# Methodology: How long do SDRs stay in the seat? Half leave by month 14

By Cristian Frunze, Founder of Ken · Oct 24, 2026 · [Ken Research](/research) No. 13 · Sales teams, Hiring and org design

The full methodology behind [How long do SDRs stay in the seat? Half leave by month 14](/research/sdr-tenure.md): the data, every method, every number and every limitation. The same text as the [methodology PDF](/research/sdr-tenure.pdf).

Data snapshot: Ken internal data, people records of June 2026. Analysis run 2026-10-16. Every number below is written by the study's analysis code to its output files.

## 1. Question

How long do sales development and business development reps (SDRs and BDRs) stay in the role, how does that compare with the time it takes them to ramp, and how does it differ by company size, industry and start year?

## 2. Data

- **People and work histories:** Ken internal data, Ken's dataset of public professional profiles and work histories, for six countries (US, GB, CA, AU, IE, NZ). We used the shared sales careers extract (built 2026-10-15 from the June 2026 snapshot for the Ken Research study *Who are GTM engineers, and where do they come from?*). It holds every experience spell of every person with at least one sales or marketing title. We kept every spell of every person with at least one SDR or BDR spell: 517,024 people and 3,977,341 spells, of which 687,214 are SDR or BDR spells at any level.
- **Companies:** Ken internal data company records for the companies that employ the SDRs (current industry and current size band), pulled by company id. 83% of headline seats matched a company record; the rest are shown as "unknown".
- **Job changes:** Ken internal data job changes detected month to month. Only one detection month exists (June 2026). We used it to measure how often an SDR job is missing from the profile at the moment the person moves on (section 7).
- **Ramp time:** published benchmarks, not Ken internal data:

| Source | Metric | Value |
|---|---|---|
| The Bridge Group, SDR Models, Motions and Metrics 2025 | average SDR ramp time (months) | 3.0 |
| The Bridge Group, SDR Models, Motions and Metrics 2025 | peak average SDR ramp time, 2014 (months) | 3.8 |
| The Bridge Group, 2023 SDR Metrics report | average SDR ramp time (months) | 3.2 |
| The Bridge Group, 2023 SDR Metrics report | median months at full productivity (tenure minus ramp, survey) | 16 |

## 3. Title rules

SDR and BDR titles are found by one rule (`sdr_bdr`) of the shared sales title taxonomy, used unchanged. Titles are lower-cased and every run of non-word characters becomes a space. The rule matches any of:

- the abbreviations `sdr`, `bdr`, `mdr`, `adr` (and plurals) as whole words;
- `sales development`, `lead development`, `market development`, `account development`, `outbound development`, `inbound development`;
- `business development` followed by `representative`, `rep` or `associate` (plain "business development manager" is a partnerships role, not an SDR seat);
- `appointment setter`, `cold caller`.

The taxonomy applies its rules in order and the first match wins. Earlier rules (recruiting, revenue operations, sales engineering, marketing operations, software engineering) take 2,202 raw titles that the SDR pattern would also match, for example "sales development recruiter". A separate level rule then reads the title. We kept only individual contributor levels (`ic` and `senior_ic`), so SDR managers, team leads, directors, VPs, founders and interns are out.

**How many titles matched:**

- 113,759 distinct raw titles carry the `sdr_bdr` function at any level; 64,417 of them are at IC or senior IC level and count as SDR seats.
- Excluded by level (distinct raw titles): manager 32,475, director 10,487, VP 3,671, intern 2,088, founder 383, C-level 238.
- In headline seats there are 39,358 distinct normalized titles; 198 are held by 50 or more people, and the top 10 cover 73% of seats.

Spells by level for all `sdr_bdr` titles:

| Level | Spells | People |
|---|---|---|
| IC | 498,936 | 407,924 |
| Manager | 103,951 | 88,563 |
| Senior IC | 45,245 | 41,755 |
| Director | 22,654 | 19,761 |
| Intern | 9,475 | 9,183 |
| VP | 6,293 | 5,804 |
| Founder | 413 | 402 |
| C-level | 247 | 239 |

Headline seats by which part of the rule matched first (groups under 50 people suppressed):

| Branch | Seats | People | Distinct titles |
|---|---|---|---|
| business development representative/rep/associate | 194,776 | 178,595 | 12,240 |
| sales development | 182,340 | 160,136 | 12,985 |
| abbreviation sdr/bdr/mdr/adr | 19,473 | 17,655 | 6,367 |
| market development | 14,782 | 14,367 | 4,372 |
| account development | 14,367 | 13,874 | 2,541 |
| appointment setter | 5,477 | 4,952 | 735 |
| cold caller | 315 | 269 | 106 |
| outbound or inbound development | suppressed (under 50 people) | | |

The 15 most common normalized titles:

| Normalized title | Seats | People |
|---|---|---|
| sales development representative | 125,426 | 113,094 |
| business development representative | 117,671 | 109,180 |
| business development associate | 39,708 | 38,853 |
| account development representative | 5,309 | 5,235 |
| senior business development representative | 4,674 | 4,511 |
| sales development | 4,655 | 4,451 |
| sales development specialist | 4,622 | 4,481 |
| sdr | 4,568 | 4,296 |
| market development representative | 4,261 | 4,230 |
| appointment setter | 4,127 | 3,797 |
| senior sales development representative | 3,336 | 3,245 |
| enterprise sales development representative | 3,231 | 3,132 |
| enterprise business development representative | 3,102 | 3,017 |
| sales development associate | 2,440 | 2,425 |
| sales development executive | 1,870 | 1,812 |

## 4. Sample and seat definition

- **Seat:** one or more consecutive SDR or BDR spells (IC or senior IC) at the same company, with a gap of at most one month, merged into one seat. A move from SDR to Senior SDR at the same company is one seat. Spells with no company id are their own seat.
- **Duration:** whole months, inclusive (end month minus start month, plus one). Profiles give dates by month. The extract's own duration field is not used: it is empty for about half of ended spells.
- **Current seats** are right-censored at the person's last profile update, capped at June 2026.
- **Dropped spells:** no start date 12,415; ended with no end date 58; start after the censor month 46; end before start 46; longer than 15 years 3,380. From 544,181 IC and senior IC SDR spells, 528,236 remain. They form 481,193 seats (40,303 merged from more than one spell).
- **Headline population:** seats started 2012 to 2025: 431,550 seats, 363,028 people, 351,366 ended and 80,184 still current. By country: US 329,629, GB 43,982, CA 34,249, AU 13,604, IE 8,863, NZ 1,223 seats.
- **Company size** uses the company's current size band, merged into 1-10, 11-50, 51-200, 201-500, 501-1,000, 1,001-5,000 and 5,001+. **Software and IT** means the 13 industry labels with 50 or more people that name software, information technology, IT services, internet, computer and network security or computer networking. The largest are software development, IT services and IT consulting, and technology, information and internet; the full list is software development; IT services and IT consulting; technology, information and internet; computer and network security; technology, information and media; data security software products; internet marketplace platforms; internet publishing; computer networking products; mobile computing software products; IT system custom software development; desktop computing software products; and embedded software products.
- **Privacy:** only aggregates leave Ken's servers, and any group under 50 people is suppressed.

## 5. Methods

- **Kaplan-Meier survival** on months in the seat, with current seats censored. The median is the first month at which survival is at or below 50%. Its 95% confidence interval comes from log-log Greenwood bounds. S(k) is the share of seats lasting more than k months. Survival and medians are not reported (n/a in the tables) where fewer than 50 seats are still at risk.
- **Group contrasts:** log-rank tests for 11-200 vs 1,001+ employees and for software and IT vs other industries.
- **Exit routes:** for each ended seat, the next job on the profile is the earliest spell that starts after the seat started, no earlier than 2 months before it ended, and is not part of the seat. Ties go to the same company, then to profile order. Routes: **moved company** (next job in sales or marketing at another company); **same-company move** (next job in sales or marketing at the same company, mostly promotions); **left sales and marketing** (next job outside the taxonomy's sales and marketing functions); **no later role** (no later job listed). We report shares of ended seats and the Aalen-Johansen cumulative incidence of each route, treating the others as competing risks.
- **Ramp comparison:** median divided by ramp, and median minus ramp ("productive months at the median"), at ramp times of 3.0, 3.2 and 3.8 months.

## 6. Results

### 6.1 Headline

- Median months in the seat: **14** (95% CI 14 to 14), 431,550 seats, 363,028 people.
- Still in the seat at 3, 6, 12 and 24 months: 95.3%, 82.3%, 57.0%, 25.6%.
- Seats ending within 3 months (inside the ramp): 4.7%; within 4 months: 8.9%.

| Ramp benchmark | Ramp (months) | Median / ramp | Productive months at median | Ramp share of median |
|---|---|---|---|---|
| The Bridge Group, 2025 | 3.0 | 4.7 | 11.0 | 21.4% |
| The Bridge Group, 2023 | 3.2 | 4.4 | 10.8 | 22.9% |
| The Bridge Group, 2014 peak | 3.8 | 3.7 | 10.2 | 27.1% |

### 6.2 By company size

| Group | People | Seats | Median (months) | 95% CI low | 95% CI high | S3 | S6 | S12 | S24 |
|---|---|---|---|---|---|---|---|---|---|
| 1-10 | 13,696 | 13,993 | 13 | 12 | 13 | 87.8% | 69.3% | 50.2% | 31.7% |
| 11-50 | 39,577 | 41,007 | 12 | 12 | 12 | 93.1% | 74.9% | 49.1% | 25.0% |
| 51-200 | 69,511 | 73,612 | 13 | 12 | 13 | 94.5% | 78.7% | 50.1% | 21.2% |
| 201-500 | 46,375 | 48,195 | 13 | 13 | 14 | 95.9% | 83.3% | 55.3% | 22.7% |
| 501-1,000 | 42,248 | 43,689 | 14 | 14 | 14 | 96.7% | 85.2% | 57.1% | 21.6% |
| 1,001-5,000 | 63,937 | 66,365 | 15 | 15 | 15 | 96.8% | 87.2% | 61.0% | 24.6% |
| 5,001+ | 68,664 | 71,083 | 17 | 16 | 17 | 97.8% | 89.3% | 68.3% | 30.7% |
| unknown | 69,706 | 73,606 | 14 | 14 | 14 | 93.6% | 79.1% | 55.8% | 29.3% |

Pooled: 11-200 employees median 12 months (114,619 seats), 1,001+ employees median 16 months (137,448 seats); log-rank chi-square 2,145.3.

### 6.3 By industry

| Group | People | Seats | Median (months) | 95% CI low | 95% CI high | S3 | S6 | S12 | S24 |
|---|---|---|---|---|---|---|---|---|---|
| Software and IT | 169,359 | 194,789 | 13 | 13 | 14 | 96.3% | 84.3% | 56.2% | 19.4% |
| Other industries | 153,667 | 163,237 | 15 | 15 | 15 | 94.8% | 81.4% | 58.4% | 31.2% |
| unknown | 69,634 | 73,524 | 14 | 14 | 14 | 93.6% | 79.1% | 55.9% | 29.3% |

Log-rank chi-square for software and IT vs other industries: 4,136.1. The 15 largest industries:

| Group | People | Seats | Median (months) | 95% CI low | 95% CI high | S3 | S6 | S12 | S24 |
|---|---|---|---|---|---|---|---|---|---|
| software development | 122,513 | 137,301 | 14 | 14 | 14 | 96.6% | 85.0% | 56.8% | 18.6% |
| it services and it consulting | 26,751 | 27,390 | 14 | 14 | 14 | 96.0% | 83.3% | 57.4% | 24.0% |
| financial services | 16,881 | 17,233 | 14 | 14 | 14 | 95.7% | 82.6% | 57.4% | 28.2% |
| technology, information and internet | 14,541 | 14,765 | 12 | 12 | 13 | 94.5% | 78.9% | 49.8% | 18.5% |
| computer and network security | 10,693 | 11,306 | 14 | 14 | 14 | 96.7% | 86.8% | 59.5% | 19.8% |
| advertising services | 10,317 | 10,563 | 13 | 12 | 13 | 92.7% | 76.0% | 50.5% | 26.1% |
| business consulting and services | 9,383 | 9,538 | 13 | 13 | 13 | 94.6% | 79.3% | 51.4% | 23.2% |
| hospitals and health care | 6,028 | 6,173 | 16 | 16 | 17 | 96.8% | 85.9% | 63.5% | 33.7% |
| e-learning providers | 5,563 | 5,656 | 12 | 12 | 13 | 88.6% | 72.4% | 49.2% | 24.3% |
| telecommunications | 4,905 | 5,008 | 16 | 15 | 16 | 96.0% | 80.7% | 60.0% | 35.8% |
| transportation, logistics, supply chain and storage | 4,214 | 4,299 | 15 | 14 | 16 | 96.8% | 84.5% | 60.2% | 30.7% |
| human resources services | 4,179 | 4,247 | 14 | 13 | 14 | 96.7% | 84.0% | 57.5% | 23.6% |
| insurance | 3,977 | 4,030 | 18 | 17 | 19 | 95.5% | 85.2% | 66.1% | 40.3% |
| construction | 3,545 | 3,583 | 18 | 17 | 19 | 95.9% | 84.4% | 63.1% | 40.2% |
| information services | 3,451 | 3,492 | 12 | 11 | 12 | 96.3% | 77.0% | 47.0% | 17.8% |

### 6.4 By start year

| Group | People | Seats | Median (months) | 95% CI low | 95% CI high | S3 | S6 | S12 | S24 |
|---|---|---|---|---|---|---|---|---|---|
| 2012 | 7,524 | 7,596 | 18 | 18 | 19 | 96.7% | 87.7% | 68.3% | 38.5% |
| 2013 | 9,249 | 9,333 | 16 | 16 | 17 | 96.0% | 86.4% | 64.9% | 33.9% |
| 2014 | 11,817 | 11,924 | 15 | 14 | 15 | 95.9% | 85.1% | 61.2% | 30.5% |
| 2015 | 15,506 | 15,700 | 14 | 14 | 14 | 95.9% | 84.3% | 59.7% | 27.9% |
| 2016 | 17,831 | 18,078 | 14 | 14 | 14 | 96.1% | 83.8% | 59.1% | 25.4% |
| 2017 | 21,608 | 21,953 | 14 | 13 | 14 | 96.1% | 83.6% | 57.7% | 24.4% |
| 2018 | 26,715 | 27,320 | 14 | 13 | 14 | 96.0% | 83.3% | 57.2% | 24.4% |
| 2019 | 32,458 | 33,276 | 14 | 14 | 14 | 95.9% | 82.9% | 58.0% | 24.1% |
| 2020 | 31,743 | 32,761 | 13 | 13 | 13 | 94.5% | 81.5% | 54.8% | 20.6% |
| 2021 | 50,799 | 52,952 | 13 | 13 | 13 | 95.4% | 82.2% | 52.6% | 21.0% |
| 2022 | 57,051 | 60,069 | 14 | 13 | 14 | 94.8% | 81.1% | 55.3% | 25.2% |
| 2023 | 43,883 | 45,875 | 14 | 14 | 14 | 94.4% | 80.2% | 55.4% | 26.4% |
| 2024 | 44,631 | 46,853 | 14 | 14 | 14 | 94.9% | 81.0% | 56.1% | 29.0% |
| 2025 | 45,465 | 47,860 | n/a | n/a | n/a | 95.1% | 82.9% | 62.7% | n/a |
| 2026 | 13,073 | 13,253 | n/a | n/a | n/a | 97.1% | n/a | n/a | n/a |

The 2025 and 2026 cohorts have at most 16 and 4 months of follow-up, so medians are not reached; they are left out of cohort comparisons. Pooled cohorts:

| Cohort | Seats | Median | S12 |
|---|---|---|---|
| 2012-2015 | 44,553 | 15 | 62.7% |
| 2016-2019 | 100,627 | 14 | 57.9% |
| 2020 | 32,761 | 13 | 54.8% |
| 2021-2022 | 113,021 | 13 | 54.1% |
| 2023-2024 | 92,728 | 14 | 55.8% |

### 6.5 By country

| Group | People | Seats | Median (months) | 95% CI low | 95% CI high | S3 | S6 | S12 | S24 |
|---|---|---|---|---|---|---|---|---|---|
| AU | 11,757 | 13,604 | 14 | 14 | 14 | 95.2% | 82.5% | 57.4% | 26.4% |
| CA | 29,475 | 34,249 | 14 | 14 | 14 | 94.7% | 79.8% | 56.0% | 26.7% |
| GB | 36,730 | 43,982 | 14 | 14 | 14 | 95.0% | 82.7% | 57.8% | 25.1% |
| IE | 7,319 | 8,863 | 15 | 14 | 15 | 96.7% | 85.8% | 60.7% | 24.6% |
| NZ | 1,083 | 1,223 | 14 | 13 | 14 | 94.5% | 80.1% | 56.1% | 27.9% |
| US | 276,664 | 329,629 | 14 | 14 | 14 | 95.3% | 82.5% | 56.9% | 25.5% |

### 6.6 Exit routes

All headline seats; shares of ended seats and cumulative incidence (CIF) at 3, 12, 24 and 36 months:

| Route | People | Seats | Share of ended | CIF 3m | CIF 12m | CIF 24m | CIF 36m |
|---|---|---|---|---|---|---|---|
| Moved company | 109,315 | 131,084 | 37.3% | 1.7% | 16.2% | 27.6% | 31.4% |
| Same-company move | 96,958 | 99,214 | 28.2% | 0.7% | 10.8% | 21.7% | 24.5% |
| Left sales and marketing | 102,639 | 105,188 | 29.9% | 1.9% | 14.0% | 22.1% | 24.8% |
| No later role | 15,537 | 15,880 | 4.5% | 0.4% | 2.0% | 3.0% | 3.5% |

Share of ended seats by route, by group:

| Dimension | Group | Moved company | Same-company move | Left sales and marketing | No later role |
|---|---|---|---|---|---|
| Company size | 1,001-5,000 | 34.4% | 37.0% | 25.4% | 3.2% |
| Company size | 1-10 | 40.2% | 7.6% | 41.2% | 11.0% |
| Company size | 11-50 | 39.0% | 20.0% | 34.5% | 6.6% |
| Company size | 201-500 | 38.2% | 31.4% | 26.6% | 3.9% |
| Company size | 5,001+ | 30.6% | 38.6% | 27.9% | 3.0% |
| Company size | 501-1,000 | 38.4% | 31.9% | 26.4% | 3.4% |
| Company size | 51-200 | 39.0% | 28.1% | 28.5% | 4.4% |
| Company size | unknown | 42.1% | 14.1% | 37.6% | 6.3% |
| Industry | Other industries | 34.5% | 23.8% | 36.1% | 5.5% |
| Industry | Software and IT | 37.7% | 36.9% | 22.3% | 3.1% |
| Industry | unknown | 42.1% | 14.1% | 37.6% | 6.3% |

Seats with no company record cannot be classed as same-company moves, so that route is undercounted in the "unknown" group. Among ended seats with a known company (291,539): moved company 36.3%, same-company move 31.1%, left sales and marketing 28.4%, no later role 4.2%.

What the next job is:

| Exit | Next job | Seats | People | Share within route |
|---|---|---|---|---|
| Moved company | Other sales or marketing | 61,891 | 60,139 | 47.2% |
| Moved company | SDR or BDR | 45,624 | 36,513 | 34.8% |
| Moved company | Account executive | 21,295 | 20,836 | 16.2% |
| Moved company | SDR or BDR lead or manager | 2,274 | 2,219 | 1.7% |
| Same-company move | Other sales or marketing | 48,969 | 48,524 | 49.4% |
| Same-company move | Account executive | 40,976 | 40,530 | 41.3% |
| Same-company move | SDR or BDR lead or manager | 8,796 | 8,707 | 8.9% |
| Same-company move | SDR or BDR | 473 | 469 | 0.5% |

17.7% of ended seats go straight to an account executive title, and 13.0% go to another SDR or BDR seat at another company.

## 7. Short jobs left off profiles

People leave short jobs off their profiles, and that makes tenure look longer. We pre-registered two estimates of the effect.

**Method 1: curve extrapolation (not used).** We fitted a Weibull and a lognormal curve to conditional survival from months 7 to 60, extrapolated it back to months 1 to 6, and read the shortfall as missing short seats:

| Model | Observed S6 | Model S6 | Implied missing share of true seats | Observed median | Corrected median |
|---|---|---|---|---|---|
| Weibull | 82.3% | 10.8% | 86.9% | 14 | 1 |
| Lognormal | 82.3% | 36.0% | 56.2% | 14 | 4 |

The two curves give corrected medians of 1 and 4 months, so the answer is set by the curve chosen, not by the data. We do not use this method.

**Method 2: SDR jobs missing at the moment of a move (used).** We took the 5,325 people detected in June 2026 moving from an IC or senior IC SDR title at one company to a job at another. For each, we checked their profile in the same June snapshot. In every case the new job was on the profile.

- The SDR job is still listed: 4,100 (77.0%).
- The SDR employer is listed, but under a different, non-SDR title: 351 (6.6%).
- No entry at the SDR employer at all: 874 (16.4%). Of these, 60 show an SDR title with no company id, so the strict share is 15.3%.
- Listed SDR jobs of these leavers have a median of 11 months; 22.9% are 6 months or shorter.

For the correction, we assume each missing job lasted 1 to 6 months. We add p/(1-p) times the number of ended headline seats as extra ended seats, spread over months 1 to 6 like the observed short exits, where p is the missing share:

| Scenario | Missing share | Added short seats | Corrected median | Corrected S3 | Corrected S6 | Corrected S12 |
|---|---|---|---|---|---|---|
| Low: no entry at the employer, excluding SDR titles with no company id | 15.3% | 63,422 | 12 | 92.4% | 71.8% | 49.7% |
| Main: no entry at the employer | 16.4% | 68,979 | 12 | 92.2% | 71.0% | 49.1% |
| High: no entry at the employer, or listed under a non-SDR title | 23.0% | 104,953 | 11 | 90.9% | 66.2% | 45.8% |

The correction lowers the median from 14 to 12 months (range 11 to 12). It is an estimate, not a fix, for three reasons. Some missing jobs were longer than 6 months, which would make the true effect smaller. Some short jobs never appear anywhere, even as a job change, which would make it larger. And June 2026 job changes may not represent all years.

## 8. Sensitivity

| Run | Seats | Median | S12 |
|---|---|---|---|
| Headline, 2012-2025 starts | 431,550 | 14 | 57.0% |
| 2012-2024 starts only | 383,690 | 14 | 56.5% |
| Current seats censored at June 2026 | 431,550 | 14 | 57.5% |
| Spells not merged into seats | 476,267 | 13 | 50.6% |
| Cohort 2012-2015 | 44,553 | 15 | 62.7% |
| Cohort 2016-2019 | 100,627 | 14 | 57.9% |
| Cohort 2020 | 32,761 | 13 | 54.8% |
| Cohort 2021-2022 | 113,021 | 13 | 54.1% |
| Cohort 2023-2024 | 92,728 | 14 | 55.8% |
| Core titles only | 410,955 | 14 | 56.5% |

The headline median is 14 months when we use only 2012-2024 starts, censor current seats at June 2026 instead of the last profile update, or drop the looser title branches (market development, appointment setter, cold caller). Counting each spell separately, without merging SDR to Senior SDR, gives 13 months, because a promotion inside the seat then counts as an exit.

