# Methodology: Which C-level seat turns over fastest?

By Cristian Frunze, Founder of Kenso · Nov 13, 2026 · [Kenso Research](/research) No. 23 · Hiring and org design, Sales teams, Founders and startups

The full methodology behind [Which C-level seat turns over fastest?](/research/executive-turnover.md): the data, every method, every number and every limitation. The same text as the [methodology PDF](/research/executive-turnover.pdf).

Data: Kenso internal data, Kenso's dataset of public professional profiles, work histories and company records.

## 1. Summary

Among 296,409 C-level seats held by 244,502 people who did not found the company, at 116,592 B2B companies with 10 or more employees at the start of the seat (seats started 2015-2025, followed to 2026-03), the Kaplan-Meier median time in the seat was: CRO 27, CPO 30, CMO 31, COO 41, CTO 41, CISO 44, CFO 46, CEO 64 months. The CRO seat turns over fastest, also after adjusting for company size, industry and start year (hazard ratio 1.70 against the CFO seat). Founder CEOs stay 96 months against 64 for non-founder CEOs under the same rules. 69% of exits go to another company within 12 months and 6% are promotions. Correcting for executive jobs left off profiles lowers the medians by 3 to 13 months and leaves the CRO the fastest seat. All numbers describe people who live in the US, UK, Canada, Australia, Ireland and New Zealand.

## 2. Question and pre-registration

**Question.** How long do CEOs, CFOs, CTOs, COOs, CMOs, CROs and CPOs stay in the seat at B2B companies, by company size and start year? The Kenso Research study *The 18-month VP of Sales is a myth: the median is 32 months* covers VPs of Sales; this compares every C-level seat (CISO added, as in the study brief).

**Pre-registration.** The hypotheses and plan were pre-registered on 2026-10-20, before any data was read; before that, only the list of data tables, their sizes and their column schemas were looked at. The hypotheses and the scored results are in Section 9.

## 3. Data

| Source | What was used | Size |
|---|---|---|
| Kenso internal data, people | Every work-history entry (title, company id, start and end month, current flag) of every profile, read from Kenso's shared person-level extract (built 2026-10-16/17). The data holds people whose current country is the US, UK, Canada, Australia, Ireland or New Zealand; their earlier jobs can be anywhere. Profiles were last refreshed between 2025-11 and 2026-04. | 555,053,752 entries of 171,326,491 people read; 4,584,770 people with a candidate C-level title kept with all their entries |
| Kenso internal data, company records | Industry label, HQ country, record type, founded year; rebuilt quarterly headcount from the same shared extract | 23,338,222 records |
| Kenso internal data, job changes | Job changes detected in 2026-06, filtered to previous titles that name a C-level seat, and those people's current work histories (short-job check, Section 7.6) | 26,654 of 1,822,157 job changes |
| Kenso Research VP of Sales tenure study | Published outputs, cited unchanged (Section 8.9) | - |

## 4. Sample

- **Companies:** company records with HQ in the six countries, an industry label in one of the six B2B industry groups of the Kenso Research study *Who gets cut first in a B2B layoff? Recruiters, then SDRs* (software and IT; other technology, telecom and media; financial services; business and professional services; manufacturing, industrial and energy; healthcare and life sciences; label map unchanged from that study), and a record type other than nonprofit, government agency, educational, self-employed or sole proprietorship. Company records carry no B2B flag; the industry groups are Kenso Research's B2B definition.
- **Seats:** a seat is one person in one C-level seat at one company. Seats started 2015-01 to 2025-12, at companies with 10 or more people in the rebuilt headcount at the end of the quarter that contains the start month (the size rule of the VP of Sales tenure study), held by people who are not founders of that company (Section 6).
- **Flow (core seats):**

| Step | Seats | People | Companies |
|---|---|---|---|
| all core seats (merged, interim and fractional left out) | 2,080,284 | 1,589,625 | 1,153,502 |
| company has a record | 1,643,513 | 1,297,592 | 848,327 |
| HQ in the six countries | 1,640,249 | 1,295,398 | 846,538 |
| B2B industry group | 1,079,958 | 873,080 | 555,098 |
| operating record type | 1,026,490 | 828,441 | 527,614 |
| started 2015-01 to 2025-12 | 669,615 | 565,368 | 377,871 |
| 10+ rebuilt employees at the start | 311,714 | 257,324 | 121,225 |
| non-founder holder (headline) | 296,409 | 244,502 | 116,592 |

Seat spells were dropped before this flow when the entry had no company id (2,717,808 of 5,361,520 seat spells), no start date, a past entry with no end date, an end before the start, a start after 2026-03, or an ambiguous acronym at an industry where it means another seat (Section 5).

