Search practice

A list is not a thesis.

The advantage is a decision you can defend, not another spreadsheet of companies.

Name the decision

A useful search thesis says which business characteristics make ownership attractive for this buyer, which conditions disqualify a target, and what observations would change that position. “Good businesses with growth potential” has no screening power. Neither does an industry label without a customer, geography, operating model and financing context.

Begin with one decision: which two subcategories deserve the next month of work? This forces comparisons and makes a research agenda more valuable than an undifferentiated queue.

Give must-haves a veto

A mandatory geography, operating capacity, revenue floor or prohibited exposure cannot be averaged away by attractive website copy. Separate hard gates from preferences. Define what evidence can satisfy each gate, how old it may be, and what happens when it is absent.

Unknown should have its own state. A company can look attractive on the two fields we know and still require evidence on the eight fields that determine whether it fits. Keep overall coverage beside the weighted score.

Write the thesis twice

First build a bottom-up thesis from sourced conversations and observations, without giving the synthesizer the principal’s existing answer. Then compare that output with the working thesis. Preserve disagreement; do not request a harmonized story merely because the original thesis came from a senior person.

For every observation retain who said it, when, what they were in a position to know, conviction, relevance and whether it supports or challenges the thesis. Repeated hearsay is not independent corroboration.

The deliverable is a test

Conclude with a shortlist of research priorities, explicit exclusions, and the cheapest falsification test for each. Our recommended rule: every strong recommendation must identify the fact most likely to overturn it. The Search Desk exports the evidence and the working thesis in separate phases so the comparison is possible without quietly leaking the answer into step one.

Your next move: replace one important assumption with a dated piece of evidence.
Open the owner tools