Why the method used to select searchers deserves the same scrutiny those same investors apply to a target company
By Javier García Manzanedo, Founder’s Risk Due Diligence for Investors
To make it easier for you to read this article, we suggest downloading it in PDF Format.
Javier García Manzanedo is an entrepreneurship psychologist specializing in founder risk due diligence for investors. His work focuses on the human factors that financial and legal due diligence often overlook, with particular attention to founder transitions, leadership assessment, and post-acquisition integration in founder-led businesses. Drawing on evidence-based assessment and psychometrics, he helps investors make safer decisions by identifying the people risks that shape continuity, performance, and long-term value creation.
In search funds, the diligence applied to a company has become genuinely rigorous. Quality of earnings, customer concentration, supplier dependency, working capital behaviour and regulatory exposure are all examined by people who know what they are looking for, using methods that can be described, questioned and repeated by someone else.
The diligence applied to the person who will run that company is a different matter, not because investors are careless, but because the method is rarely written down, rarely applied identically across candidates, and almost never revisited afterwards.
The previous articles in this series examined the searcher: how the nature of human risk shifts across the four phases, what happens when a buyer becomes a successor inside an inherited organisation, and how accumulated commitment distorts judgment as a deal approaches. All of them treat the searcher as the object of assessment, which leaves one question unasked. The searcher is selected, by someone, using something, and it is worth asking what that something is and how anyone would know whether it works.
This is the central argument: the process by which searchers are selected would not survive the standard the same investors apply to a target company, and the environment in which that process operates systematically denies its practitioners the feedback that would otherwise correct it over time.
Experience is not the same as calibration
Kahneman and Klein (2009) set out two conditions under which intuitive expertise becomes trustworthy: the environment must provide adequately valid cues to the nature of the situation, and the individual must have the opportunity to learn those cues, which in practice means prolonged exposure with feedback that is both rapid and unambiguous. Where those conditions hold, expert intuition is reliable and often superior to formal analysis; where they do not, confidence and accuracy come apart.
Searcher selection fails both conditions, and it fails them structurally rather than through anyone’s negligence.
The outcome arrives too late to teach anything, since a judgment made at the point of funding is tested five to seven years afterwards, by which time the result has been shaped by the company acquired, the price paid, the sector, the macro environment, the quality of the board, and a hundred decisions the searcher made in between. Attributing that outcome back to the assessment made at the outset is not a matter of care but of impossibility.
The counterfactual, meanwhile, is never observed at all, because an investor who declines a searcher does not learn what that searcher went on to do, so there is no record of the strong candidates who were passed over and no way to distinguish a selection process that identifies talent from one that merely identifies familiarity. Selection systems that never see their rejected cases cannot compute their own error rate, and therefore cannot improve.
Kahneman and Klein’s conclusion is the uncomfortable part, since they found that subjective experience is not a reliable indicator of judgment accuracy. Twenty years of selecting searchers produces twenty years of confidence, and whether it produces twenty years of skill depends on conditions this market does not currently provide.
Highhouse (2008) gave the pattern a name in the personnel selection literature — the myth of expertise, or the belief that experience improves one’s ability to predict human behaviour — and argued that practitioners resist structured selection aids less because of practical constraints than because deferring to a procedure feels like a concession about one’s own judgment. The resistance is understandable, but it is not supported.
The least examined instrument in the process
The interview sits at the centre of most searcher selection, and its properties are the least discussed part of the whole exercise.
Dana, Dawes and Peterson (2013) ran an experiment worth describing in full, because the design is what makes it persuasive. Participants interviewed candidates and predicted their academic performance, and in one condition the candidate answered at random, with yes or no determined by the last letter of the question. The interview was, literally, noise. Participants formed impressions from those interviews with the same confidence as from real ones, conducting interviews made their predictions less accurate than the background data alone would have made them, and in a third study participants preferred a random interview to no interview at all.
Kausel, Culbertson and Madrid (2016) reached a converging conclusion by a different route, finding that decision-makers given unstructured interview information alongside standardised test scores became measurably more overconfident in their predictions of job performance, and that in a task where participants bet real money on those predictions the overconfidence cost them.
None of this argues for abandoning interviews, since Levashina, Hartwell, Morgeson and Campion (2014), reviewing the evidence on employment interviews, found that their predictive value depends heavily on structure — consistent questions, consistent order, defined evaluation criteria and independent scoring — which suggests that the problem lies not with the interview but with the unstructured one.
The evidence has moved, and in a useful direction
For two decades the reference point in this field was Schmidt and Hunter (1998), which placed general mental ability at the top of the validity hierarchy for selection methods, appears in the reference list of the first article in this series, and has been enormously influential.
It has since been revisited, and substantially. Sackett, Zhang, Berry and Lievens (2022) argued that the standard corrections for range restriction had systematically overstated validity, and produced revised estimates that are substantially lower across the board; in their corrected table, structured interviews emerge with the highest validity among widely used selection methods, above general mental ability.
That result points somewhere unexpected, because the most defensible instrument available turns out to be neither proprietary nor expensive nor technically demanding. It is a conversation conducted the same way with every candidate, against criteria written down in advance.