## 5. Title rules

Titles are normalized as in the shared sales title taxonomy (lower case, every run of non-word characters replaced by one space). The rules are checked by 83 test cases.

- **Spelled seats.** Every "chief ... officer" span (up to 5 words between) is read word by word: executive (CEO); financial, finance, fiscal (CFO); technology, technical, tech (CTO); operating, operations, operational (COO); marketing (CMO); revenue (CRO); product (CPO); security, cyber, information security (CISO). "Chief executive" without "officer" is CEO. Phrases are read first: information security and product security are CISO, information technology is the chief information officer, product marketing is marketing, revenue cycle (healthcare billing) names no seat. A compound title such as "chief product and technology officer" counts in every seat it names.
- **Variants,** each its own seat: chief growth officer (and CGO), chief customer officer (customer, customer success, client, customer experience), chief commercial officer, chief sales officer, chief business officer, chief strategy officer, chief information officer, chief people officer (people, human resources, HR, talent, culture; CHRO), chief data officer (data, analytics), chief digital officer.
- **Acronyms.** CEO, CFO, CTO, COO, CMO, CRO, CPO, CISO, CIO, CGO and CHRO count as whole words, except when the acronym is the initials of a "chief ... officer" span in the same title ("chief risk officer CRO" is a risk officer), when the next word makes it another job ("CRO manager", "CMO relations", "CEO office"), or when the previous word does ("production technician CRO"). Ambiguous bare acronyms depend on the company: a bare CMO does not count at health and life-sciences companies (chief medical officer), a bare CRO not at financial-services companies (chief risk officer) or health and life-sciences companies (mostly chief research officer; this goes beyond the pre-registered text and was logged as a decision), a bare CIO not at financial-services companies (chief investment officer), and a bare CPO counts only at software and IT companies. The holders' other titles back these rules.
- **Not seats:** deputy, assistant, associate and vice seats; titles that point at someone else's seat (assistant to the CEO, chief of staff to the CEO, office of the CTO); field CTOs and CISOs; regional and divisional qualifiers (regional, EMEA, APAC, Americas, North America, Europe, division, country and similar); former, retired, aspiring, student and intern titles; advisory board and non-executive titles; "CEO coach" and similar. Interim, acting and fractional seats are flagged and left out of the main build (a sensitivity row includes them).
- **Audit.** Before any tenure was computed, the most common matched titles per seat, the most common unmatched titles containing "chief" or a seat acronym, and the most common titles dropped by a negating rule were read (titles held by 50 or more people only), and the rules were fixed once. 549,415 of 898,272 distinct candidate titles name a seat. Match counts over every candidate spell, any company:

| Seat | Distinct titles | Spells | People | Spells named by a bare acronym | Interim spells | Fractional spells | Titles dropped by a negating rule |
|---|---|---|---|---|---|---|---|
| CEO | 241,322 | 3,109,458 | 2,585,776 | 2,203,461 | 36,709 | 2,158 | 81,291 |
| CFO | 84,158 | 707,886 | 474,649 | 325,964 | 22,527 | 12,928 | 34,746 |
| CTO | 41,996 | 297,044 | 226,520 | 148,668 | 4,815 | 4,814 | 11,180 |
| COO | 87,871 | 671,886 | 582,292 | 221,409 | 9,407 | 4,605 | 26,821 |
| CMO | 24,328 | 130,666 | 102,951 | 39,823 | 2,595 | 9,333 | 6,116 |
| CRO | 9,656 | 45,739 | 38,015 | 11,249 | 711 | 1,463 | 2,468 |
| CPO | 9,874 | 43,998 | 37,245 | 14,025 | 603 | 957 | 2,046 |
| CISO | 10,713 | 31,302 | 22,167 | 10,446 | 1,222 | 749 | 3,017 |
| Chief Growth Officer | 2,775 | 16,228 | 14,899 | 859 | 154 | 405 | 581 |
| Chief Customer Officer | 2,933 | 13,110 | 11,851 | 0 | 179 | 72 | 823 |
| Chief Commercial Officer | 3,666 | 28,781 | 25,603 | 0 | 381 | 188 | 1,092 |
| Chief Sales Officer | 1,851 | 11,963 | 11,065 | 0 | 51 | 82 | 401 |
| Chief Business Officer | 3,007 | 16,183 | 14,770 | 0 | 117 | 59 | 586 |
| Chief Strategy Officer | 7,684 | 38,271 | 35,032 | 0 | 292 | 249 | 1,603 |
| Chief Information Officer | 33,916 | 131,098 | 98,242 | 63,081 | 5,267 | 2,187 | 15,922 |
| Chief People Officer | 9,259 | 50,144 | 38,616 | 3,986 | 1,822 | 807 | 4,117 |
| Chief Data Officer | 3,251 | 9,241 | 7,966 | 0 | 239 | 75 | 815 |
| Chief Digital Officer | 2,590 | 8,522 | 7,364 | 0 | 284 | 64 | 811 |

