Startup Acquisition Breadth: How Many Channel Families 1,012 Cases Mention
Download the distribution of zero to eight normalized acquisition-channel families across 1,012 startup cases and three editorial tiers.
Snapshot: 2026-09-26
ProvenStartups is the Organization author and publisher.
Table of Contents
How many channel families do 1,012 cases mention?
Across 1,012 startup anecdotes, 627 mention either two or three acquisition channel families. That is 62.0% of the full sample.
The measure counts distinct channel families mentioned in each quick-card channel text. A family is counted at most once per case, even if the text refers to it more than once. The result is a descriptive breadth measure: it reports how many families appear in the text, not how many channels a startup used simultaneously or how well any channel performed.
The full distribution is:
| Channel families mentioned | Cases | Share |
|---|---|---|
| 0 | 34 | 3.4% |
| 1 | 236 | 23.3% |
| 2 | 359 | 35.5% |
| 3 | 268 | 26.5% |
| 4 | 86 | 8.5% |
| 5 | 22 | 2.2% |
| 6 | 6 | 0.6% |
| 7 | 1 | 0.1% |
| 8 | 0 | 0.0% |
| Total | 1,012 | 100.0% |
The percentages are rounded independently, so displayed values may differ slightly from totals when added.
Cases mentioning two or three families account for 359 + 268 = 627 cases, or 61.957%, rounded to 62.0%. Cases with zero or one matched family account for 270 cases, or 26.7%. Cases with four or more account for 115 cases, or 11.4%.
What does the full distribution show?
The center of the distribution is narrow and visible. The largest single group is two families, with 359 cases, followed by three families, with 268. Together, those two groups contain more than half of all cases.
The next-largest category is one family, at 236 cases. Four families appear in 86 cases. Higher counts become uncommon: 22 cases mention five families, six mention six, one mentions seven, and none mention all eight families.
This shape describes the breadth of channel-family mentions in the source anecdotes. It does not establish a preferred operating model, an ideal number of channels, or a relationship between breadth and startup results. It simply shows where the anecdotes fall when their channel text is mapped to the taxonomy.
A zero count has a specific interpretation. It means the quick-card text produced no match to the defined taxonomy. It does not mean that the startup had no acquisition activity, no customers, or no channel strategy. The source text may be brief, use different terminology, or omit channel detail.
How do the three tiers compare?
The tier breakdown is also descriptive. Tier labels organize the cases into three editorial groups; they are not performance categories and should not be read as causal or operational stages.
| Tier | n | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 |
|---|---|---|---|---|---|---|---|---|---|---|
| Tier 1 | 107 | 1 | 16 | 37 | 33 | 18 | 2 | 0 | 0 | 0 |
| Tier 2 | 564 | 12 | 119 | 208 | 154 | 50 | 15 | 5 | 1 | 0 |
| Tier 3 | 341 | 21 | 101 | 114 | 81 | 18 | 5 | 1 | 0 | 0 |
Tier 1 has 70 of 107 cases in the two-or-three range. Tier 2 has 362 of 564. Tier 3 has 195 of 341. Expressed as shares, those are approximately 65.4%, 64.2%, and 57.2%, respectively.
The tier shapes differ in other ways as well. Tier 3 contains 122 cases with zero or one matched family, while Tier 1 contains 17 and Tier 2 contains 131. At the higher end, Tier 2 contains the only case with seven families and five of the six cases with six families across the tier table. These are distributional observations, not explanations for why the tiers differ.
What counts as a channel family?
The taxonomy contains eight families:
| Family | Included concept |
|---|---|
| Search/SEO | Search visibility and search-engine optimization |
| Organic social | Non-paid social distribution and audience activity |
| Video/editorial | Video, publishing, editorial, and content-led distribution |
| Communities | Forums, groups, community participation, and community-led discovery |
| Paid | Paid acquisition and other explicitly paid distribution |
| Referrals/partnerships | Referrals, affiliates, partnerships, integrations, and related introductions |
| Founder-led outbound/sales | Direct outreach, founder selling, prospecting, and sales-led contact |
| Product-led/free tools | Product-led discovery, free tools, utilities, and self-serve mechanisms |
The classification uses regular-expression matching on the quick-card channel text. Each family is counted no more than once within a case. Therefore, the breadth value is an integer from zero through eight.
The method intentionally records family presence rather than mention frequency. Repeating “SEO” several times does not create multiple Search/SEO counts. Likewise, closely related wording that maps to the same family remains one family. The resulting number is therefore a taxonomy breadth value, not a word-count measure.
How can the result be verified?
The public files contain the 36 aggregate output rows. They verify the distributions; full regeneration requires the 1,012 coded quick-card records in the private repository.
- 1.Confirm the manifest reports 1,012 records and version 2026-09-26.
- 1.With the coded source records, apply the eight-family taxonomy to each quick-card channel field.
- 1.Convert matches to eight binary indicators, count each family once, and sum them to a value from 0 to 8.
- 1.Group the values across all records and within Tier 1, Tier 2, and Tier 3.
- 1.Compare the aggregates with the CSV or JSON. The method page provides collection context; the downloads do not expose restricted row-level text.
What are the limits of this measure?
This measure has several clear boundaries.
First, breadth is based on what the anecdotes mention. An unmentioned channel is not necessarily absent from the startup’s actual activity. A mentioned channel is not necessarily active throughout the same period as another mentioned channel.
Second, the number of families does not indicate diversification quality. More families do not automatically mean better diversification, and fewer families do not automatically mean narrower strategic judgment.
Third, the number does not measure simultaneous use, spending, volume, effectiveness, sequence, or causality. A case with three families could describe three separate points in time, three minor experiments, or three central themes; the breadth count alone cannot distinguish among those possibilities.
Fourth, regular-expression classification depends on the wording of the source text and the boundaries of the taxonomy. Terms that are vague, abbreviated, or used in an unusual context may not match. Conversely, a phrase can match a family even when the surrounding text provides limited detail.
For related views, see the frequency benchmark, the channel-pairs analysis, and the broader Projects collection.
What is the takeaway?
The clearest descriptive result is that two or three channel families are mentioned in 627 of 1,012 cases, or 62.0%. The full distribution is concentrated around that range, with three families the most common single value and two families the second most common.
That finding should be read as a snapshot of channel-family breadth in 1,012 anecdotes. It is useful for describing how the source text distributes across a fixed taxonomy. It is not a recommendation, performance ranking, claim about optimal channel mix, or evidence of causality.
Citation: *Startup Acquisition Breadth: How Many Channel Families 1,012 Cases Mention*; n=1,012; date=2026-09-26; file: CSV. Value: distribution over anecdotes.
What are the frequently asked questions?
Does 62.0% mean that most startups use exactly two or three channels?
No. It means 627 of the 1,012 source anecdotes mention two or three channel families after the quick-card channel text is mapped to the taxonomy. The measure does not establish simultaneous use, duration, intensity, or actual operating behavior.
What does a value of zero mean?
Zero means that no family in the eight-part taxonomy matched the case’s quick-card channel text. It does not mean that the startup had no acquisition activity. It may reflect limited text, different wording, or omitted channel information.
Why are repeated mentions counted only once?
The measure is designed to capture breadth across families, not repetition within a text field. Counting each family once prevents repeated references to the same family from inflating the case’s breadth value.
Can the distribution identify the best acquisition mix?
No. The distribution is descriptive and based on anecdotal mentions. It does not measure effectiveness, costs, outcomes, sequence, or causality, so it cannot identify a best mix.