Skip to content
Optible Help Center home
Optible Help Center home

How Assessment Works

When an application is submitted, Optible runs a three-stage assessment pipeline. Each stage must complete before the next begins.

The three stages

Stage 1 — Document validation

Optible reads every uploaded document and extracts key values: amounts, dates, names, and so on. It also checks whether required documents have been provided and whether they meet the grant's requirements.

Stage 2 — Enriched fields

Using the values extracted from documents and the applicant's answers, Optible derives new fields that your scoring criteria need. For example, it might total up expense line items, classify a project type from a description, or pull a verified spend figure from a bill. These derived values are saved alongside the applicant's answers.

Stage 3 — Criteria scoring

With all the above in hand, Optible works through your scoring rubric and answers each yes/no question for every criterion. Scores roll up from individual questions to criteria, from criteria to groups, and from groups to an overall rank.

Key things to know

  • Order is fixed — documents are read first because enriched fields may depend on them. Enriched fields run next because criteria scoring may reference them. Scoring always runs last.

  • Everything is configured per grant — you decide which documents get validated, what enriched fields do, and what the scoring rubric looks like. Two grants in the same organisation can behave differently.

  • Reviewers always have the final say — every score Optible produces is a starting point. Reviewers can override any criterion. Their overrides are preserved when a re-run happens.

When does a run happen?

The pipeline runs automatically when an applicant submits their form. It also runs when a progress report is submitted, or when a reviewer manually triggers a re-run.

How long does it take?

Typically a few minutes for a standard application. Complex applications with many documents or many criteria may take longer. The applicant's view does not change during the run — scoring happens in the background.