All Insights

LP Diligence & Workflow

LP Reference Calls: How Institutional Allocators Run Manager References That Actually Surface Risk

Most LP reference calls confirm what the GP wants confirmed. Here is how serious allocators design references that test judgement, partnership behaviour, and operational truth.

Matthew S. · Founder, Allocator Desk · April 2026 · 12 min read

Most LP reference calls do not fail because allocators are naïve or because references are dishonest. They fail because the process is unstructured, inconsistent, and weakly connected to the actual underwriting questions in the investment case. The result is a familiar pattern: polite conversations that validate what the deck already said, then a file note that no one reads again.

Institutional teams also tend to treat references as a late stage validation step, rather than a primary risk discovery tool. By the time calls happen, the narrative is anchored, committee calendars are set, and the human instinct to confirm is strong. A well run reference process can still surface decisive information, but only if you design it to capture signal under realistic constraints.

The best LPs approach references as a system: hypothesis driven, repeatable, and cumulative across years and relationships. That system does not remove judgement, it creates the conditions where judgement can work.

What a reference call is supposed to do

A reference call is not a character check and it is not a substitute for attribution, legal diligence, or operational due diligence. It is a structured attempt to observe behaviour through the lens of someone who has had to live with the manager through good markets, bad markets, and organisational change. Done properly, it tests whether the GP behaves as the LP is underwriting.

The output of the process should be decision usable signals. That means specific examples, context on when and why something happened, and enough detail to triangulate with other facts. If the call cannot change your view of key risks, it is not due diligence, it is theatre.

A reference programme also creates institutional memory. The goal is not only to decide on this fund, it is to understand how this firm operates so that future interactions, re-ups, co-investments, and governance decisions are grounded in observed evidence.

> "A reference call is not a verdict on a manager, it is a test of your underwriting assumptions under real world conditions."

Why most LP reference calls underperform

On-list references are coached, even when the person on the other end is trying to be fair. GPs select references who are satisfied, aligned, and familiar with what can be shared. Many will have had prior calls and will default to safe, high level language unless pressed for specifics.

Time pressure is the silent killer. Reference calls are often squeezed into the last two weeks before an IC, when the team has already absorbed the marketing narrative and is focused on getting a memo out. Under those conditions, references become confirmation rather than investigation.

Junior delegation without context is another common failure mode. A capable associate can run a tight call, but only if they understand the investment thesis, the key risk hypotheses, and the internal history with the firm. Without that, the call becomes a generic questionnaire, heavy on process and light on where failures actually happen.

Many teams lack question discipline. They ask too many questions, too quickly, and accept generalities. The reference senses the pace and responds in kind. A small number of open ended prompts, followed by structured follow ups, usually produces more signal than a long list.

Finally, most organisations do not have structured capture. Notes are inconsistent, stored in personal folders, or reduced to a single sentence in a CRM. Without a disciplined record, lessons do not compound, and the same weak calls get repeated fund after fund. This is exactly the institutional memory failure that platforms like Allocator Desk are built to address through consistent workflows and durable decision records.

On-list, off-list, and back-channel: a working taxonomy

On-list references

On-list calls are the baseline. They can be useful for understanding how the GP wants to be perceived, what the reference considers important, and what topics are clearly rehearsed. They can also clarify mundane but material points: reporting quality, fund admin competence, pacing of capital calls, and the GP’s posture in difficult conversations.

Their limitation is selection bias. Even if the reference is credible, the set is curated. You should treat on-list calls as necessary but never sufficient, and you should avoid letting a clean set of on-list conversations close the diligence loop.

Off-list references

Off-list references are the workhorse for surfacing risk. These are people who have direct experience with the firm but were not selected by the GP, such as former LPs, co-invest partners, portfolio company executives, former employees, or service providers. The value is not that off-list people are more negative, it is that they are less constrained by the GP’s framing.

Off-list calls require more care on confidentiality, motivation, and context. Some off-list sources have an agenda, including unresolved commercial disputes or internal politics. The skill is to separate grievance from pattern, and to triangulate.

Back-channel references

Back-channel references are informal, fast, and often the earliest signal you will get. They include peers in the allocator community, former colleagues, and sector specialists who have intersected with the firm. These conversations can be highly informative on reputation, culture, and market behaviour, especially around fundraising conduct and how the GP handles difficult stakeholders.

The limit is verifiability. Back-channel information can be stale, second hand, or coloured by competitive dynamics. Use it to generate hypotheses and direct further work, not as standalone evidence.

The five categories of risk a reference call should test

1. Investment judgement under stress The reference should help you understand what the team does when a deal breaks, a thesis changes, or exits close. Ask for examples where the GP changed its mind, wrote down an asset, or chose liquidity over optics. The key is whether they learn and adapt, or defend and rationalise.

