay, which is rarer than it should be and better than our own examples page.
- An unusually candid help centre. It says outright that Cathoven is not an official test centre, that its scores are predictions, and even that real exam performance often lands around half a band below its own scores. Very few products in this market volunteer that.
- Cathoven's pay-per-day option at $3.00 with a three-day minimum does not auto-renew, which is a genuinely low-commitment way in if you have a test next week.
- A clear separation between what renews and what does not, plus a documented 14-day cancellation right.
- Handwritten essay checking, so candidates sitting the paper test can photograph what they actually wrote.
- Inline numbered annotations on your own text, each with a named error type and a corrected rewrite.
- An unusually open description of its scoring architecture, which most products treat as a secret.
- A documented account-deletion route with concrete in-product steps.
Stay with Cathoven if
- You want to see a full, detailed sample report before paying anything. Cathoven does this better than we do.
- You have a test in a few days and want the lowest possible commitment, in which case pay-per-day is a sensible way in.
- You want a personalised study plan with daily assignments and are willing to pay for it.
Where we are different
One accuracy figure, said the same way everywhere
Cathoven's 98 per cent appears as independently validated across four skills on its marketing pages and as its own internal testing in its help centre, and no sample size or method is published for either. We publish one number, from one method, re-measured weekly, with the count of essays beside it and the limits written underneath. You do not have to take our word for it, which is the entire point.
You can try it before paying anything
Three writing scores and three speaking scores on the free tier, plus a three-minute level test that needs no card and no password.
All four modules actually scored
Including 81 complete reading tests, free, with a written explanation on every question rather than an answer key.
Named teachers
Two DELTA-qualified teachers, named, with profiles and bylines on the guides they wrote.
The accuracy figure, the method behind it and the limits we put on it are all on the scoring accuracy page, updated every Saturday. It is written by Manish Sharma and Anjali Tripathi.
Where these Cathoven facts came from
Each of these was read on Cathoven's own pages on the date shown. Nothing here came from a review site, and nothing came from behind a login, because we did not create an account or pay. If any of it is out of date, tell us and we will fix it.
Cathoven's published plans are $21.00 a week and $67.50 a month, both auto-renewing, a $172.80 three-month plan that does not renew, and a pay-per-day option at $3.00 a day with a three-day minimum.
www.cathoven.com/pricing/ · checked 14 September 2026
The IELTS product is paid. After onboarding, practice requires a plan, and any free reports are a one-time allowance that does not refresh.
“Free reports on Cathoven are a one-time allowance”
cathoven.zendesk.com/hc/en-us/articles/47591142422163 · checked 14 September 2026
Cathoven's marketing pages claim 98 per cent accuracy under independent validation across all four skills, while its help centre attributes the same figure to its own internal tests and says the system currently scores Writing and Speaking.
“Independent validation shows the AI band scores match real examiner scores with over 98% accuracy across Writing, Speaking, Reading, and Listening.”
www.cathoven.com/ielts/ · checked 14 September 2026
No sample size, dataset, error metric, method or validating body is published for that figure on the marketing site, the help centre or the pricing page.
www.cathoven.com/ielts/ · checked 14 September 2026
To its credit, Cathoven states clearly that it is not an official IELTS test centre and that its scores are practice predictions rather than official results.
“Our scores are practice predictions, not official IELTS scores.”
cathoven.zendesk.com/hc/en-us/articles/47592396997523 · checked 14 September 2026
Questions about accuracy
Is Cathoven 98% accurate?+
Cathoven's marketing pages state 98 per cent accuracy under independent validation across all four skills. Its own help centre attributes the same figure to internal testing. Those two descriptions of where the number came from are different, and the validating party is not named on either page. No sample size or method is published. We have not tested their scorer and make no claim about whether 98 per cent is right or wrong.
Why does BandNine not publish a percentage accuracy?+
Because on a band scale a percentage does not mean anything on its own. Accurate to within what, over how many scripts, marked by whom? We publish a mean absolute error in bands with the sample size and the confidence interval, so the number can be argued with. It is currently measured against scripts labelled in house, not by a certified IELTS examiner, and we say so on the same page as the figure.
Try it before you move anything
Three writing scores and three speaking scores free, or a three-minute level test with no card and no password.