Knowledge work automation
15 min read
—
Imogen Jones
Content Writer
A due diligence questionnaire is the primary structured information exchange between a private equity fund and the institutional investors considering allocating capital to it.
When a limited partner (a pension fund, endowment, sovereign wealth fund, or fund of funds) evaluates a new manager relationship, they don't rely on the deck. They send a DDQ: a structured set of questions organized into sections covering the firm's organizational structure, investment process, team composition, risk management, operations, compliance history, ESG practices, track record, and fee terms. The fund manager answers every question with supporting documentation and returns it. The LP's investment and operational due diligence teams analyze the responses. Then they decide whether to proceed.
This sounds like a clean information exchange. In practice, it's a significant operational burden on both sides of the table.
Institutional LPs each maintain their own DDQ format, their own section structure, and their own depth of questioning. A fund manager with active relationships across 50 LPs may receive DDQs from 10 to 15 of them in any given quarter, arriving as Excel workbooks, PDF exports, Word documents, and proprietary LP portal formats, each asking broadly similar questions in completely different structures.
Completing each one accurately, consistently, and with reference to verifiable documentation is a material drain for all but the largest firms.
The analysis challenge on the LP side is symmetrical. A large institutional allocator reviewing 30 fund managers simultaneously receives 30 DDQ responses, each structured differently, each presenting the same categories of information in different sequences and at different levels of detail. Extracting comparable data across the manager universe, identifying gaps, flagging inconsistencies, and producing a structured assessment of each manager's operational profile is time-intensive work that scales poorly as the programme grows.
This guide covers the DDQ from both sides of that exchange.
In this article:
What a due diligence questionnaire is and where the standard frameworks (ILPA, AIMA, SBAI) come from
What the core sections of a PE DDQ cover, and what each section is actually probing for
Why completing and analyzing DDQs is harder than it looks, on both the GP and LP side
How AI agents are handling the extraction and answering layer (and what that changes for teams running high DDQ volume)
What to look for in an AI DDQ tool built for fund operations
Chat with your files and knowledge hubs
Expert AI agents that understand your work
Get started today
What is a due diligence questionnaire?
A due diligence questionnaire is a standardized set of written questions submitted by one party to another as part of a structured evaluation process.
In private equity and alternative investments, the DDQ flows primarily from limited partner to general partner before a capital commitment is made.
In corporate M&A transactions, a DDQ is submitted by the buyer to the target company, covering the same workstreams as the broader due diligence process.
In vendor and cybersecurity assessments, organizations send DDQs to third-party suppliers and service providers before entering into material commercial relationships.
The DDQ is the beginning of due diligence, not the end. A fund manager's response identifies areas that require deeper investigation: a brief answer about the compliance program suggests further questioning; a team section that lists names without context invites follow-up; a track record section presenting returns without attribution raises questions about methodology. The DDQ creates the structured information base from which the investment team, legal counsel, and operational due diligence specialists identify priority areas for expert inquiry and management presentation.
DDQ vs Request for Proposal
The DDQ is distinct from a request for proposal. An RFP is a competitive solicitation asking multiple vendors or fund managers to describe their capabilities and propose terms. A DDQ is a risk assessment: it evaluates a specific party against a defined set of criteria, with the goal of identifying risk factors, verifying representations, and establishing whether the relationship meets the investor's or buyer's threshold requirements.
In practice, the two documents often overlap (RFPs frequently include DDQ-style sections, and DDQs often include commercial questions that resemble RFP content) but their primary purposes are different.
The DDQ in private equity: who sends it and why
In private equity, the due diligence questionnaire flows primarily from LP to GP. Before an institutional allocator commits capital to a fund, it conducts a structured evaluation of the manager as an organization: not just the investment strategy and track record, but the operational infrastructure, governance, and risk management practices that determine whether the firm can manage capital at the scale and over the duration that the investment requires.
The DDQ is the primary instrument for gathering this information.
