Time-consuming reading
Relevant passages are scattered across extensive documents.
The module prepares reading and comparison in one consistent structure. Evaluators can work more consistently and verify every finding in the original document.
A building block within iSyNAPS
Goals, roles, working methods and validated knowledge provide the context for this building block. During onboarding, we agree what is needed. That profile can then grow with further applications, within the permissions agreed for your organisation.
Explore the ecosystem ↗The challenge
Evidence is spread across documents and suppliers structure responses differently. Evaluators must still apply the same published criteria and explain every score.
Relevant passages are scattered across extensive documents.
Evaluators may read requirements and wording differently.
Scores and reasons must be carefully traceable to the requirements and tender.
Benefits to test
In a pilot, test whether relevant passages are found, tenders follow the same comparison structure and findings remain traceable.
Use the same structure for every tender.
Review suggested findings in the original document.
Evaluators determine the score, reasoning and decision.
Who it is for
For configuring a consistent evaluation and source review.
For comparing submitted answers per criterion.
For reviewing the process, reasoning and use of supporting technology.
For a more transferable record after the evaluation is completed.
How it works
The module prepares the comparison within the established evaluation process. Expert evaluators review every finding and make the decision.
Import or enter the approved criteria and priorities.
Use only documents that may be processed under the applicable procedure.
Build relevant passages, draft comparisons and source references.
Have evaluators review every finding and approve the final score and reasoning.
Capabilities
Make tenders searchable, compare responses per requirement and retain source references. We agree supported document types in advance.
Work from the approved requirements and one consistent set of criteria.
Make text from supported tender formats searchable.
Compare findings per criterion and supplier.
Connect a finding to the relevant passage in the submitted tender.
Allow experts to amend, reject and justify findings.
Record who changed an evaluation and why where the pilot capability supports it.
Frequently asked questions
No. AI output is supportive. Only authorised evaluators can approve scores, reasoning and the final decision.
No model score is inherently objective. Criteria, input, interpretation and model behaviour can all affect it. Source checks and human review are therefore essential.
This varies by pilot version. Formats, maximum size and extraction quality are tested in advance.
Not until the purpose, access, hosting, retention, confidentiality and other procurement and privacy conditions have been agreed in writing.
OfferteBeoordeling is being tested further with municipalities as a pilot. Availability and conditions are confirmed for each pilot.