This system map is part of the assessment appeal resolution guide. The main guide explains why appeal records matter; this page shows where each record lives and what holds a case together.
The systems that hold one appeal, and who runs each
Most organizations that mark written or practical assessments split appeal handling across three to five systems. The table is the map in list form: one row per system, the record it holds, the key it shares with its neighbors, and the team that administers it.
| System | Record it holds | Linking key it shares | Usual owner |
|---|---|---|---|
| Assessment platform | Candidate response, attempt, sitting date, per-criterion scores, marker comments | Candidate ID, attempt ID | Assessment operations or exams team |
| Scheme or content repository | Scheme versions, examiner instructions, effective dates, change notes | Scheme ID and version number | Assessment design or standards team |
| Appeal case manager or ticketing tool | Appeal form, stated reasons, filing date, board minute, decision letter | Case ID, plus the attempt ID if the form asked for it | Academic operations or quality team |
| Gradebook or score history | Revised score, who changed it, when | Attempt ID with an audit entry | Platform administrator |
| Email or shared drive | Second-examiner notes, board papers, candidate correspondence | Whatever people typed: candidate number, name, date | Individual staff, unless filed |
The owner column matters for two reasons. It tells the advisor who can describe each system accurately, and it shows that nobody on the list owns the whole case. Scheme design, platform administration and the appeal process often sit in different teams.
The linking keys, join by join
A key is only useful if it was recorded on both sides at the time. Here is what each join depends on and what happens without it.
| Join | Key on one side | Key on the other side | If the key is missing |
|---|---|---|---|
| Candidate to attempt | Candidate ID in the platform | Candidate ID or email on the appeal form | Appeals cannot be matched to the sitting they dispute |
| Attempt to scheme version | Scheme ID and version stored with the scores | Version number in the repository | Nobody can show which criteria were in force when the mark was given |
| Attempt to appeal case | Attempt ID, if captured on the form | Case ID in the case manager | Appeal outcomes float free of the scores they changed |
| Decision to revised score | Score change with grader and timestamp | Board decision and minute | A changed score looks unexplained, or a decision looks unapplied |
| Scheme change to cause | Change note on the new version | Board recommendation in the minute | The revision cannot be traced to the appeal that prompted it |
Score history is the easiest join to check, because some platforms document it. Moodle, for example, has a Grade history report that filters grading changes by student, grade item and grader, can show only revised grades, and downloads as CSV or a spreadsheet. The same documentation notes that administrators can switch grade history off or keep it only for a set period, so whether the history exists for older sittings is a configuration question, not a given.
The scheme-version join is harder. Moodle documents its rubric total as the sum of the level picked on each line, which makes the point general: a number of points is meaningless once the lines behind it have been edited. Products differ on whether an old version stays attached to the attempts it scored, so confirm with the vendor's own documentation instead of assuming it.
One fully joined appeal case (Illustrative)
A fictional 260-person IT training and certification company runs a practical lab exam. Each candidate's lab is scored on three tasks, 3 points per task, 9 in total, with 6 needed to pass. Every ID below is invented to show the shape of a joined case, not a real record.
| Step | System | Record | Key that carries the join forward |
|---|---|---|---|
| 1 | Assessment platform | Attempt A-2291 for candidate C-58014, lab sat on Feb 9 | Attempt A-2291 |
| 2 | Assessment platform | Task 2 scored 1 of 3 under scheme R-12 version 3; total 5 of 9 | Scheme R-12 v3 |
| 3 | Scheme repository | R-12 v3 task 2 names one configuration method only | Scheme R-12 v3 |
| 4 | Case manager | Appeal case AP-0417 filed Feb 23, claiming an alternative method also meets the requirement; form cites A-2291 | Case AP-0417 citing A-2291 |
| 5 | Email, filed to case | Second examiner's note supporting the alternative method | Case AP-0417 in the subject line |
| 6 | Case manager | Board minute, Mar 8: method accepted; task 2 rescored 3 of 3 | Case AP-0417 |
| 7 | Gradebook history | A-2291 revised from 5 to 7 of 9, pass; grader and timestamp recorded | Attempt A-2291 |
| 8 | Scheme repository | R-12 v4 adds "or an equivalent configuration method"; change note cites AP-0417 | Scheme R-12 v4 citing AP-0417 |
Read down the last column and the case holds together: every step names a key that the next step also recorded. Remove step 8's citation and the scheme change has no recorded cause. Remove the attempt ID from the appeal form at step 4 and someone has to match the appeal to the attempt by name and date, by hand.
Where the keys usually break
Most appeal histories are partly joined. The breaks tend to fall in the same places, and naming them helps an advisor describe the history honestly.
- Typed identifiers. If staff copied candidate numbers into a ticket by hand, a share of them will be wrong or missing. Ask whether the case tool pulls IDs from the platform or relies on typing.
- Overwritten schemes. A repository that edits in place keeps today's lines and loses the ones a disputed score was based on.
- Board papers outside the case tool. Minutes saved to a personal drive or sent by email are often the only record of reasoning.
- Score edits without an audit entry. A revised score with no grader or timestamp cannot be tied to the board decision.
- A second request type for U.S. schools. The federal education records rule (34 CFR 99.20) gives families and eligible students a route to request a correction, with a hearing offered if the school declines. A vendor that tests on behalf of such schools may hold those correction requests next to ordinary appeals; give them their own node on the map.
None of these rules a company out. A history with three solid joins and one weak one is still a clear thing to describe.
Mapping questions for the head of academic operations
The head of academic operations can usually answer these without opening a single record. Keep every question at metadata level: what exists, how much, how long, which system, who owns it.
- Which systems hold attempts, marking schemes, appeals and score changes, and since when?
- Does the appeal form capture the attempt ID, or only the candidate's name and number?
- Are superseded scheme versions kept, with effective dates?
- Is score history switched on, and for how many years does it go back?
- Roughly how many appeals are filed a year, and how many change a mark?
- Who administers each system, and who owns the appeal process overall?
The answers become the system map in a referral brief. Who can approve a licensing discussion is a separate question, covered in who can license assessment appeal records. A data audit uses the same metadata as its starting point, and more context for the sector is on the education and training page.
The introduced company needs 50+ full-time employees at peak (contractors excluded), documented operating history, rights to license and an authorized contact. Companies in the U.S., Canada and the EU qualify, as do companies in the UK, Australia, New Zealand, India, Singapore, South Africa, Latin America, the UAE and Saudi Arabia when at least 70% of the data is in English. Partners sign the Channel Partner Agreement before submitting introductions. What stays unknown until SourceX reviews it: whether the records can be licensed under candidate terms and privacy law, how much de-identification is needed, and whether any buyer wants them.
Next step
Refer a company. No cap on rewards. Default reward on a company's first qualifying transaction: the greater of $50,000 plus 5% and 15% of Actual Supplier Payout, with no cap; further qualifying transactions first agreed within 12 months of the first Closing pay 15%. Eligibility and payment conditions apply, including the USD 100,000 Minimum Transaction, and a signed Schedule A or addendum can differ. A lead, meeting or signature alone does not trigger payment.
If the head of academic operations can name the systems and confirm that at least the attempt and case IDs join, register as a partner, check the company on who qualifies and read the program terms before you make the introduction.