## 6. Seats, censoring and founders

- **Merging.** A person's spells in the same seat at the same company that overlap or start at most one month after the previous one ends are one seat (the rule of the VP of Sales tenure study): earliest start, latest end, current if the latest spell is current.
- **Tenure** in months = end month - start month + 1. Observation ends at 2026-03, the last complete quarter before the profile refreshes: seats marked current, and seats whose end month is after 2026-03, are censored there.
- **Founders** (the rule of the Kenso Research study *The first 20 hires: engineers first, a seller by hire 7*): a seat holder is a founder of the company when any of their entries there has a founder title (founder, co-founder, founding partner, owner and similar, but not job titles such as product owner), or when their earliest dated entry at the company starts before 1 January of the year after the founded year on the company record and has a C-level or managing director title. Headline comparisons use non-founder seats.

## 7. Methods

1. **Tenure.** Kaplan-Meier survival in the seat; the median is the first month at which the survival curve is at or below 50%, with a 95% interval from the log-log Greenwood band; shares still in the seat at 12, 18, 24, 36 and 60 months.
2. **Cuts.** Size band at the start (10-49, 50-199, 200-999, 1,000+), industry group, start year and start-year group, country of the person.
3. **Adjusted comparison.** Cox proportional hazards model on the headline core seats with seat, size band, industry group and start-year group (references CFO, 200-999, software and IT, 2015-2017), Breslow ties, robust errors clustered by person.
4. **Exits.** Headline seats that ended 2015-01 to 2024-12, so the full 12-month window after the exit ends by 2025-12. The exit is classified from the person's other entries that start from 2 months before to 12 months after the exit month, first match in this order: promoted (same company, CEO, president, managing director or chair title; for a CEO, a chair title), other move inside the company, moved to another company (C-level seat or other role), left (no new entry in the window).
5. **Start years** 2010-2025 for the series (any size, since rebuilt headcount starts in 2015).
6. **Short jobs left off profiles** (method 2 of Kenso's earlier study of sales tenure). Among executives in the June 2026 job-change data whose previous title is a core seat and whose new job is on their profile, the missing share m is the share whose profile no longer lists the previous company at all (low variant: not counted as missing when the profile lists, with no company, a title naming one of the previous title's core seats). Correction: m / (1 - m) times the number of ended headline seats is added as extra seats lasting 1-6 months (low, with the low m) or 1-12 months (high, with m), spread evenly, and each median is recomputed.
7. **Sensitivities:** observation end 2025-10; unmerged spells; bare acronyms excluded; chief growth officer in CMO and chief sales officer in CRO; founders by title only; any HQ country; interim and acting spells included; interim and acting seats alone; the VP of Sales tenure study's basis; any size at the start.
8. **Privacy.** Groups under 50 people are suppressed; published title lists hold only titles with 50 or more holders; only aggregate tables leave Kenso's servers.

