Proprietary deal flow and the evidence on auctions

Proprietary deal flow can describe direct owner relationships, early access to a sale process, or an invitation to a limited auction. Those situations offer different advantages. Research on takeover processes helps distinguish the value of early access from the assumption that avoiding an auction necessarily produces a lower price.

Defining proprietary deal flow

The term is used for at least four situations. Define which one you mean when assessing an origination strategy or reporting its results.

• Genuinely unintermediated. No banker, no process, no other bidder. The owner was not planning to sell this quarter and is talking to you because you turned up credibly.
• Pre-process. The owner has decided to sell, has spoken to advisers, and you are having the conversation before the book is written. There will very likely be a process; you are trying to pre-empt it.
• Limited process. Three to six buyers, chosen by an adviser, no public marketing. Competitive, but quiet. This is often described as proprietary by the people in it.
• Relationship-sourced. A broad auction that you heard about early through a relationship. Early knowledge of a competitive process is not the same as the absence of one.

The economics differ sharply across those four, and so does the work required to get there. Being precise about which one you are actually building for is the first discipline, because the third and fourth are largely a function of who returns your calls, while the first two are a function of whether you know something about the company before anyone tells you.

How sale processes work

The best empirical answer to “what does the sale process look like?” comes from Audra Boone and Harold Mulherin, who hand-collected the pre-public stages of takeover processes from merger background sections in SEC filings [1] [11]. Their point is methodological before it is economic: if you count bidders by looking at who publicly announced an offer, the takeover market looks uncompetitive. If you read the Background of the Merger section, where a target has to disclose how many parties were contacted and how many signed confidentiality agreements, it does not.

In their sample, roughly half of targets were auctioned among multiple bidders and the remainder negotiated with a single bidder [1]. The sample shows that both processes were common in the transactions studied. It also demonstrates why public bids alone do not capture all competition. These findings should not be treated as a direct estimate of the process mix for smaller private-company sales.

The same paper reports that wealth effects for target shareholders were comparable in auctions and in negotiations [1]. Read carefully, that is not a claim that competition does not matter. It is evidence that sellers and their advisers choose the process that suits the asset, and that the choice is endogenous: a target likely to attract many bidders gets auctioned, and a target with one natural buyer gets negotiated with that buyer, often with the credible threat of a process standing in for the process itself. The seller’s adviser is not passive, and neither is the board, whose conduct in a sale is the most litigated question in Delaware corporate law [7] [6].

Why the merger background section is underused

For anyone doing origination research, the practical takeaway from that body of work is not the finding. It is the data source. Merger background sections are a public, searchable, first-person account of how a specific company came to be sold, how many parties were approached, who declined, and what the timetable was. They are full-text searchable back to 2001 [11]. If you want to know which buyers get called in a sector, and how quickly a process moves once it starts, that is written down.

It is only available where a public registrant was involved, which biases it towards larger deals. Use it to develop questions about smaller transactions, while recognising that their process and adviser coverage may differ.

Evidence on the winner’s curse

The standard argument for avoiding auctions is the winner’s curse: in a common-value auction, the bidder who wins is disproportionately likely to be the one who overestimated the asset. It is a real phenomenon, it is well documented in settings like offshore oil leases, and it is the reason many buyers say they will not play in broad processes.

Boone and Mulherin tested it directly in corporate takeovers and did not find the pattern the theory predicts [2]. Acquirer returns in auctions were not systematically worse than in negotiations. This does not establish that auctions are preferable in every case. Bidder behaviour and the selection of assets into different processes matter when interpreting the comparison.

Who values what

Gorbenko and Malenko estimated bidder valuations from actual bids in takeover auctions and separated strategic from financial buyers [3]. Their headline is that a typical target is valued more highly by a strategic acquirer. There was also a substantial exception: in their sample 22.4% of targets were valued more highly by financial bidders, and those targets were, in the authors’ description, mature and poorly performing companies.

They also found that valuations across different strategic bidders were more dispersed, while financial bidders’ valuations moved with aggregate economic conditions [3]. These findings suggest questions for an origination strategy:

1. Where might a financial buyer have an advantage? The characteristics associated with higher financial-buyer valuations in the study can inform research questions. They do not determine which buyer will value any particular company most highly.
2. Strategic dispersion means the identity of the other bidder matters more than its category. If strategic valuations are widely dispersed, the question is never “will a strategic outbid us” but “is this specificstrategic the one for whom this asset closes a gap.” That is answerable with public evidence: what they have bought before, what their site says they cannot do, where their footprint has a hole.
3. Sponsor competition is correlated, so timing is a real variable.If financial bidders’ valuations track macro conditions, then the number of sponsors chasing the same asset moves together. Consider how changes in financing conditions may affect competition for the same assets.

Fidrmuc and co-authors looked at the same question from the seller’s side, examining how the chosen sale process differs when the likely buyer is a sponsor rather than a strategic [4]. The general lesson is that sellers are not indifferent between buyer types and structure processes accordingly, so buyer type is relevant when assessing a sale process.

The potential value of early access

Four potential benefits can be assessed separately from any expected price discount:

These benefits depend on both research and relationships. Research helps identify relevant companies and prepare for a discussion; continued contact establishes whether and when the owner wants to engage.

Three routes to being early

Buy the list

Commercial databases will sell you a filtered universe in an afternoon. It is a useful starting point that benefits from further research, for a reason that has nothing to do with vendor quality: everybody else buys the same list with the same filters. A screen that any subscriber can reproduce produces a target set that any subscriber has. The database’s coverage of small private firms is also systematically incomplete in ways that are well documented in the research literature on firm-level databases, and the gaps are not random.

Develop intermediary relationships

Intermediary relationships, conference circuits and banker coverage are the traditional answer, and they work. They also scale badly, cost a partner’s calendar, and deliver mostly the third and fourth definitions of proprietary from the top of this page. A fund that is well covered by intermediaries gets called early about processes. It does not get called about companies that are not for sale.

Build the knowledge

The third route is to maintain a research base on the sector: relevant operators, their ownership, activities, scale, and fit with the mandate. Updating that record preserves both current findings and the history of changes, supporting direct owner outreach over time.

Being early is a research problem

Before approaching a company, research can help establish:

• That the company exists and is what you think it is. Harder than it sounds. Trading names, holding companies, same-named businesses in other states and acquired entities that still run a website all conspire against you.
• Who owns it. Not who runs it. Founder-owned, family-owned, management-owned, sponsor-backed, subsidiary of a listed parent: these have entirely different conversations attached, and their relevance depends on the mandate.
• Whether it clears your hard constraints before anyone spends time on the soft ones. Size, geography, regulatory status, ownership structure.
• What is distinctive about it in terms your investment committee already uses, with the evidence attached, so the first email is about their business rather than about your fund.
• Whether anything has changed.A principal’s tenure, a recent hire in a finance seat, an adviser engagement, a licence transfer. Timing in origination is mostly a monitoring problem.

Every one of those is a question with a findable, citable answer for a large fraction of companies. None of them requires a relationship. All of them, historically, required an analyst, which is why the work was done for the forty companies somebody already liked rather than the four thousand nobody had looked at. The population is the point: the number of US firms in the size bands most lower-middle-market funds target runs to the hundreds of thousands [12], and a hundred-name list is a rounding error against it.

The arithmetic of an origination programme

A funnel makes the work and conversion assumptions explicit. The stages below are illustrative; use your own volumes, outcomes, and capacity constraints to assess the programme.

Two considerations help interpret the funnel.

1. Identify the limiting stage. A larger researched list may not generate more conversations if contact coverage, approval capacity, or follow-up capacity remains unchanged.
2. Evaluate coverage and conversion together. A larger universe increases potential volume only if downstream conversion and delivery capacity hold. Compare the cost of expanding coverage with the cost of improving a particular stage.

There is a third observation that is less comfortable. If the number of deals reaching diligence per year is small — and for a lower-middle-market fund it is — then the sample is far too small to learn from quickly. You cannot A/B test your way to a better origination programme on closings. You can only learn from the intermediate outcomes you get in volume: which segments reply, which messages get read, which research fields actually predicted an owner engaging. That requires recording those intermediate outcomes consistently, with the relevant segment and research context.

What the seller’s adviser is optimising

Origination strategy is a game against an intelligent counterparty, and it helps to be explicit about what that counterparty wants. A sell-side adviser is not maximising price in the abstract. They are maximising some combination of price, certainty of closing, speed, and the probability of being hired again — and the weights differ by adviser and by mandate.

That has direct consequences for a buyer trying to pre-empt a process:

• Certainty is worth real money. A buyer who has already done the work, who can describe the business accurately in the first meeting, and whose approvals are not contingent on a committee that meets quarterly, is offering something the adviser can price. This is the strongest practical argument for researching before contact rather than after.
• A pre-emptive bid needs a clear rationale. An adviser who takes a pre-emptive offer to a client is spending credibility. If the number is merely acceptable, running the process is the safer recommendation for them personally, whatever it does for the client.
• Being known reduces the cost of including you. The cheapest thing a buyer can buy from an adviser is a place on the call list, and the currency is being legible: a clear mandate, a track record of closing what you sign, and not wasting their time on assets outside your stated criteria. A clear mandate helps advisers judge whether an opportunity is relevant.
• The board’s duties are a real constraint, not a formality. Where a target has outside shareholders, the conduct of a sale process is the most litigated area in corporate law [7] [6], and an adviser will not accept a pre-emptive deal that leaves the board exposed. Understanding that is the difference between a proposal that can be accepted and one that cannot, regardless of price.

What early looks like operationally

A workable origination function has five properties. They support consistent research and review across the target universe.

1. The universe is defined before the screening starts. Otherwise the screen is measuring where you happened to look. A market map built from public statistics and licence registers is checkable; a list assembled from memory is not.
2. Disqualifying questions are asked first. If a listed parent is fatal, find out in the first minute of research rather than the third week. The cost asymmetry is enormous and it is entirely within your control.
3. Apply the same questions across the list. When criteria change, review existing records as well as new companies. Show which records still need the additional research.
4. Nothing is scored without evidence. A blank should sort a company down, not silently score as a pass. A high score built on four answers out of twenty is a statement about your coverage, not about the company.
5. Contact is governed. The list of who gets approached, by whom, saying what, is a decision the buyer makes and can audit afterwards. This is a reputational asset and it is easy to spend by accident.

Notice that items two through four are really one idea: the ranking has to be produced by a consistent process applied to every candidate, or the ordering carries information about the process rather than about the companies. That is an old result in the judgement literature, not a new one in deal-making [16].

Add-ons make the arithmetic worse, not better

Add-on acquisitions have accounted for a large and growing share of US private equity deal count for years [9] [8]. A platform doing four tuck-ins a year needs a live map of its sector, not a project. And because most add-ons fall well below the Hart-Scott-Rodino notification thresholds [13], they close without any public filing at all, which means the competitive map you are working from is silently going stale in a way no data vendor will tell you about. The agencies have themselves become more attentive to serial acquisition strategies [14], which is one more reason to know precisely what your platform and its competitors have already bought.

What this does not claim

Three limits matter when interpreting these findings.

• Research does not create willingness to sell. It finds the companies where the conversation is possible and makes the first contact credible. Whether an owner wants to talk is theirs to decide, and most will say no. Survey work on owner readiness consistently finds that a large share of private business owners have no written transition plan at all [15], so a demographic opportunity does not imply a near-term transaction.
• Negotiated does not mean cheap. A seller with one buyer still has an adviser, a number in mind and the option of waiting. The evidence that target shareholders do comparably well in negotiations [1] should be read as a warning to any buyer whose model depends on a discount for exclusivity. Deal-protection terms exist precisely because the threat of another bidder never fully disappears [5].
• Coverage is not insight. Knowing about every company in a sector is necessary and not sufficient. It removes the excuse of not having looked; it does not tell you which one to buy. What it does do is let the judgement happen on a complete board rather than on the subset somebody had time for, which is a different and better problem to have. Your limited partners, incidentally, are entitled to ask how the pipeline was built [10].

A maintained research record supports relationships over the life of a fund. It gives the team shared context on companies, previous contact, and changes worth reviewing, while leaving transaction judgement and the owner relationship with the people responsible for them.

Sources

References for the research and standards discussed in this guide. Some publications require a subscription or institutional access.

1. [1]
   How Are Firms Sold?

   The Journal of Finance · 2007

   The canonical study of how takeover targets actually reach a buyer: full auctions, limited auctions and negotiations, hand-collected from SEC filings.

2. [2]
   Do auctions induce a winner's curse? New evidence from the corporate takeover market

   Journal of Financial Economics · 2008

   Tests whether buyers overpay in competitive takeover processes relative to negotiated ones.

3. [3]
   Strategic and Financial Bidders in Takeover Auctions

   The Journal of Finance · 2014

   Separates how strategic and financial buyers value the same target, using bid-level data from takeover contests.

4. [4]
   One size does not fit all: Selling firms to private equity versus strategic acquirers

   Journal of Corporate Finance · 2012

   Why the sale process a seller chooses differs when the likely buyer is a sponsor rather than a strategic.

5. [5]
   Termination fees in mergers and acquisitions

   Journal of Financial Economics · 2003

   On the deal-protection devices that shape what a competing bidder can realistically do after a deal is signed.

6. [6]
   Harvard Law School Forum on Corporate Governance

   Harvard Law School

   Practitioner and academic commentary on deal process, fiduciary duties and disclosure; heavily cited by the M&A bar.

7. [7]
   Delaware Court of Chancery opinions

   Delaware State Courts

   Where the duties a board owes when running a sale process are actually litigated and defined.

8. [8]
   Global Private Equity Report

   Bain & Company

   Annual industry review: dry powder, holding periods, exit conditions and deal multiples.

9. [9]
   PitchBook research and reports

   PitchBook Data

   Quarterly US PE breakdowns, including add-on share of deal count.

10. [10]
    Institutional Limited Partners Association

    ILPA

    LP-side standards for reporting, fees and governance; the counterparty view of how a fund is judged.

11. [11]
    EDGAR company and filing search

    U.S. Securities and Exchange Commission

    The public search interface. Free, rate-limited, and authoritative for anything a registrant had to disclose.

12. [12]
    Statistics of U.S. Businesses (SUSB)

    U.S. Census Bureau

    Firm counts and employment by enterprise size, which is how you size the population of buyable companies.

13. [13]
    Premerger Notification Program (Hart-Scott-Rodino)

    U.S. Federal Trade Commission

    Filing thresholds are adjusted annually; below them a deal closes without notifying anyone.

14. [14]
    2023 Merger Guidelines

    U.S. Department of Justice and Federal Trade Commission · 2023

    How the agencies say they analyse a transaction, including concentration thresholds and serial acquisition.

15. [15]
    State of Owner Readiness research

    Exit Planning Institute

    Survey series on how prepared private business owners are to transition, and how few have a written plan.

16. [16]
    Judgment under Uncertainty: Heuristics and Biases

    Science · 1974

    Representativeness, availability and anchoring, including the base-rate neglect that makes a good story beat a good prior.

Keep reading