The parties sending DDQs are typically pension funds, university and charitable endowments, sovereign wealth funds, insurance companies, banks, family offices, and funds of funds. Each has its own DDQ format, reflecting its own investment policy and the specific operational concerns relevant to its beneficiaries and regulatory context.
A public pension fund may weight governance and conflicts of interest heavily. A sovereign wealth fund may focus on the currency and cross-border dimensions of the investment structure. A fund of funds may place particular emphasis on operational due diligence, given its own obligations to its own investors.
Three industry bodies have produced standardized DDQ frameworks that many allocators use as a baseline.
The Alternative Investment Management Association (AIMA) publishes the industry-standard DDQ for hedge funds and alternative investment managers, organized in modular sections that can be combined depending on the fund structure and strategy type.
The Institutional Limited Partners Association (ILPA) publishes a DDQ specifically for private equity fund managers, developed with input from the LP community to reflect the specific governance and alignment concerns of long-term closed-end fund structures.
The Principles for Responsible Investment (PRI) publishes a responsible investment DDQ focused on ESG integration, used by LPs with sustainability mandates as either a standalone document or a supplement to the AIMA or ILPA questionnaire.
Many institutional LPs use these standard frameworks as a starting point and add their own supplemental sections and questions. The result is a practical DDQ landscape in which every LP sends a document that looks broadly similar in content but differs in format, sequencing, depth, and terminology.
For a GP managing a large LP base, this format heterogeneity is a structural operational challenge rather than an exceptional case.
Core sections of a PE due diligence questionnaire
While each LP's DDQ is structured differently, the substantive content of virtually all institutional PE DDQs covers the same eight areas. Understanding what each section is actually asking for (not just what it says on the cover) is essential for GPs completing responses and for LPs interpreting them.
This section documents the legal and organizational structure of the management company: the registered entities, the beneficial ownership, the governance bodies (advisory boards, investment committees, limited partner advisory committees), and the management of conflicts of interest.
The LP is not just collecting organizational facts here. It is assessing whether the firm is structured in a way that protects LP interests: whether the governance provides genuine oversight of the investment team, whether conflicts between the GP's interests and those of the fund are identified and managed rather than undisclosed, and whether the ownership structure is stable enough to sustain the firm over the life of the fund.
This section asks the fund manager to describe what it invests in, how it sources and evaluates investments, and how the investment process translates from initial screening through to execution and portfolio management.
A fund that claims a focused strategy but describes a broad and vague investment process signals one thing; a fund that articulates a specific, repeatable sourcing and evaluation methodology with documented criteria signals another. LPs with experience in the strategy compare the described process to what the track record actually reflects.
The team section profiles investment professionals, senior operational and compliance staff, and support infrastructure. It covers experience and credentials, but more importantly it covers organizational stability: how long has the team been together, have there been departures, who are the individuals that the firm's investment capability depends on, and what arrangements exist to retain them over the fund's life.
Key-person risk is a dominant concern in smaller and emerging manager contexts, where one or two individuals often represent the majority of the firm's investment capability and LP relationships.
Risk management questions ask the GP to describe how it identifies, monitors, and mitigates risk at the portfolio and firm level. This includes market risk frameworks, portfolio concentration limits, the use of leverage and how its limits are set and monitored, valuation policy and independence, and liquidity risk management for funds with redemption mechanisms.
LPs assess not just whether policies exist but whether they are implemented consistently and whether the described framework matches the risk profile evident in the track record.
Operational due diligence questions cover the infrastructure of the management company: the administrator, custodian, prime broker, and auditor relationships; trade execution and settlement processes; IT systems and data management; business continuity and disaster recovery; cybersecurity policy and controls; regulatory registrations and their scope; and the compliance monitoring program. This section is where LP operational due diligence teams focus their analysis.
For more on the distinct field of operational due diligence, V7 Go's comprehensive guide to AI in due diligence covers how AI is changing the operational diligence process across the full investment cycle.