2. Partnership behaviour and culture Culture shows up in decision rights, internal dissent, and how credit is allocated. References can reveal whether the firm is founder dominated, whether juniors can challenge, and how investment committees actually function. Look for patterns in turnover, promotions, and whether success depends on proximity to a rainmaker.

3. Operational and ODD signals Operational issues often appear first as small frictions: inconsistent reporting, weak controls around valuation, or slow responses on compliance. References can indicate whether the GP invests in infrastructure ahead of growth, or patches problems after investors complain. Service provider interactions, including fund administrators and auditors, are particularly informative when a reference has seen transitions.

4. LP service and communication quality You are underwriting a long relationship, not a trade. References should describe how the GP communicates when performance disappoints, when a deal goes wrong, or when a key person leaves. The most telling moment is not the quarterly call, it is the unsolicited call when the GP has bad news.

5. Succession and key-person continuity Many organisations market depth, but behave like a small partnership. References can clarify who really owns deals, who speaks for the firm with lenders and management teams, and what happens when a senior person is distracted by fundraising or personal matters. You want evidence of continuity across fund vintages, not assurances.

Designing the question set

Start with your underwriting hypotheses, then design prompts that force specific examples. If you are underwriting a strategy shift, you need references who experienced the transition. If you are underwriting a complex co-invest programme, you need people who have executed with them under time pressure, not just sat on quarterly calls.

Keep the call structured but human. Begin by setting confidentiality expectations and clarifying the reference’s relationship to the GP, including timeframe, fund exposure, and role. Then move quickly to examples: one good investment, one difficult situation, one governance moment, one resourcing change.

Sequence matters. Ask for narrative first, then drill into gaps. People disclose more when they feel listened to, but they disclose specifics when they are pushed politely toward facts, dates, and decisions.

Questions that work

• Tell me about a time the GP was wrong on a deal. What happened next, and what did they change as a result?
• When performance disappointed, how did they communicate, and what did they say before they had a solution?
• Who are the real decision makers in the room, and how does dissent show up?
• If you could change one thing about how they operate, what would it be, and why?
• Describe the hardest negotiation you had with them, fees, terms, reporting, valuation, or governance. How did they behave?
• How stable has the team been during your relationship, and what were the reasons for any departures?
• What is the best example of them doing the right thing at their own expense?
• If you were re-underwriting them today, what would you pay closest attention to?

Questions to avoid

• Would you invest with them again?
• Are they good people?
• Do they have strong deal flow?
• Any concerns?

These prompts invite polite answers and give you little you can test. Replace them with questions that require a story, a trade-off, or a decision point.

<aside><strong>TAKE NOTES LIKE YOU WILL BE DEPOSED</strong><p>Write down the example, the setting, who was involved, and what changed afterwards. Separate what the reference directly observed from what they inferred or heard elsewhere. If you cannot attribute the insight to a specific experience, treat it as a hypothesis, not a finding.</p></aside>

How to find off-list references

Off-list sourcing is a craft, but it is not mysterious. Start with public filings and work outward. For managers registered as investment advisers, the SEC’s Form ADV search can provide legal entities, ownership, and sometimes clues on affiliates and key personnel that help map the organisation. It is not a directory of LPs, but it is a reliable anchor for names and structures.

Use fund and deal databases in a Pitchbook style workflow to identify portfolio companies, co-invest partners, and financing counterparties, then map individuals who were involved at the time. LinkedIn alumni searches are practical for finding former employees of the GP and former executives at portfolio companies, especially those who served during restructurings or leadership transitions. Treat former employees as high value but potentially high bias, and always triangulate.

Look for prior-fund LP overlap within your network. Advisory committee membership is often not public, but conferences, annual meeting attendee lists, and manager hosted events can indicate who has been in the ecosystem. You can also use known service providers, administrators, auditors, and placement agents as routing points for introductions, while respecting confidentiality and conflicts.

Warm-introduction etiquette matters. Be clear that you are not asking for confidential fund information and that you will keep the conversation discreet. Offer context: strategy, vintage, and what you are trying to understand. Do not forward a long question list by email, and do not ask someone to breach their own policies. A good reference process protects the reference as much as it protects you.

Capturing the call so it compounds

The difference between competent and excellent LP reference work is capture quality. Most teams treat notes as temporary, but reference insights are often most valuable years later when something changes: a partner leaves, a valuation is challenged, or a co-invest opportunity arrives with no time for a fresh diligence cycle. This is the institutional knowledge problem described in Why Endowments Lose Institutional Knowledge, and it is solvable with a disciplined operating system.

