Review & release
Every result waits for a person
Read the evidence, override a criterion, then release. The respondent sees nothing before that.
Competency sign-off for training cohorts
It reads each response against your rubric, cites the lines it scored on, and holds every result until your assessor releases it.
1 assessment · 10 evaluations a month · No card
The Flow view
Show or hide questions based on previous answers. The Flow view draws every route a respondent can take, so you can check the path before anyone sees it.
Conditional branching is on Starter and above.
The pipeline
Every mark on Scorafy goes through the same seven steps. You can inspect all of them.
Your criteria, levels and weights. Up to 10 criteria and 6 levels.
Written, spoken, recorded or uploaded, across 26 question types. No account needed.
Claude by Anthropic reads each answer against your rubric. Text in an answer is treated as an answer, never as instructions.
Each criterion gets a level, the evidence from the answer, and how sure the model is.
Your assessor checks, overrides and comments. The AI score stays on record beside the override.
The score reaches the respondent only when an assessor releases it. Who released it, and when, is logged.
The respondent gets their report. The audit pack (PDF and JSON) and your Accuracy page keep the record.
Assessment should be fast and fair. Scorafy does the reading and the first draft. Your assessor keeps the judgement.
Built for accountability
Review & release
Read the evidence, override a criterion, then release. The respondent sees nothing before that.
Live rooms
Everyone joins by QR. You move them on.
Evidence
The words in the answer that earned it.
Data residency
Hosted in the EU, or in Australia under a residency addendum.
Audit trail
Who scored it, who overrode it and who released it, with times.
The hidden problem
On a borderline response, judgement fills the gap the rubric left. Scorafy does not remove the assessor. It hands them a first draft with the evidence already attached.
Assessor A
Read the answer generously. Counted the QR code as an invite link.
Assessor B
Read it strictly. No integration named, so no marks for it.
Same answer, same rubric · illustrative example
Markers disagree on open answers. When England's exam regulator checked exam marking against senior examiners, the chance of a candidate getting the “definitive” grade ranged from 0.96 for a maths qualification to 0.52 for an English language and literature qualification. Ofqual notes some answers have more than one legitimate mark. (Ofqual, 2018)
Nothing reaches a respondent until you release it.
The AI drafts. Your assessor decides. Every override is logged with who changed what, and why.
From a live pilot
An online education non-profit is piloting Scorafy in Australia for in-program assessments with adult learners. Every answer is scored against their own criteria, and an assessor signs off before a learner sees a result.
“Both respondents and staff have found the Scorafy recommendations highly actionable, and the platform itself very easy to navigate and use.”
“Feedback has been well received and immediately actioned by Scorafy.”
“As the Product Lead, I have found working with Scorafy to be brilliant - it's been an exciting and dynamic process”
Free sample marking
Send 10 anonymised answers you have already marked, plus your marking guide. We mark them against your own criteria and send them back with the evidence behind every mark. No sign-up, no card.
Pricing
Start free. Scale to cohort and enterprise volume when you're ready.
1 assessment, 10 AI evaluations per month. No card needed.
New yearly allowances apply from 1 January 2027. If you subscribe before then, you keep your current allowance until your first renewal after that date.
Up to 3 team members, plus unlimited assessors
For a single programme or team
Up to 10 team members, plus unlimited assessors
For training and enablement teams running several cohorts
Unlimited team members and assessors
For programmes and companies certifying at volume
For organisations
From US$12,000 a year
A Scale response pool plus the service around it, for teams running assessments at scale. Enterprise is quoted.
Talk to usNot ready to commit? Start with a paid pilot: US$3,000 for 90 days and up to 300 responses, credited in full to your first year.
Calibration service: US$900 per assessment. We mark up to 50 of your already-marked answers, report agreement by criterion, and tune the rubric before release.
All prices in USD. One marked response is one completed respondent assessment, evaluated. Annual plans get the whole year's allowance up front. When the allowance is used, marking pauses until you move up a plan or buy a top-up - no surprise bills. Scale plans (annual only) from US$5,400 a year and top-ups from US$150 - see full pricing. Not-for-profits and accredited education providers get 40% off the software.
Questions
One AI evaluation is one completed respondent assessment, evaluated. It is counted when someone completes your assessment and Scorafy generates their report. Partial or abandoned attempts do not count towards your limit. A resit is a new attempt, so it is a new evaluation, and so is regenerating a report. Appeals and changes an assessor makes to marks never count.
No. Respondents click a link and start the assessment. No signup, no friction. If an assessment is timed, respondents see a start screen first - the clock only starts when they press Start - and if they return after finishing they see a completion screen rather than dropping back into the questions.
When your assessor releases them. Scorafy holds every result - score, evidence, and report - until a human reviewer signs off. You can override any score before release, and overrides are kept on the record.
Yes. Set a time limit for the whole assessment, individual questions, or both. The whole-assessment limit is enforced on our server, so it holds even if a respondent's browser is manipulated; per-question limits are too in live sittings.
We use Claude by Anthropic. Every report is generated fresh from the respondent's specific answers - not matched to pre-written text buckets. The AI's score is a first draft - your assessor reviews and releases every result. See how scoring works at /how-scoring-works.
Yes. In an assessor-paced live sitting you open one question at a time and everyone in the room answers it together, with instruction screens for briefing and pause-and-discuss beats in between. Sequencing is enforced on the server - the content of a question is not sent to anyone’s browser until you open it, and the platform refuses answers to questions that have not been opened yet. Candidates join by link or QR code, and live sittings are available on every plan, including free.
Every mark carries the model’s own confidence for that question, and every criterion carries an explanation of which rubric level was awarded and why. Reviewers work the low-confidence marks first rather than reading every result at the same depth. Your organisation also gets an Accuracy page showing how often your assessors agree with the AI, which questions attract the most overrides, and how each assessor moves scores.
Yes, if you want that. An assessment can be set to share the written feedback immediately while holding the numeric score back until an assessor signs it off. The score is withheld server-side - it does not reach the browser, and it is kept out of the AI’s written prose too.
Yes. You can add context, frameworks, rubrics, and grading schemas. The AI uses these to mark against your own standards, in your own terms.
No. All plans allow unlimited questions. We recommend 8 - 25 for the best balance of respondent experience and AI analysis quality.
All data is encrypted in transit and at rest, with row-level security isolating every organisation’s data, and MFA available on all accounts. You control response retention per assessment, from 30 to 365 days. We never use your data to train AI models. Full detail on our security page at /security.
Choose an assessment, answer it, and read the marking Scorafy writes back. No card needed.