Lesson 6 of 6 · Who sells and why
Ranking a list: scoring vs filtering
6 min read · Updated Nov 16, 2026
The short answer
A hard filter removes every company that fails any one rule, including companies that only just miss it or whose data is missing. Scoring weighs all the criteria together and ranks the whole list, so a strong company with one near miss stays near the top and a weak one that passes every rule drops down.
By the end of this lesson you will be able to:
- Explain how a hard filter and a score treat the same company differently
- Show with a worked example how a filter drops good targets that a ranked list keeps
- Build a simple score and decide which rules should remain hard exclusions
Hard filter versus ranked score
Hard filter
Pass or fail
- One miss removes the company
- Missing data counts as a fail
- Just under the line is treated like far under it
- You never see what was cut
Ranked score
Every company gets a number
- Strengths can outweigh one weakness
- Missing data is marked unknown
- Near misses rank just below hits
- The whole list stays visible, in order
Key takeaways
- Filters fail companies on near misses and on missing data. Both are common in private company information.
- A score ranks every company, so judgment is applied to the top of a full list, not to what survived a cut.
- Keep a few true exclusions, such as the wrong country. Score everything else.
- Even official data has gaps: 14.4% of owner responses on year of acquisition in the 2023 Census survey were "Don't know".
Two ways to narrow a list
Once you have a target list built from your thesis, you have to decide who to contact first. A hard filter sets rules and deletes every company that fails one: revenue above $5 million, owner over 60, founded before 2000. A score gives each company points for how well it matches each criterion and sorts the list by the total.
What a hard filter gets wrong
- Near misses. A company with $4.8 million of revenue fails a $5 million floor exactly as a $1 million company does, although revenue figures for private companies are usually estimates.
- Missing data. A filter cannot tell "no" from "unknown". Gaps are normal: in the Census Bureau's Annual Business Survey for 2023, 14.4% of owner responses on the year they acquired the business were "Don't know".
- One rule outvotes the rest. A company that is ideal on five criteria and short on one is removed. A company that is mediocre on all six stays.
- Signals are not rules. In the IBBA and M&A Source Market Pulse for Q2 2026, between 60% and 90% of sellers had done less than one year of exit planning or none, so a filter that demands visible preparation is likely to remove owners who go on to sell.
Worked example: six companies
Suppose the thesis is specialty food distributors with $5 million to $20 million of revenue, and the filter also requires an owner aged 60 or older. The score uses the same facts plus fit with the thesis.
| Company | Facts | Hard filter | Score | Rank |
|---|---|---|---|---|
| A | $12M revenue, owner 66, strong fit | Pass | 88 | 2 |
| B | $4.6M revenue, owner 71, no successor, ideal product mix | Fail: revenue | 91 | 1 |
| C | $9M revenue, owner age unknown, same owner for 34 years | Fail: age missing | 84 | 3 |
| D | $15M revenue, owner 62, one customer is 60% of sales | Pass | 52 | 5 |
| E | $21M revenue, owner 58, general manager hired this year | Fail: revenue and age | 79 | 4 |
| F | $7M revenue, owner 45, bought the company in 2021 | Fail: age | 30 | 6 |
The filter keeps A and D. The ranked list puts B, A and C on top. The filter dropped the best target over $400,000 of estimated revenue, dropped another because one field was blank, and kept D, whose customer concentration puts it near the bottom.
How to build a score
- Take the criteria from your thesis: business fit, size, location, ownership and sell-readiness signals.
- Decide how much each one matters. Fit with the business you want usually matters most.
- Give partial credit near a threshold, and mark missing facts as unknown, not as zero.
- Keep the reason next to each score, so a colleague can see why a company ranks where it does.
- Review the top of the list and a sample from the bottom, then adjust.
When a filter is still right
Some rules are true exclusions: the wrong country, a company already owned by a private equity firm, a line of business your fund cannot hold. Filter on those, because no strength elsewhere changes the answer. Score everything that is a matter of degree.
How Equitle helps
Equitle scores every company against your own description of a great fit and ranks the list, with the reasons shown for each score. Close matches are kept and scored lower, not thrown away, and every keep and cut you make teaches Hawk what you mean, so the next list starts closer. The scoring runs on Equitle's side while you do something else.
Key terms
Sources
- U.S. Census Bureau (2025). 2024 Annual Business Survey (reference year 2023), Characteristics of Business Owners, table AB2300CSCBO: age of owner and year acquired ownership, for owners of respondent employer firms, overall and by sector. Source of the owner age and tenure shares. The estimates cover responding owners of employer firms, not all business owners. View source
- IBBA and M&A Source (2026). Market Pulse Report, second quarter 2026, Executive Summary: a survey of 255 business brokers and M&A advisors, fielded July 1 to 15, 2026, covering 181 transactions. Source of the reasons sellers went to market by deal size and the exit planning figures. Read in the full Executive Summary as hosted by a member firm. View source
Common questions
What is the difference between scoring and filtering a list?
Filtering removes every company that fails a rule. Scoring gives each company points across all criteria and sorts the list, so nothing is hidden and near misses stay visible.
Why do hard filters miss good acquisition targets?
They treat a near miss and missing data as failures. A company just under a revenue floor, or one whose owner's age is not public, is deleted even if it fits on every other count.
Should I ever use a hard filter?
Yes, for true exclusions such as the wrong country or an ownership type you cannot buy. Score anything that is a matter of degree.
How should missing data be treated in a score?
Mark it as unknown, give it neutral or partial credit and research it. Do not score it as zero, because that turns a gap in the data into a judgment about the company.