One caution is worth adding, since the figures circulate widely: an updated set of estimates attributed to Schmidt, Oh and Shaffer (2016), frequently cited in practitioner material as though it were a published meta-analysis, is an unpublished working paper that never completed peer review. Anyone building an argument on selection validity should know which of these sources has been reviewed and which has not.
What structure would look like here
Applied to searcher selection, none of this requires new instruments; it requires the discipline these same investors already apply to a target company.
Criteria can be defined before meeting the candidates, and defined functionally — not qualities in the abstract, but what the role will demand across search, deal, integration and operation, which is the argument the first article in this series made in detail.
Every candidate can be asked the same questions in the same order, and scored independently before anyone discusses anyone, since group discussion held before individual judgment converts a set of independent estimates into a single correlated one, which is the opposite of what a panel is for.
Introduction can be separated from assessment, given that the investor who found and championed a searcher is, by the argument of the third article in this series, the least well placed to judge whether that searcher clears the bar.
And the decision can be recorded with its reasons, including the rejections. This is the step that is always skipped, and the only one that creates the possibility of ever knowing whether the process works, since a market that documented its searcher selection decisions for a decade would be able to answer questions it currently cannot even frame.
This is an assessment problem before it is a governance problem
The real question, then, is not whether investors exercise judgment about people. They obviously do, and much of it is good. The more important question is whether that judgment is supported by anything capable of correcting it.
Here the argument returns to measurement, because a characteristic that matters enough to influence a funding decision should be assessed in a way that is reliable, valid for the intended interpretation, and comparable across candidates. Those are not consultancy claims but the requirements set out in the Standards for Educational and Psychological Testing (AERA, APA, & NCME, 2014), and they apply to a selection process regardless of who is running it.
That does not mean an instrument resolves the decision; it means the decision rests on evidence someone else could examine. This is where an ordered methodology can make sense, and an approach such as APF® may be useful not to score the candidate, but to define what the role actually demands, organise hypotheses about functional fit, and make the resulting judgment explicit enough to be revisited later, which is the condition under which a process eventually learns something.
Conclusion
In search funds, the quality of an investment depends on the company and on the person who will run it. The first is examined with method that can be inspected, while the second is examined with judgment that the environment gives almost no opportunity to calibrate.
That asymmetry is not a criticism of investors but a description of a market that has professionalised one half of its own diligence, while leaving the other half to operate under conditions in which, according to the evidence, nobody’s intuition reliably improves with time.
Having spent three articles on the searcher, it seems reasonable to turn the same question around.
References
American Educational Research Association, American Psychological Association, & National Council on Measurement in Education. (2014). Standards for educational and psychological testing. American Educational Research Association.
Dana, J., Dawes, R., & Peterson, N. (2013). Belief in the unstructured interview: The persistence of an illusion. Judgment and Decision Making, 8(5), 512–520. https://doi.org/10.1017/S1930297500003612
Highhouse, S. (2008). Stubborn reliance on intuition and subjectivity in employee selection. Industrial and Organizational Psychology, 1(3), 333–342. https://doi.org/10.1111/j.1754-9434.2008.00058.x
Kahneman, D., & Klein, G. (2009). Conditions for intuitive expertise: A failure to disagree. American Psychologist, 64(6), 515–526. https://doi.org/10.1037/a0016755
Kausel, E. E., Culbertson, S. S., & Madrid, H. P. (2016). Overconfidence in personnel selection: When and why unstructured interview information can hurt hiring decisions. Organizational Behavior and Human Decision Processes, 137, 27–44. https://doi.org/10.1016/j.obhdp.2016.07.005
Kelly, P., & Heston, S. (2024). 2024 search fund study: Selected observations (Case No. E-870). Stanford Graduate School of Business.
Kowalewski, A.-S., Kelly, P., Simon, J., & Johnson, R. (2024). International search funds — 2024: Selected observations (Technical Note ST-658-E). IESE Business School.
Levashina, J., Hartwell, C. J., Morgeson, F. P., & Campion, M. A. (2014). The structured employment interview: Narrative and quantitative review of the research literature. Personnel Psychology, 67(1), 241–293. https://doi.org/10.1111/peps.12052
Sackett, P. R., Zhang, C., Berry, C. M., & Lievens, F. (2022). Revisiting meta-analytic estimates of validity in personnel selection: Addressing systematic overcorrection for restriction of range. Journal of Applied Psychology, 107(11), 2040–2068. https://doi.org/10.1037/apl0000994
Schmidt, F. L., & Hunter, J. E. (1998). The validity and utility of selection methods in personnel psychology: Practical and theoretical implications of 85 years of research findings. Psychological Bulletin, 124(2), 262–274. https://doi.org/10.1037/0033-2909.124.2.262
Schmidt, F. L., Oh, I.-S., & Shaffer, J. A. (2016). The validity and utility of selection methods in personnel psychology: Practical and theoretical implications of 100 years of research findings [Unpublished working paper].
Simon, J., & Kowalewski, A.-S. (2023). Nurturing leadership in search funds (Technical Note ST-647-E). IESE Business School.