Capture should be structured. Record the reference’s relationship to the GP, time period, fund exposure, and what they directly observed. Tag insights to the risk categories you underwrite and to specific people, strategies, and vehicles. Store negative and positive evidence with equal clarity, then link it to subsequent events so you can evaluate whether early signals were predictive.

This is where workflow and relationship management tools earn their keep. In Allocator Desk, the goal is not simply to store a note, it is to connect reference insights to the manager record, the relationship history, and the decision trail so that future team members can see how prior conclusions were formed. The same logic applies when you scale GP interactions across portfolios, as discussed in How LPs Can Manage GP Relationships at Scale.

Reference insights should flow into the IC memo as explicit risk statements, not as appendices that no one reads. Use a consistent format that distinguishes facts, reference observations, and your interpretation. Then link the conclusions into the decision record, including what you would need to see to change your mind. For a practical framework, align capture with the structure in How to Write an Investment Committee Memo for LPs.

Finally, use technology carefully. AI can help summarise, deduplicate, and retrieve prior notes, but it cannot judge credibility or motivation without your context. A sensible approach is outlined in How LPs Should Use AI in Investment Due Diligence: treat AI as an assistant for recall and patterning, then apply human scepticism to the substance.

Reading what is not said

The most valuable part of a reference call is often the boundary of what the person will say. Hedging language, long pauses, and careful legalistic phrasing can signal unresolved issues, even when the reference remains polite. Pay attention to where the reference becomes non-specific, and ask what makes it hard to answer.

Silence around certain partners is a tell. If the reference enthusiastically praises the firm but cannot describe how one senior person operates, you may have a key-person dependency or internal politics. Similarly, if they praise communication but cannot recall a time the GP delivered bad news proactively, you may be underwriting a relationship that only works when everything is going well.

Refusal to be specific is not always negative. Some institutions have strict policies that limit what staff can share, and some professionals are simply cautious. The question is whether the person can at least describe categories of behaviour, decision processes, and the shape of events without disclosing confidential details. If they cannot, you may need a different reference, not a different conclusion.

Listen for asymmetry. If the reference describes the GP’s strengths in detail but describes weaknesses in vague terms, or vice versa, ask yourself what incentives are operating. Then triangulate with a second source that had a different role, such as an operating executive versus an LP, or a former employee versus a co-invest partner.

Governance and conflicts: keeping the process clean

Reference work sits in a sensitive zone. Confidentiality expectations are real, and professionals remember who handled a call responsibly. Start each call by stating that you are not seeking material non-public information, and steer away from deal-specific forward looking questions or anything that sounds like it could elicit MNPI. If the reference starts sharing information that feels inappropriate, stop and redirect.

Conflicts also matter. A service provider may be auditioning for work. A former employee may be aggrieved. A competing GP may have commercial motives. None of that makes the information useless, but it requires explicit tagging of potential bias and additional triangulation.

ILPA’s guidance frames the professional conduct expected between LPs and GPs, including transparency, governance, and alignment. For diligence standards and a shared language, anchor your approach in ILPA Principles 3.0 and the ILPA Diligence Questionnaire. Use these documents to align your reference prompts with what you expect to be true, and to identify where behaviour diverges from policy.

Governance extends beyond the call. If reference work indicates side letter complexity, inconsistent MFN handling, or uneven disclosure practices, that is an operational risk, not just a legal detail. Make sure reference insights connect to your contracting workflow and side letter controls, particularly if you are scaling across multiple vehicles. It is a common blind spot, and worth treating as part of the diligence system, as outlined in Side Letter Management for LPs: Hidden Operational Risk.

A closing standard for serious LP reference work (final closing section)

A serious LP reference programme has a clear standard: every call must test a defined risk hypothesis, must produce at least one specific example, and must be captured in a way that can be retrieved and audited later. If you cannot meet that standard, you are better off doing fewer calls, with better sourcing and tighter execution.

The second standard is triangulation. No single reference, on-list or off-list, should decide an outcome. Patterns should. When multiple independent sources describe the same behaviour under different conditions, you have something you can underwrite.

The third standard is accountability to the decision record. Reference findings should appear in the IC memo as explicit risks, mitigants, and open questions, not as folklore. Over time, your organisation should be able to look back and evaluate whether reference signals predicted outcomes, then refine the question set accordingly.

Reference calls will never be perfect, but they can be decisive. Treat them as a signal-capture process, run with the same discipline you apply to portfolio construction, and you will surface more risk before you pay for it.

Topics

• Reference Calls
• Manager Diligence
• LP Operations
• Off-List References
• GP Diligence
• Investment Committee
• Private Equity
• Institutional Memory