## 8. Results

### 8.1 Tenure by seat (headline sample)

| Seat | Seats | People | Exits | Median months (95% CI) | In seat 12 m | 24 m | 36 m | 60 m | Completed seats only |
|---|---|---|---|---|---|---|---|---|---|
| CRO | 15,812 | 14,002 | 10,017 | 27 (26-27) | 85% | 56% | 37% | 19% | 20 |
| CPO | 9,213 | 8,349 | 5,765 | 30 (30-31) | 88% | 62% | 42% | 21% | 23 |
| CMO | 18,642 | 15,869 | 12,238 | 31 (31-32) | 88% | 63% | 44% | 25% | 24 |
| COO | 71,836 | 68,211 | 39,437 | 41 (40-41) | 89% | 70% | 55% | 36% | 25 |
| CTO | 31,815 | 28,890 | 17,948 | 41 (40-41) | 90% | 71% | 55% | 35% | 25 |
| CISO | 8,682 | 7,255 | 4,829 | 44 (43-45) | 91% | 74% | 58% | 37% | 27 |
| CFO | 80,170 | 65,209 | 42,529 | 46 (46-47) | 92% | 75% | 59% | 40% | 26 |
| CEO | 60,239 | 56,117 | 25,434 | 64 (63-65) | 94% | 82% | 70% | 52% | 30 |

### 8.2 Variant seats (same sample rules)

| Variant seat | Seats | People | Median months (95% CI) | In seat 24 m |
|---|---|---|---|---|
| Chief Growth Officer | 4,914 | 4,722 | 30 (29-31) | 60% |
| Chief Customer Officer | 5,490 | 5,136 | 32 (31-33) | 63% |
| Chief Commercial Officer | 9,815 | 9,310 | 32 (31-33) | 63% |
| Chief Sales Officer | 2,923 | 2,831 | 34 (32-36) | 64% |
| Chief Strategy Officer | 8,584 | 8,269 | 35 (34-36) | 65% |
| Chief Business Officer | 3,249 | 3,111 | 36 (34-38) | 65% |
| Chief Digital Officer | 2,214 | 2,119 | 36 (34-37) | 70% |
| Chief Data Officer | 2,748 | 2,564 | 37 (36-38) | 69% |
| Chief Information Officer | 14,230 | 12,681 | 47 (46-48) | 77% |
| Chief People Officer | 14,434 | 12,451 | 47 (46-48) | 76% |

### 8.3 Founders and non-founders

| Seat | Basis | Non-founder people | Non-founder median | Founder people | Founder median | Founders in seat at 60 m |
|---|---|---|---|---|---|---|
| CEO | 10+ employees at the start | 56,117 | 64 | 7,960 | 96 | 65% |
| CEO | any size at the start | 111,745 | 112 | 148,427 | not reached | 82% |
| CFO | 10+ employees at the start | 65,209 | 46 | 1,218 | 46 | 41% |
| CFO | any size at the start | 82,194 | 50 | 7,390 | 92 | 61% |
| CTO | 10+ employees at the start | 28,890 | 41 | 1,950 | 49 | 42% |
| CTO | any size at the start | 45,146 | 46 | 25,817 | 72 | 55% |
| COO | 10+ employees at the start | 68,211 | 41 | 2,407 | 47 | 41% |
| COO | any size at the start | 103,965 | 48 | 22,193 | 81 | 58% |
| CMO | 10+ employees at the start | 15,869 | 31 | 436 | 35 | 25% |
| CMO | any size at the start | 22,344 | 34 | 4,796 | 61 | 50% |
| CRO | 10+ employees at the start | 14,002 | 27 | 498 | 36 | 26% |
| CRO | any size at the start | 17,076 | 28 | 2,115 | 55 | 45% |
| CPO | 10+ employees at the start | 8,349 | 30 | 635 | 39 | 32% |
| CPO | any size at the start | 10,404 | 32 | 3,593 | 56 | 48% |
| CISO | 10+ employees at the start | 7,255 | 44 | <50 | suppressed | |
| CISO | any size at the start | 8,148 | 46 | 321 | 115 | 67% |

Under the headline rules most founder seats drop out, because a founder's seat starts when the company is small; basis (b) compares all seats at frame companies started 2015-2025.

### 8.4 Company size at the start (median months)

| Seat | 10-49 | 50-199 | 200-999 | 1,000+ |
|---|---|---|---|---|
| CRO | 26 | 26 | 27 | 30 |
| CPO | 31 | 30 | 30 | 31 |
| CMO | 29 | 30 | 34 | 36 |
| COO | 44 | 40 | 38 | 36 |
| CTO | 43 | 40 | 39 | 38 |
| CISO | 50 | 42 | 43 | 44 |
| CFO | 50 | 44 | 44 | 41 |
| CEO | 75 | 61 | 58 | 49 |

### 8.5 Industry group (median months)

| Industry group | CRO | CPO | CMO | COO | CTO | CISO | CFO | CEO |
|---|---|---|---|---|---|---|---|---|
| Software and IT | 25 | 30 | 26 | 35 | 38 | 40 | 37 | 52 |
| Other technology, telecom and media | 29 | 31 | 29 | 37 | 43 | 46 | 42 | 56 |
| Financial services | 31 | 30 | 39 | 44 | 42 | 48 | 54 | 69 |
| Business and professional services | 26 | 30 | 37 | 45 | 42 | 55 | 49 | 70 |
| Manufacturing, industrial and energy | 32 | 31 | 34 | 42 | 46 | 45 | 49 | 71 |
| Healthcare and life sciences | 30 | 36 | 29 | 37 | 41 | 46 | 39 | 55 |
| Consumer, public sector and the rest (contrast, not B2B) | 28 | 29 | 30 | 38 | 39 | 44 | 43 | 59 |

### 8.6 Start years

Share still in the seat at 24 months, by start-year group:

| Seat | 2015-2017 | 2018-2020 | 2021-2023 | 2024-2025 |
|---|---|---|---|---|
| CRO | 61% | 60% | 52% | 57% |
| CPO | 67% | 65% | 58% | 64% |
| CMO | 65% | 64% | 61% | 63% |
| COO | 71% | 71% | 69% | 72% |
| CTO | 72% | 73% | 70% | 72% |
| CISO | 76% | 75% | 72% | 73% |
| CFO | 77% | 76% | 72% | 76% |
| CEO | 83% | 84% | 80% | 82% |

Share of seats that ended within 6 months, by start year (frame, any size, non-founders; Kaplan-Meier):

| Start year | CRO | CPO | CMO | COO | CTO | CISO | CFO | CEO |
|---|---|---|---|---|---|---|---|---|
| 2010 | 0% | 0% | 1% | 1% | 1% | 0% | 1% | 1% |
| 2011 | 1% | 1% | 2% | 2% | 2% | 1% | 1% | 1% |
| 2012 | 4% | 1% | 2% | 1% | 2% | 0% | 1% | 1% |
| 2013 | 1% | 3% | 2% | 2% | 2% | 1% | 1% | 1% |
| 2014 | 3% | 1% | 2% | 2% | 2% | 2% | 1% | 1% |
| 2015 | 2% | 2% | 2% | 2% | 2% | 1% | 1% | 1% |
| 2016 | 3% | 2% | 3% | 2% | 2% | 1% | 2% | 1% |
| 2017 | 3% | 2% | 3% | 2% | 3% | 3% | 2% | 1% |
| 2018 | 4% | 2% | 3% | 3% | 2% | 2% | 2% | 1% |
| 2019 | 3% | 3% | 3% | 2% | 3% | 2% | 2% | 1% |
| 2020 | 3% | 3% | 3% | 2% | 2% | 2% | 2% | 1% |
| 2021 | 3% | 3% | 3% | 3% | 2% | 2% | 2% | 1% |
| 2022 | 4% | 3% | 4% | 3% | 3% | 3% | 2% | 1% |
| 2023 | 4% | 3% | 5% | 3% | 3% | 3% | 3% | 2% |
| 2024 | 4% | 3% | 5% | 3% | 3% | 3% | 3% | 2% |
| 2025 | 4% | 4% | 5% | 3% | 4% | 3% | 3% | 2% |

### 8.7 Country of the person (median months)

| Seat | US | GB | CA | AU | IE and NZ |
|---|---|---|---|---|---|
| CRO | 27 | 25 | 28 | 24 | 28 |
| CPO | 30 | 31 | 32 | 30 | 31 |
| CMO | 32 | 29 | 33 | 31 | 31 |
| COO | 42 | 37 | 44 | 35 | 39 |
| CTO | 41 | 39 | 41 | 37 | 43 |
| CISO | 44 | 43 | 47 | 39 | 53 |
| CFO | 48 | 42 | 48 | 40 | 49 |
| CEO | 68 | 56 | 75 | 51 | 63 |

### 8.8 Exits

| Seat | Exits | Promoted | Other move inside | Moved: C-level seat | Moved: other role | Left | Moves into a founder title |
|---|---|---|---|---|---|---|---|
| CEO | 20,281 | 2% | 12% | 35% | 33% | 17% | 11% |
| CFO | 35,331 | 4% | 9% | 47% | 27% | 13% | 4% |
| CTO | 14,712 | 2% | 14% | 31% | 41% | 11% | 12% |
| COO | 32,236 | 14% | 15% | 25% | 34% | 12% | 9% |
| CMO | 10,276 | 2% | 12% | 38% | 36% | 12% | 11% |
| CRO | 7,815 | 6% | 11% | 36% | 38% | 9% | 10% |
| CPO | 4,579 | 4% | 15% | 32% | 40% | 9% | 14% |
| CISO | 3,944 | 0% | 10% | 41% | 39% | 10% | 6% |
| All core seats | 129,174 | 6% | 12% | 36% | 34% | 13% | 9% |

With a same-company role held at the exit and continuing after it counted as an internal move:

| Seat | Promoted | Other move inside | Moved: C-level seat | Moved: other role | Left |
|---|---|---|---|---|---|
| CEO | 2% | 16% | 34% | 32% | 16% |
| CFO | 4% | 11% | 47% | 26% | 13% |
| CTO | 2% | 16% | 31% | 41% | 10% |
| COO | 14% | 17% | 25% | 34% | 10% |
| CMO | 2% | 13% | 37% | 36% | 11% |
| CRO | 6% | 12% | 36% | 38% | 9% |
| CPO | 4% | 16% | 32% | 40% | 8% |
| CISO | 0% | 12% | 40% | 39% | 9% |
| All core seats | 6% | 14% | 35% | 33% | 12% |

### 8.9 VPs of Sales (VP of Sales tenure study) next to the C-level seats on that study's basis

| Row | People | Median months (95% CI) | In seat 24 m |
|---|---|---|---|
| VP of Sales (VP of Sales tenure study) | 102,652 | 35 (35-36) | 65% |
| CRO family (VP of Sales tenure study) | 14,675 | 26 (26-27) | 56% |
| Head of Sales (VP of Sales tenure study) | 28,863 | 26 (26-27) | 54% |
| Chief Sales Officer (VP of Sales tenure study) | 3,288 | 33 (32-35) | 63% |
| All top sales leaders (VP of Sales tenure study headline) | 143,761 | 32 (32-32) | 62% |
| CEO | 119,384 | 67 (66-68) | 82% |
| CFO | 107,693 | 47 (46-47) | 74% |
| CTO | 39,568 | 40 (40-41) | 71% |
| COO | 121,161 | 40 (40-40) | 70% |
| CMO | 26,098 | 32 (31-32) | 63% |
| CRO | 17,426 | 27 (27-28) | 56% |
| CPO | 10,888 | 30 (30-31) | 62% |
| CISO | 10,737 | 45 (44-46) | 75% |

### 8.10 Short jobs left off profiles

| Seat | Leavers in the check | Missing share | Low variant | Observed median | Corrected median, low | Corrected median, high | In seat at 12 m, high correction |
|---|---|---|---|---|---|---|---|
| CRO | 716 | 21% | 18% | 27 | 24 | 23 | 72% |
| CPO | 442 | 19% | 17% | 30 | 26 | 26 | 77% |
| CMO | 1,177 | 22% | 17% | 31 | 27 | 25 | 74% |
| COO | 3,758 | 20% | 16% | 41 | 36 | 34 | 78% |
| CTO | 2,422 | 26% | 21% | 41 | 34 | 32 | 75% |
| CISO | 276 | 14% | 12% | 44 | 40 | 39 | 83% |
| CFO | 3,093 | 18% | 12% | 46 | 42 | 39 | 82% |
| CEO | 11,326 | 27% | 19% | 64 | 55 | 51 | 81% |

Pooled over core seats: 22,921 leavers, missing share 23.8% (low variant 17.8%).

### 8.11 Cox model

| Term | Hazard ratio | 95% CI |
|---|---|---|
| seat=ceo | 0.71 | 0.70-0.72 |
| seat=ciso | 0.95 | 0.92-0.98 |
| seat=cmo | 1.44 | 1.41-1.47 |
| seat=coo | 1.14 | 1.13-1.16 |
| seat=cpo | 1.46 | 1.42-1.50 |
| seat=cro | 1.70 | 1.66-1.74 |
| seat=cto | 1.05 | 1.03-1.07 |
| size_band=1,000+ | 1.10 | 1.08-1.12 |
| size_band=10-49 | 0.87 | 0.86-0.88 |
| size_band=50-199 | 0.97 | 0.96-0.99 |
| group=business_services | 0.78 | 0.77-0.79 |
| group=financial | 0.74 | 0.73-0.75 |
| group=industrial | 0.78 | 0.77-0.79 |
| group=life_sciences | 0.95 | 0.93-0.98 |
| group=tech_media | 0.89 | 0.87-0.91 |
| year_group=2018-2020 | 0.99 | 0.97-1.00 |
| year_group=2021-2023 | 1.04 | 1.03-1.05 |
| year_group=2024-2025 | 1.01 | 0.98-1.03 |

### 8.12 Sensitivities (median months)

| Sensitivity | CRO | CPO | CMO | COO | CTO | CISO | CFO | CEO |
|---|---|---|---|---|---|---|---|---|
| main (headline) | 27 | 30 | 31 | 41 | 41 | 44 | 46 | 64 |
| observation ends 2025-10 | 26 | 30 | 31 | 40 | 40 | 43 | 45 | 63 |
| unmerged spells | 26 | 29 | 31 | 39 | 39 | 41 | 43 | 61 |
| bare acronyms excluded | 27 | 30 | 32 | 41 | 42 | 45 | 48 | 63 |
| bare CRO dropped only at financial-services companies (Plan wording) | 27 | 30 | 31 | 41 | 41 | 44 | 46 | 64 |
| chief growth officer in CMO, chief sales officer in CRO | 28 | 30 | 31 | 41 | 41 | 44 | 46 | 64 |
| founders by founder titles only | 27 | 30 | 31 | 41 | 41 | 44 | 46 | 65 |
| any HQ country | 27 | 30 | 31 | 41 | 41 | 44 | 46 | 64 |
| interim and acting spells included | 26 | 30 | 30 | 40 | 39 | 42 | 44 | 61 |
| interim and acting seats only | 9 | 9 | 10 | 11 | 10 | 11 | 9 | 8 |
| VP of Sales tenure study basis (all industries, any HQ, 10+, founder titles in the seat excluded) | 27 | 30 | 32 | 40 | 40 | 45 | 47 | 67 |
| any size at the start | 28 | 32 | 34 | 48 | 46 | 46 | 50 | 112 |

## 9. Hypotheses

| # | Pre-registered claim | Result | Verdict |
|---|---|---|---|
| 1 | CRO shortest median of the 8 core seats; CMO among the 3 shortest; both 6+ months below non-founder CEO | ranking CRO 27, CPO 30, CMO 31, COO 41, CTO 41, CISO 44, CFO 46, CEO 64 | holds |
| 2 | CMO median above 24 months | CMO 31 months | holds |
| 3 | founder CEO median at least 1.5 x non-founder CEO median (headline rules) | 96 vs 64 months, ratio 1.50 | holds |
| 4 | median shorter at 10-49 than at 1,000+ employees in 6+ of 8 seats | 2 of 8 (CMO, CRO) | wrong |
| 5 | 24-month in-seat share lower for 2021-2023 starts than 2015-2017 in 6+ of 8 seats | 8 of 8 | holds |
| 6 | promotion under 10% of exits for every non-CEO seat; COO highest; moved > left for every seat | promotion CFO 3.8%, CTO 2.1%, COO 14.1%, CMO 2.4%, CRO 5.6%, CPO 3.9%, CISO 0.3%; COO is the most promoted non-CEO seat; moving to another company outnumbers leaving with no new role for every seat | wrong |
| 7 | high short-job correction moves every median by under 6 months and keeps the fastest seat | changes (high) CEO -13, CFO -7, CTO -9, COO -7, CMO -6, CRO -4, CPO -4, CISO -5; fastest after correction CRO (high), CRO (low) | wrong |

## 10. Verification

An independent re-analysis, written from the study plan alone without seeing our code, tables or report, recomputed 178 numbers. It received the pre-registered plan, the definitional decisions, the title rules, the industry map and the same data extracts. The recomputed numbers cover every headline median with its interval, the seat, people and exit counts, the in-seat shares, the ranking, the founder comparison, the size and start-year cuts, the exit routes, the short-job shares and corrections, two Cox hazard ratios and the medians on the VP of Sales tenure study's basis. The match rules were fixed before any of its numbers were read: counts equal, medians equal in whole months, interval bounds within 1 month, shares within 0.1 percentage point, hazard ratios within 0.01 and corrected medians within 1 month.

In the first comparison 132 of 178 numbers matched. The differences traced to this study's code or to definitions the plan left open: a tie in the current-seat flag (two exit counts off by 1), company records with a founded year after 2026 (one founder count), which spells count in an exit and which titles count as a promotion or a C-level destination (the exit-route shares), and two details of the short-job check (its base and the low variant). Each was fixed and the analysis rerun, and the independent re-analysis recomputed the 61 affected numbers under the pinned rules. Now 178 of 178 match. No headline median or interval moved in the fixes. One rule in this study is broader than the plan's text: a bare "CRO" is also left out at healthcare and life-sciences companies, where most of those titles belong to research officers. Under the plan's wording the CRO median and interval are the same (Section 8.12).

## 11. Alternative explanations and limitations

1. *Profiles miss short jobs, so tenure is too long.* Measured directly (Section 8.10): about a quarter of executives who changed jobs in June 2026 no longer listed the job they left. If those were short stints, the medians fall by 3 (CRO) to 13 months (CEO). The ranking of the fastest seats does not change. Limitation for the levels, not the ranking.
2. *Older cohorts look more stable because old profiles lost their short jobs.* The share of seats ending within 6 months rises with the start year, for CMOs from 1% for 2010 starts to 5% for 2025 starts and for CEOs from 1% to 2% (Section 8.6), which fits this explanation as well as a real change. The 2015-2017 against 2021-2023 comparison (hypothesis 5) is therefore read as partly a data effect.
3. *Company mix: CROs work at young software companies that churn everyone.* The Cox model adjusts for size, industry and start years and keeps the CRO hazard 1.70 times the CFO's (Section 8.11). CRO has the shortest median in every size band and every country (Sections 8.4 and 8.7), and in four of the seven industry groups, software and IT included; in the other three one seat is shorter by 1 month (CPO in financial services and in manufacturing, industrial and energy, CMO in healthcare and life sciences; Section 8.5).
4. *A new title: CRO seats skew to recent starts, which turn over faster.* CRO has the lowest 24-month in-seat share in every start-year group (Section 8.6) and the shortest median in every start year from 2015 to 2023, and the Cox model adjusts for start years.
5. *Retitling inside the company: a CRO renamed chief commercial officer or president ends the seat without leaving.* 6% of CRO exits are promotions and 11% other moves inside the company, close to the CFO's 4% and 9% and the CMO's 2% and 12% (Section 8.8). Merging chief sales officer into CRO gives 28 months (Section 8.12).
6. *Title inflation: some "CRO" or "CMO" titles are small-company or vanity titles.* The 10+ employee rule removes companies too small to have such a team; CRO and CMO have the two shortest medians at 10-49 employees, and CRO is still the shortest at 1,000+ (Section 8.4).
7. *Late reporting: people add exits months late, so recent seats look longer.* Ending observation at 2025-10 instead of 2026-03 moves each median by at most 1 month (Section 8.12).
8. *Title edits at promotion: a CRO who becomes CEO may overwrite the old title.* This shortens nothing that is visible and cannot be measured (title-change history is not in the data); the promotion shares in Section 8.8 are lower bounds.
9. *Ambiguous acronyms.* Excluding every bare acronym moves each median by at most 2 months and keeps the CRO the fastest seat, and the plan's narrower rule for a bare "CRO" gives the same CRO median (Section 8.12).
10. *Founders counted as hired executives.* Classifying founders by founder titles only moves each median by at most 1 month (Section 8.12).
11. *Profiles that exist today.* Executives who deleted their profiles are missing; if fast leavers delete more often, the levels are biased. This cannot be measured here.
12. *Six countries.* Kenso internal data holds people living in the US, UK, Canada, Australia, Ireland and New Zealand, so seats held from elsewhere are missing and rebuilt headcount undercounts companies with staff abroad. The frame is limited to companies with HQ in these six countries.
13. *Causation.* The study compares seats; it does not show why executives leave.

## 12. Sources

- Kenso internal data (Kenso's dataset of public professional profiles, work histories and company records), through the shared headcount history extract built for the Kenso Research study *Did AI kill the SDR role?* and two small filtered pulls (Section 3).
- The shared sales title taxonomy, built for the Kenso Research study *Who are GTM engineers, and where do they come from?*
- Other Kenso Research studies: the VP of Sales tenure study, the first 20 hires study (founder rule), the layoffs study (industry groups) and the earlier tenure study (short-job method).