ESG questions have expanded significantly in scope and depth across most institutional DDQ frameworks over the last five years. The current standard extends well beyond a policy statement: LPs often now ask for evidence of ESG integration at the deal selection stage, portfolio monitoring frameworks, measurement methodologies, reporting to LPs, and alignment with recognized frameworks (TCFD, SFDR, the PRI).
Track record questions request historical returns at the fund, deal, and portfolio company level: gross and net IRR, multiple of invested capital, DPI and RVPI, and the attribution of returns across individual investments.
LPs with analytical resources will reconstruct the track record from deal-level data rather than relying on the summary returns in the marketing materials.
The fees section covers the management fee structure (rate, step-down provisions, basis), the carried interest mechanism (hurdle rate, catch-up, distribution waterfall), co-investment rights and their terms, key-person provisions and fund suspension or dissolution mechanisms, LP Advisory Committee rights, and GP commitment to the fund.
Why due diligence questionaires become a manual bottleneck
The content of institutional DDQs has been substantially standardized through the AIMA and ILPA frameworks. The format has not.
The same LP may send the AIMA DDQ in its original Word format one year and switch to an Excel-native version the following year. A different LP with a similar underlying questionnaire may have built its DDQ into a custom portal that requires answers to be entered field by field through a web interface, with no Excel export available. A third LP may send a scanned copy of a PDF questionnaire generated from a legacy Word template, requiring recipients to type their answers into a new document.
For a GP completing DDQ responses, this format diversity is an operational headache managed through manual labor: reading the format, copying questions into a usable structure, matching them to existing library answers, adapting the language, and returning the response in the format the LP requires. For an LP analyzing DDQ responses received from multiple GPs, the challenge is different but structurally similar: the responses arrive in incompatible formats, requiring the analyst to first extract and normalize the content before any comparative analysis can begin.
Several problems compound this at any scale:
The staleness problem. An answer about the compliance programme structure given six months ago may no longer reflect the current programme if there has been a staff change or regulatory update. A master library that isn't actively maintained produces outdated answers.
The evidence problem. DDQ answers describing policies and processes are only as credible as the supporting documentation. A well-written description of the investment process that cannot be verified against a documented investment policy statement, deal screen, or IC template will be treated with appropriate skepticism.
The consistency problem. Without centralized control, the same GP may give subtly different descriptions of its strategy to different LPs, and a sophisticated allocator reviewing multiple responses from the same firm will notice the divergences.
The volume problem. Average response windows have compressed. LP DDQ response expectations have shortened from 14 days to 7 days for standard requests at many institutional allocators. The combination of more DDQs, more questions per DDQ, and shorter turnaround windows is where the manual process breaks down most visibly.
These problems compound at scale. A fund manager that receives 15 DDQs simultaneously from LPs who all want responses within 30 days faces a coordination and resourcing challenge that is structurally similar to the data room extraction problem in M&A.
The content exists somewhere in the organization. The challenge is extracting it accurately, matching it to the questions asked, and presenting it in a format that meets each LP's requirements without introducing inconsistencies.
AI for due diligence questionaires
This is the specific technical problem that AI-based DDQ tools are designed to address. Rather than requiring the DDQ to arrive in a specific format, or requiring the analyst to manually normalize the input, an AI ingestion layer that can handle format diversity across the full range of common DDQ structures eliminates the manual extraction step and allows the analysis to begin at the question level rather than the document level.
For teams running large LP programmes or managing simultaneous DDQ processes across a portfolio, the V7 Go PE due diligence automation covers how this fits within a broader systematic diligence approach.
How V7 Go processes DDQs at scale
V7 Go's DDQ Diligence Agent was built to address both the format ingestion problem and the answering problem that follows. It operates in two sequential stages that convert an unstructured DDQ in any format into a structured set of answered questions, each with a traceable evidence chain and an explicit flag for whether the question could be answered from the available materials. Note that the selection of LLMs is optimized for the best outputs while increasing performance.
