All Insights

LP Diligence & Workflow

Building an LP Co-Investment Program: A Repeatable Playbook Without Straining GP Relationships

Most LP co-invest programs fail quietly. Not because deals go bad, but because the process cannot keep up with the sponsor. Here is how sophisticated allocators build a repeatable pipeline that GPs actually want to share.

Matthew S. · Founder, Allocator Desk · July 2026 · 9 min read

Co-investment sits in an awkward spot on the allocator desk. It looks like private equity, moves at the speed of direct investing, and is judged against a portfolio that already exists. Programs that work treat co-invest as its own discipline with its own decision cadence, not a bolt-on to fund selection.

This piece walks through what a repeatable LP co-invest program looks like in practice: how deals arrive, how a small team makes fast decisions, and how you stay a preferred partner without becoming a rubber stamp.

<figure style="margin:40px 0"><img src="/blog/co-invest-program.jpg" alt="Abstract visualization of a co-investment deal pipeline funneling into a decision point" loading="lazy" width="1600" height="912" style="width:100%;height:auto;border-radius:10px;display:block" /><figcaption style="font-family:'SF Mono',Monaco,monospace;font-size:11px;letter-spacing:0.1em;text-transform:uppercase;color:#6A7080;margin-top:10px;text-align:center">A working co-invest program is a pipeline, not a to-do list.</figcaption></figure>

Why most LP co-invest programs stall

The common failure modes are boring and predictable.

• **Slow response times.** A GP sends a deal on a Wednesday with a two-week window. The IC does not meet for three. The LP passes by default.
• **No standing thesis.** Every deal is underwritten from a blank page, so decisions are inconsistent and slow.
• **Fund team owns the process.** The fund analyst who cleared the manager also has to underwrite the deal. They are already booked with re-ups and quarterly reviews.
• **Sponsor fatigue.** After two or three passes with no clear reason, the GP stops calling.

The pattern is not a lack of capital. It is a lack of process muscle.

The three-lane model

Allocators with functioning programs almost always run some version of the same structure. Three lanes, each with different bar heights and decision speeds.

Lane 1: Fast-follow. Deals from managers you already back, in strategies you already underwrite. Target 48 to 72 hour turnaround. Decision maker is a single senior investor with post-facto IC ratification.

Lane 2: Extended platform. Deals from managers you know but in adjacent sectors, or from a GP that is new to you but comes highly referenced. Full memo, standing IC slot within 10 business days.

Lane 3: Opportunistic. Direct or club deals outside the fund program. These require the full underwriting stack. Assume most will not close and staff accordingly.

The magic is not in the lanes themselves. It is that every deal is triaged into a lane within 24 hours of arrival. Nothing sits waiting to be classified.

<figure style="margin:40px 0"><img src="/blog/co-invest-workflow.jpg" alt="Editorial illustration of memo pages routing into a decision hub with three outcomes" loading="lazy" width="1600" height="912" style="width:100%;height:auto;border-radius:10px;display:block" /><figcaption style="font-family:'SF Mono',Monaco,monospace;font-size:11px;letter-spacing:0.1em;text-transform:uppercase;color:#6A7080;margin-top:10px;text-align:center">Every deal enters a lane within 24 hours. Nothing sits in triage.</figcaption></figure>

What sits inside a standing thesis

Programmatic co-invest requires you to know, before the deal arrives, what a good deal looks like. That means writing down:

1. Sectors you will underwrite. Not just industries, but sub-verticals and business models. If you do not do consumer, say so. 2. Check size bands. By strategy, by risk profile, and by concentration limit against total portfolio NAV. 3. Structural bright lines. Minority only, board observer minimum, tag-along rights, information rights, and what fee terms you will accept. 4. Sponsor concentration caps. How much of your co-invest book any single GP can represent.

A good standing thesis is short. Two pages, updated annually, signed off by the IC. It is the reason you can move in 48 hours without cutting corners.

<aside><strong>Practitioner note</strong><p>The biggest unlock is not analytical horsepower. It is agreeing in advance on what you will not do. Every no you can pre-clear turns into speed on the deals you actually want.</p></aside>

Sourcing without begging

GPs share the best co-invest with LPs who behave like partners on the previous ones. That means five behaviours worth building into your program:

• **Reply within one business day.** Even a soft pass is better than silence.
• **Give clean reasons for a pass.** Not exhaustive, just directional. It helps the sponsor calibrate.
• **Do not renegotiate at the last minute.** If you flagged concerns at the IOI stage, do not resurface them the night before signing.
• **Show up at the reference stage.** Being available for calls with management makes you look like a co-underwriter, not a passive check.
• **Track outcomes and share them.** A short annual note back to the GP on how their deals performed for you builds trust faster than any pitch.

Being a preferred LP is a compounding advantage. The best allocators treat sponsor relationships as a book of business with its own KPIs.

Underwriting compression: what to cut, what to keep

A 48 hour Lane 1 decision does not mean a shallow memo. It means a compressed one. What matters in that first pass:

• **Sponsor conviction.** You already know the GP. Confirm the deal team is who you expect and who owns the value creation plan.
• **Alignment.** Fee load, GP commitment on top of the fund position, and whether the co-invest sits pari passu with the flagship.
• **One-page valuation view.** Entry multiple against a public comp set and against the sponsor track record for the strategy.
• **Two downside scenarios.** A specific-to-the-company case and a macro case. Not exhaustive, just enough to see where you break.
• **Portfolio fit.** Concentration at the sponsor, sector, and geography level.

What you leave to Lane 2 or later: exhaustive market sizing, deep expert calls, and full financial model rebuilds. Trust the sponsor to have done that work. Your job is to pressure test it, not repeat it.

The infrastructure question

Co-invest programs live or die on operational discipline. A short checklist of the systems that separate the working programs from the aspirational ones:

1. Deal log with source, date, lane, decision, and time-to-decision. Reviewed monthly. 2. Sponsor scorecard. How many deals shown, how many closed, how the closed deals are performing, what the pass reasons were. 3. Standing IC memo template. Same shape every time. The reader knows where to look. 4. Post-close monitoring cadence. Co-invest positions are usually reported inside the sponsor fund cadence. That is not enough. Set your own quarterly touch. 5. Institutional memory. Every pass should be searchable a year later. The GP will show you a similar deal in Fund V and you need to know why you passed on the Fund IV version.

Without this infrastructure the program is a series of one-off decisions. With it, you have a compounding data asset.

:::related /insights/the-reup-decision-how-lps-evaluate-the-next-fund The Re-Up Decision: How LPs Should Evaluate a Manager's Next Fund Co-invest performance feeds directly into re-up conviction. A working co-invest book is one of the best pre-re-up diligence tools you have. :::

Common structural traps

A handful of terms that regularly cause pain and are worth negotiating hard on:

• **Broken-deal expense pass-throughs.** Understand your exposure if the deal falls apart between IOI and close.
• **Follow-on obligations.** If the company needs more capital, are you compelled or invited? Get it in writing.
• **Exit dragging rights.** How much control does the sponsor have over your exit, and at what price mechanics?
• **Reporting frequency.** Co-invest positions often report less than the flagship fund. Set expectations up front.
• **MFN scope.** If you are getting different terms than other LPs on the deal, confirm the MFN language covers you.

Most of these are boilerplate for larger allocators. If you are building a program for the first time, they are the terms your counsel will thank you for flagging early.

What good looks like at 24 months

A program running for two years should be able to show:

• 30 to 60 deals reviewed, with lane assignment and time-to-decision on each.
• A close rate of 15 to 25 percent in Lane 1, lower in Lanes 2 and 3.
• Sponsor concentration inside the caps set in the standing thesis.
• A portfolio that looks like the thesis, not a scrapbook of whatever crossed the desk.
• At least one honest post-mortem on a deal that did not work.

That last one is underrated. The programs that stay good are the ones that can look at a loss without flinching.

Bottom line. A co-invest program is not a strategy. It is a system. The allocators who compound in this asset class are the ones who have made peace with the fact that speed, discipline, and sponsor trust are the product. The deals are just the output.

Topics

• Co-Investment
• LP Operations
• Deal Flow
• GP Relationships
• Portfolio Construction
• Private Equity
• Investment Committee