Due diligence questionnaires are designed to help firms understand the risks associated with counterparties, service providers and other third parties. The challenge is not only getting a questionnaire completed, but also making sure nothing important is missed during the review process. A questionnaire may contain hundreds of responses, many of which are routine and require little further investigation, while others may raise questions that deserve closer attention.
For compliance and oversight teams, that creates a familiar problem: significant time can be spent working through responses line by line, even though only a relatively small proportion may ultimately require follow-up. Zeidler Group’s new AI Questionnaire Review, built directly into Zeidler Due Diligence (ZDD), is designed to make that process more focused by helping reviewers identify where their attention is most needed.
One of the most time-consuming aspects of DDQ review is the sheer volume of information that has to be worked through. Reviewers need to check whether questions have been addressed, identify missing information, notice unclear responses and distinguish routine answers from those that may require further investigation. On a large DDQ, this can mean spending considerable time on responses that ultimately require no further action, while more significant points risk being buried among hundreds of otherwise unremarkable answers.
AI Questionnaire Review helps reduce that scanning burden. At the click of a button, the feature analyses the questionnaire and assigns each response one of three statuses: Satisfactory, Flagged or Unanswered. Flagged and unanswered responses are brought together in a dedicated panel, allowing reviewers to navigate directly to areas that may require closer attention.
This can be particularly useful where an issue is easy to understand once seen, but easy to overlook within a large questionnaire. A statement such as “We do not maintain a cybersecurity policy”, for example, may be perfectly clear; the challenge is ensuring that it is noticed and considered in the context of the wider assessment. By surfacing these responses, the AI allows reviewers to spend more time assessing significance, requesting evidence and deciding whether an issue warrants escalation.
Prioritisation is only useful if the reviewer understands why a response has been highlighted. A generic warning or risk score can create additional work if the reviewer then has to reconstruct the reasoning behind it.
AI Questionnaire Review therefore provides a short, plain-English explanation alongside each assessment. A response might be flagged because it does not fully address the question, because important information appears to be missing, or because the content may indicate a potential compliance, regulatory or operational concern.
This gives the reviewer something concrete to assess. A Flagged status is not a conclusion that the counterparty has failed the assessment; it indicates that the response merits closer consideration. The reviewer can then decide whether clarification is needed, whether the concern is addressed elsewhere or whether further action should be taken.
Identifying an issue is only one part of due diligence. Once something has been highlighted, the reviewer needs to decide what further information is required. A vague response may need clarification, a statement about a control may require supporting evidence, and an operational arrangement may raise further questions about oversight or contingency planning.
By bringing potentially problematic responses together, AI Questionnaire Review can make that follow-up easier to organise. Instead of completing a full read-through and then separately compiling a list of outstanding points, reviewers can work directly from the areas that have been highlighted. If a counterparty states that important activities have been outsourced but provides little detail about oversight, for example, the reviewer can quickly determine whether more information is required.
The AI does not prescribe the follow-up or make the final assessment. It helps reviewers identify where those decisions are needed and move more quickly into the substantive part of the review.
DDQ review is rarely completed in a single round. Reviewers may request clarification, ask for supporting documentation or receive updated answers, all of which then need to be assessed.
AI Questionnaire Review can be re-run when responses or supporting information are updated, allowing the latest version of the questionnaire to be reviewed again. This can make subsequent review rounds easier by helping the reviewer reassess areas of concern without having to approach the questionnaire as if starting from scratch.
This reflects the reality that due diligence is often iterative. Questions are clarified, evidence is added and issues may be resolved over several rounds. Supporting that cycle is therefore just as important as improving the first review.
Another challenge for due diligence teams is maintaining consistency across reviewers. Different levels of experience, areas of expertise and familiarity with a counterparty can influence what stands out during an initial review.
Applying the same first-pass analysis across a questionnaire can provide a more consistent baseline. Reviewers begin with the same structured view of responses that may require attention, while still applying their own judgement to the significance of those issues.
This can be particularly useful for organisations managing large volumes of DDQs across different teams or offices. The aim is not to produce identical decisions, but to make the initial review more structured and reduce the risk that important issues are identified inconsistently.
There is also a practical question when introducing AI into an existing compliance process: does it simplify the workflow, or create another system to manage? If questionnaires need to be exported, analysed elsewhere and then brought back into the due diligence record, much of the efficiency gained can be lost.
AI Questionnaire Review is built directly into ZDD, so the questionnaire, AI assessment, flagged responses and wider review functionality remain within the same environment. Reviewers can move directly from the overview of potential issues to the relevant response without switching systems or maintaining a separate record.
That matters because efficiency is not only about completing individual tasks faster. It is also about reducing the unnecessary steps around them.
The value of AI in DDQ review is not in replacing the reviewer, but in reducing the time they spend searching for the issues that require their attention. By scanning the questionnaire, surfacing potential concerns and explaining why they may matter, AI Questionnaire Review helps reviewers spend more of their time on investigation, judgement and follow-up — the parts of due diligence where their expertise matters most.
Discover how Zeidler Due Diligence can help your team reduce manual DDQ review, identify potential risks faster and focus attention where it matters most.