Stage 1: format detection and question extraction. The agent accepts a DDQ file in any format: Excel workbooks (including multi-tab structures with merged cells, nested sub-questions, and conditional logic), PDFs (digital-native and scanned), Word documents, and plain text files. It detects the file type automatically and routes it through the appropriate processing path. Excel files are converted to structured XML that preserves the workbook hierarchy, allowing section boundaries and question relationships to be identified from the original structure. PDF and document files are processed using Gemini 2.5 Pro with OCR support, handling both clean digital PDFs and scanned hard copies with equivalent accuracy. Text-based files are read directly.
Once the file is processed, GPT-5 extracts every question with its section label, producing a standardized JSON structure for each item: Section Name and Question. For a 350-question DDQ, this produces 350 individually addressable records, each with its provenance in the original document preserved, available in V7 Go's project view for team review before answering begins.
Stage 2: answering against the data room or knowledge base. Each extracted question is processed by a grounded AI agent that retrieves evidence from an indexed data room or knowledge base. For each question, the agent identifies the data fields required, forms three to seven targeted search queries using document-specific terminology and time qualifiers, and searches the indexed materials. If initial results are insufficient, it refines its queries and searches again. Source selection follows a defined hierarchy: audited materials over management accounts; signed documents over drafts; the most recent version over earlier ones. Where sources conflict, the agent selects the most authoritative, notes the discrepancy, and reconciles where the evidence allows.
The output for each question is a structured record containing an executive answer in bullet-point format, a reasoning section explaining the logical chain from retrieved evidence to conclusion, a sources table listing each document, section, and verbatim snippet used (each snippet under 25 words), and an "Able to Answer" flag set to Yes or No. GPT-5 Mini serves as the primary answering model; Gemini 2.5 Flash processes each question independently as a secondary pass, allowing the team to compare outputs for questions where additional scrutiny is warranted.
The "No" output is as strategically valuable as the "Yes" output. Questions flagged as unanswerable become an Information Request List: a structured document, organized by section and question, specifying exactly what evidence would be needed to answer each flagged item. For an LP analyzing a GP's DDQ response, the Information Request List turns the gap analysis from an informal note into a formatted follow-up document that can be submitted to the GP directly. For a GP completing its own DDQ against a knowledge base, the list identifies exactly which documentation is missing or insufficient before the response is finalized.
In one deployment, the agent processed 401 questions drawn from multiple DDQ formats simultaneously, covering AIMA-style regulatory and operational sections, fund strategy and governance workstreams, and corporate due diligence sections, all processed against the indexed materials of a single mid-market technology company. Of the 401 questions, 293 (73%) were answered with traceable evidence. The remaining 108 questions were flagged as unanswerable from available materials, with each item specifying the document type and content required. The extraction and answering phase that would typically take a small due diligence team several days was completed systematically and consistently in a fraction of that time.
The project view in V7 Go allows the team to filter results by section, by answer status, and by individual question, with the full evidence chain accessible for any answer that requires verification. Via API or MCP-connected systems, the structured output can populate a deal management system, CRM, or LP relationship management platform, making the findings available across the full investment workflow. The guide to AI in virtual data rooms and best AI data rooms for due diligence cover the data room infrastructure that the DDQ answering layer operates within.
What LPs look for when reviewing a DDQ response
Understanding what an experienced LP investment or operational due diligence team is actually doing with a DDQ response puts the GP's completion challenge in sharper context. The review process is not primarily about reading for content: it is about reading for signals that the content does not explicitly provide.
Internal consistency: Responses should align across strategy, portfolio construction, governance, and risk sections. Mismatches reveal how the firm actually operates versus how it presents itself.
Inconsistencies often indicate poor internal coordination or misrepresentation.Evidence over claims: Statements alone carry little weight without supporting documentation.
LPs expect policies, data, and third-party validation in areas like compliance and valuation.
The depth and quality of evidence is a direct signal of institutional maturity.Consistency over time: LPs compare current DDQs to prior submissions to track changes.
Unexplained shifts in team, strategy, or performance raise concerns quickly.
Clear, proactive explanations build trust over multiple fund cycles.Risk management specificity: Generic statements about “disciplined risk management” are a red flag.
LPs look for clear limits, monitoring processes, and decision authority.
Lack of specificity suggests either immaturity or discomfort with scrutiny.Team transparency: Undisclosed departures or unclear role definitions undermine credibility.
LPs track team changes closely across DDQs and supporting materials.
Transparency around changes—and how gaps are filled—is critical.Track record depth: High-level metrics without loss detail or deal-level data are insufficient.
LPs want to understand how losses occurred and what was learned.
Avoiding this detail signals discomfort with the full performance picture.ESG with substance: Broad ESG commitments without implementation detail are easy to spot.
LPs expect deal-level integration, measurable KPIs, and reporting processes.
Policy without practice signals a superficial program.Operational proof: Policies alone (cybersecurity, compliance, BCP) are not enough.
LPs look for evidence of testing, updates, and real-world application.
Mature firms can demonstrate how policies function in practice.DDQ red flags: what inconsistencies and gaps reveal
Experienced LPs have developed pattern recognition for DDQ-level signals that correlate with more serious issues discovered in deeper diligence. The following recur most frequently across institutional allocator experience.
Vague risk management answers. A risk management section that describes risk management in aspirational terms ("we take a disciplined approach to risk") without specifying the actual mechanisms (what the concentration limits are, how they are monitored, who has authority to breach them and under what conditions) suggests either that the risk management framework is immature or that the team is not comfortable being specific about its constraints. Both are worth investigating.
Undisclosed team changes. A team section that lists individuals who have departed the firm, or that presents the team's credentials in a way that obscures the departure of a key person since the last DDQ cycle, is a credibility issue beyond the operational concern it represents. LPs who maintain detailed DDQ archives will identify undisclosed changes. GPs who manage those changes proactively, with a transparent explanation of how the capability has been preserved or rebuilt, are distinguished from those who hope the change is not noticed.
Track record without loss attribution. A track record section that presents IRR and MOIC without loss detail, or that presents aggregated returns without the deal-level data needed to understand them, tells the LP that the GP is not comfortable with what the full picture looks like. Losses in a PE track record are not inherently disqualifying: the question is whether the losses were consistent with the stated strategy, whether they were managed well, and what the firm learned. Track records that suppress this information are more concerning than track records that present it honestly.
ESG statements without measurement. An ESG section that describes the firm's commitment to responsible investment without specifying how ESG considerations are incorporated at the deal level, what metrics are tracked, and how performance is reported to LPs tells an experienced ESG-focused LP that the program exists on paper only. As institutional LPs have systematized their ESG assessment processes, the gap between substantive ESG integration and policy-without-practice has become easier to identify in the DDQ response itself.
Operational policy without implementation evidence. The cybersecurity section of a DDQ is now a standard component of institutional LP review. A response that says "the firm has a cybersecurity policy and provides staff with annual training" without indicating who the policy was drafted by, when it was last updated, whether penetration testing has been conducted, and by whom, tells a different story than one that provides dated policy documents and test results. The same principle applies to business continuity, disaster recovery, and incident response. Policies that exist but have never been tested or documented in practice provide limited assurance.
DDQ automation: what AI changes and what it does not
AI-based DDQ processing changes the extraction and answering layer of the DDQ workflow. What it does is accept a DDQ in any format, extract every question with its section context, search available knowledge materials for relevant evidence, produce structured answers with traceable sources, and flag where evidence is insufficient. The output is produced at a speed and consistency that manual processes cannot match at scale, and the evidence chain produced for each answer is more auditable than the informal attribution typical of manually assembled DDQ responses.
What AI does not change is the judgment layer. The strategy narrative that a fund manager presents in its DDQ is a function of how the firm chooses to position itself, what it emphasizes, and what it knows about this LP's priorities. The compliance positions taken in the legal section may have nuanced regulatory dimensions that require the firm's legal counsel to review. The ESG commitments made in the responsible investment section carry reputational and legal weight that requires deliberate review before submission. For these elements, the AI agent produces a draft that surfaces what the indexed materials support; the investment and legal team reviews and finalizes. The same applies to LP-side analysis: AI surfaces the patterns in the data, identifies gaps, and flags inconsistencies; the investment committee decision requires human judgment about what those signals mean in the context of the overall investment case.
For a complete overview of how AI is being applied across the full due diligence workflow, including data room analysis, contract review, and fund document processing, V7 Go's AI legal due diligence agent covers the contract and legal document layer, and the broader guide to AI in due diligence maps the full application landscape.
What is a due diligence questionnaire (DDQ)?
A due diligence questionnaire (DDQ) is a structured set of questions submitted by one party to another as part of a formal evaluation process. In private equity, LPs send DDQs to fund managers before allocating capital, covering firm structure, investment strategy, team, risk management, operations, compliance, ESG, track record, and fees. In M&A transactions, the buyer sends a DDQ to the target company. In vendor assessments, organizations send DDQs to third-party suppliers before entering into material commercial relationships.
+
What is the difference between an AIMA DDQ and an ILPA DDQ?
The AIMA (Alternative Investment Management Association) DDQ is the industry-standard questionnaire for hedge funds and alternative investment managers, organized in modular sections covering firm overview, governance, operations, and fund-specific topics. It is widely used for hedge fund and multi-strategy manager diligence. The ILPA (Institutional Limited Partners Association) DDQ is designed specifically for private equity fund managers in closed-end fund structures, with a stronger focus on LP governance rights, alignment of interests, fee disclosure, and the ILPA Principles framework. Many institutional LPs use one or both as a baseline and add their own supplemental sections.
+
How long is a typical private equity DDQ?
A typical institutional PE DDQ contains between 100 and 300 questions organized into 8 to 15 sections. Large LPs with comprehensive operational due diligence programs may send questionnaires that reach 400 or more questions when supplemental sections on cybersecurity, ESG, and regulatory compliance are included. The AIMA standard DDQ in its full form runs over 100 pages. The ILPA DDQ is somewhat shorter but covers PE-specific governance and alignment topics in greater depth. Custom LP questionnaires built on these frameworks can vary significantly in length and structure.
+
What are the main sections of a PE due diligence questionnaire?
The core sections of a private equity DDQ are: firm overview and organizational structure (legal entities, ownership, governance, conflicts); investment strategy and process (what the fund invests in and how); team and key personnel (experience, stability, key-person risk); risk management (portfolio risk framework, leverage, valuation, liquidity); operations, compliance, and legal (service providers, regulatory status, compliance program, cybersecurity); ESG and responsible investment (ESG integration, reporting, framework alignment); track record and performance (fund and deal-level returns, loss history, attribution); and fees, terms, and alignment (management fee, carry, hurdle rate, co-investment rights, GP commitment).
+
How do GPs typically manage DDQ responses?
AI is changing the DDQ process primarily by addressing format fragmentation and extraction bottlenecks. AI agents can ingest DDQ files in any format (Excel, PDF, Word, scanned images), detect the file structure, extract every question with its section label, and match questions to relevant evidence in an indexed data room or knowledge base. For each question, the agent searches available materials, applies a source hierarchy, produces a structured answer with traceable evidence, and flags questions that cannot be answered from available materials as an Information Request List. This automates the extraction and initial classification phase that typically consumes the majority of DDQ completion and analysis time, allowing GP and LP teams to focus on the judgment-intensive review and decision work that requires domain expertise.
+
How is AI changing the DDQ process?
Go is more accurate and robust than calling a model provider directly. By breaking down complex tasks into reasoning steps with Index Knowledge, Go enables LLMs to query your data more accurately than an out of the box API call. Combining this with conditional logic, which can route high sensitivity data to a human review, Go builds robustness into your AI powered workflows.
+