## 9. Hypotheses vs results

| # | Pre-registered expectation | Result | Verdict |
|---|---|---|---|
| 1 | Median 14-20 months; productive months 11-17; ramp 15%-21% of median | Median 14; productive 11.0; ramp 21.4% of median | Held, at the low edge (ramp share just above the range) |
| 2 | S(3) at least 90%, S(12) under 70% | S(3) 95.3%, S(12) 57.0% | Held |
| 3 | Moved company largest (35%-55%), same-company second (25%-40%), left sales smallest (10%-30%) | 37.3%, 28.2%, 29.9% | Partly: moved company is largest and every share is in range, but leaving sales is second and same-company moves third |
| 4 | 1,001+ at least 2 months longer than 11-200 | 16 vs 12 | Held |
| 5 | 2021-2022 lowest 12-month survival; 2023-2024 higher | 2021 lowest single year (52.6%); 2020 (54.8%) is below 2022 (55.3%); pooled 2023-2024 above 2021-2022 | Partly: fails as stated, because 2020 is below 2022 |
| 6 | Software and IT shorter than other industries | 13 vs 15 months | Held |
| 7 | Short-job correction lowers median by 1-3 months | Method 2: by 2 (range 2 to 3); method 1 not usable | Held for method 2 |

## 10. Verification

An independent re-analysis, written from the pre-registered plan, the decisions and the extract paths alone, without this study's code, recomputed the headline numbers. Every number matched to the reported precision: seats, people, median, S(3), S(12), S(24), size and industry medians, missing share and corrected median. The first comparison found the exit-route shares off by up to 0.3 percentage points. The cause was a bug here: a check that returned "unknown" when a company id was missing dropped some later SDR jobs as next-job candidates. After the fix, all four shares match to 4 decimals. One tied seat still differs between the two codes (99,214 vs 99,215 same-company moves), which is below the reported precision.

## 11. Alternative explanations and limitations

- **Short jobs left off profiles** make tenure look longer. We estimate the effect at 2 to 3 months on the median (section 7), but cannot measure it exactly.
- **Survivorship.** Only active profiles exist; deleted profiles drop out. Older profiles have had more time to drop early short jobs. Older cohorts therefore look more stable, and part of the decline in 12-month survival since 2012 may come from the data, not from behaviour.
- **Late updates.** People add new jobs months late. Recent ended seats may still show as current, which makes recent cohorts look longer; that is why 2025 and 2026 are out of the cohort comparisons.
- **Current company attributes.** Size and industry are today's values, not the values when the SDR worked there. A company that grew from 50 to 2,000 people is counted as 1,001-5,000.
- **Selection by company size.** Large companies may hire SDRs with more experience or run longer programs. The size gap shows a link, not a cause.
- **Exit classes come from the next job listed.** A same-company move includes both promotions and sideways moves. "Left sales and marketing" includes career breaks recorded as other jobs, founding a company and going back to study. "No later role" mixes unemployment, unlisted jobs and people who stopped updating.
- **Titles.** The shared rule catches a few non-seats, such as "sales development student" and "sales development program" (about 1,500 seats, 0.35%). Market development titles include some non-SDR jobs in pharma and consumer goods. Dropping the looser branches leaves the median unchanged (section 8). English titles are classified best; non-English SDR titles are mostly missed.
- **Coverage.** Six English-speaking countries, 76% of seats in the US. Results may not hold elsewhere.
- **Ramp** comes from company surveys (The Bridge Group), not from the same people, so the comparison is between a typical ramp and a typical stay, not each SDR's own ramp.

## 12. Sources

- Ken internal data: Ken's dataset of public professional profiles, work histories and company records; June 2026 snapshot; six countries.
- The Bridge Group, Sales Development Models, Motions and Metrics: 2025 Research Report. https://www.bridgegroupinc.com/research/2025-sdr-models-metrics-report-the-bridge-group
- The Bridge Group, 2023 SDR Metrics: Sales Development Models, