seastone

36 Research-Supported Signals. One Pre-Exit Intelligence System.

Richard Besier11 June 2026

Seastone combines 36 empirically grounded indicators to identify when a company may be moving toward an ownership transition—before that opportunity becomes obvious to the wider market.

Most acquisition sourcing begins after a company has entered a formal sale process. By then, an adviser may have been appointed, marketing materials prepared and likely buyers contacted. The opportunity is visible, but it is no longer proprietary.

Seastone is built around a different premise: business exits are rarely isolated events. They are usually preceded by a combination of observable changes in ownership, leadership, operations, financing, regulation, stakeholders and market conditions.

Seastone has developed a proprietary signal architecture around this premise. It brings together 36 research-supported signal families, each with a defined role in evaluating exit propensity, timing or transaction state.

The individual signals and their source mappings remain proprietary. The research foundation behind the system, however, produces a clear conclusion: observable business changes can identify populations with materially different transaction and exit probabilities.

How the 36-signal architecture works

The signals do not operate as 36 separate yes-or-no predictions. They work together as a system.

Some signals indicate that the probability of an exit is increasing. Others identify where a company sits within a longer ownership or investment cycle. Some provide negative evidence that reduces the likelihood of a sale. Others distinguish a potential ownership transition from closure, distress, refinancing or ordinary corporate activity.

Together, the 36 signals perform three essential functions.

1. Detecting forward movement

The largest part of the architecture identifies changes that can precede an ownership transition. These signals cover company-level events, stakeholder changes, operating developments and external shocks. They help answer:

  • Has something changed?
  • When did it change?
  • Is the company entering a historically relevant window?
  • Are several independent developments pointing in the same direction?

The evidence includes longitudinal company studies, administrative ownership records, event-history models, matched company panels and natural experiments.

2. Adjusting the underlying probability

A company’s exit probability does not begin at zero. Ownership structure, company maturity, market conditions and other persistent characteristics influence the starting probability before a new event occurs.

Seastone therefore uses propensity and suppression signals to adjust that baseline. A relevant development may increase the score, while negative evidence can reduce it or indicate that the company is more likely to remain independent.

This is important because a predictive system that only searches for positive evidence will produce too many false positives.

3. Confirming and classifying the transition

Some observable developments become most informative when a transition is already underway. These signals help distinguish:

  • an emerging sale from an ordinary company event;
  • an ownership transfer from a closure;
  • a strategic exit from financial distress;
  • a seller from a likely acquirer; and
  • a genuine transaction from administrative noise.

Confirmation evidence strengthens the engine’s ability to classify opportunities and validate earlier observations.

What the published evidence shows

Across the complete research base, the studies report materially different outcomes when relevant conditions are present.

Transaction effects of up to 3.48×

In published research measuring an actual sale, acquisition or ownership change, reported effect estimates range from modest but significant increases to 3.48 times higher acquisition odds. Examples of published comparisons include:

  • 5.8% versus 4.4% annual transaction incidence;
  • 5.0% versus 3.1% annual ownership-change incidence;
  • 5.0% versus 2.5% annual ownership-change incidence;
  • 3.2% to 5.8% around a dated external event; and
  • 32.3% versus 16.6% over a four-year ownership window.

These results represent effect ratios ranging from approximately 1.07× to 3.48×, depending on the population, signal and statistical method.

Measurable changes in exit timing

The research also indicates that relevant developments can alter when an ownership transition occurs. Published findings include:

  • approximately 21.5% higher annual exit hazard;
  • external transitions occurring roughly two years sooner;
  • observable pre-transaction activity beginning around five months before an announcement; and
  • operating trajectories developing over several years before a terminal business event.

For Seastone, this timing dimension is as important as the direction of the signal. A recent cluster of developments should not be treated the same way as an isolated event from several years ago.

Broader business-transition effects

Not every business exit becomes a conventional sale. Research covering company survival, distress, closure, default and restructuring helps the engine distinguish different pathways. Across the reviewed studies, reported effects include:

  • approximately 1.2× higher business mortality;
  • 1.33× to 1.67× higher broader exit likelihood;
  • 2.0× bankruptcy incidence in one comparison;
  • approximately 3× exit rates following certain external shocks;
  • 8× to 16× closure rates in severe enforcement categories; and
  • up to 10.3× baseline failure risk in the highest-risk classification.

These findings do not treat closure and sale as identical. They allow the engine to separate clean succession opportunities, strategic acquisitions, distressed transactions and companies that are more likely to disappear without a sale.

Strong conditional transaction evidence

Several studies also report a high transaction share once a more advanced condition has been observed. Depending on the company state and transaction context, published findings include:

  • approximately 21.4% completing a substantial going-concern asset sale;
  • roughly 25% ending in acquisition within a defined distressed population;
  • 51.2% of one termination category being caused by an acquisition offer;
  • around 53% of cases involving a formal sale process; and
  • up to approximately 85% of mature cases ending in a sale.

These signals are especially useful for validation and classification. They may become visible later than the earliest forward indicators, but they can materially increase confidence that an ownership transition is occurring.

Negative evidence improves prediction

The research also contains evidence associated with a lower probability of business disappearance or exit. One published comparison reported a 0.76× disappearance rate, while other company characteristics produced lower transition incidence than the relevant baseline.

Seastone retains this negative evidence deliberately. A system that knows when to reduce a company’s score is more useful than one that interprets every event as a buying opportunity.

Why combining signals matters

No individual signal can describe every company. A business may exhibit one relevant development and remain independent for years. Another may display several mutually reinforcing changes within a short period. A third may appear attractive until negative or contradictory evidence is considered.

Seastone evaluates the complete pattern. The engine considers:

  • whether the evidence is forward-looking;
  • the strength of the published relationship;
  • the age of the signal;
  • whether several independent observations converge;
  • whether the signal predicts a sale, distress, closure or another outcome;
  • the company’s underlying ownership and market context; and
  • whether negative evidence should reduce the score.

This turns isolated information into structured exit intelligence.

From research to a sourcing advantage

The studies behind the architecture use different populations, time periods and statistical methods. Their percentages should not be added together or averaged into a synthetic “probability of sale.” Their combined importance lies in what they establish collectively:

  1. observable developments can precede ownership transitions;
  2. companies exhibiting those developments can have materially different exit rates;
  3. the effect of a signal depends on its timing and context;
  4. negative and confirming evidence improve classification; and
  5. combining signals provides a more complete view than relying on a single trigger.

Seastone operationalizes those findings across a proprietary library of 36 signal families.

Conclusion

Published research repeatedly shows that ownership transitions are preceded, shaped or confirmed by observable changes. Across Seastone’s research base:

  • transaction-specific effects reach 3.48×;
  • broader terminal-business risk reaches 10.3× in high-risk categories;
  • absolute transaction comparisons differ by as much as 15.7 percentage points;
  • certain external transitions occur approximately two years earlier;
  • measurable pre-transaction activity appears months before announcement; and
  • advanced conditions show conditional sale shares of up to approximately 85%.

Seastone combines all 36 signal families—not as 36 isolated predictions, but as one evidence-driven system that raises, lowers, times and validates exit probability.

The objective is simple: identify companies entering a pre-exit state before the wider market recognizes the opportunity.

Research methodology

The evidence review includes academic research, administrative-record studies, longitudinal ownership data, cohort analyses, natural experiments and institutional transaction research.

The underlying studies measure different outcomes, including ownership transfers, acquisitions, broader business exits, distress events and transaction-confirmation states. Results are retained in their published form and are not mathematically pooled.

The research validates the role of each signal within Seastone’s predictive architecture. Product-level recall, precision and average lead time are measured separately through historical and live-market validation.