Finding the ‘next Max Verstappen’ with data and smart algorithms.
Van Amersfoort Racing
Report.IQ
AI for document analysis.
Assess large volumes of documents against a set of standard questions and criteria.
Trusted by leading organisations
Why we built this,
Grant applications, permits, reports and audit documents contain a great deal of information that needs to be checked manually. Staff look for the same data, compare passages against assessment criteria and record their findings separately.
This takes a lot of time and makes it difficult to handle documents consistently. As the number of applications or reports grows, so does the pressure on experienced reviewers.
Report.IQ analyses documents against predefined questions and criteria. The application finds relevant information, structures the findings and shows where further review is needed.
What you can expect,
Review documents using a consistent approach
Report.IQ processes documents based on questions or assessment criteria agreed in advance. This means staff do not have to read every document in full to find the relevant information.
The application can be configured for different document workflows, provided the purpose and assessment method are clearly defined.
Find and structure information
The information found is organised by question or criterion. This gives staff an overview they can use for further assessment.
The analysis can, for example, help with:
- Finding required information
- Checking standard sections
- Comparing documents against assessment criteria
- Flagging missing or unclear information
Human review for uncertainties and exceptions
Not every conclusion can be drawn from a document alone. Unclear wording, conflicting information or exceptions still require subject-matter expertise.
We therefore configure the application to make uncertainties visible, so staff can check the relevant information.
Our approach,
-
Define the scope of the assessment first
We map out which types of documents are processed and which questions staff answer when reviewing them. We also determine which parts are sufficiently clear-cut for Report.IQ to prepare. This means we automate clearly defined analysis tasks, not an entire role.
-
Align documents and criteria
We configure the application using representative documents and the applicable assessment criteria. We then test the analyses with subject-matter experts.
We check whether it finds relevant information and makes clear when an answer cannot be derived from the document. -
Start with one document stream
We start with one type of report, application or assessment. Once validated, Report.IQ can be expanded to cover additional document types, questions or assessment criteria. We also agree on how it will be managed, so changes to documents or criteria can be handled in a controlled way.
Not an automated final decision-maker, but a focused application that structures the preparatory analysis.
Report.IQ supports the preparatory work, while the substantive decision remains with the staff member.
What we've built before,
Three projects from our work with data and AI.
Further reading
Blogs and in-depth articles from our team.
24 Aug 2026
Three lessons from our AI projects
What we learned from dozens of AI projects: start with the decision, treat data quality as an organisational issue and arrange management in advance.
18 Aug 2026
From pilot to production: why most AI projects stall
Most AI projects stall not because of the technology, but because of data, teams and processes. Read what it takes to move from pilot to production.
25 Jun 2026
Responsible AI: from principles to practice
AI is developing rapidly. New models emerge almost every week, and more and more organisations are experimenting with AI.
Get in touch
Stef Stoelwinder
Business DevelopmentE-mail info@datasciencelab.nl
Phone +31 (0)20 3636 724
Frequently asked questions,
Report.IQ can be configured for clearly defined document workflows, such as reports, grant applications, permit applications and audit documents. Whether it is suitable depends on the structure of the documents, the questions you want to ask and the criteria used in the assessment.
No. It helps you find relevant information, structure findings and flag missing elements. The final assessment remains with an authorised staff member when interpretation, additional context or a formal decision is needed.
Report.IQ can be configured around predefined questions and criteria. If these vary by document type or process, they must be documented and tested separately. Any subsequent changes to the criteria must be incorporated into Report.IQ in a controlled way.
A broad instruction such as ‘assess this document’ leaves considerable room for interpretation. Specific questions and criteria make it easier to check what information Report.IQ needs to find and how staff can assess the results. This makes the results more useful and helps reveal errors sooner.
That depends on the complexity, the cost of errors and the consequences of the outcome. Finding information is generally easier to support than making a substantive or legal judgement. The more interpretation a conclusion requires, or the greater its consequences, the more important human review becomes.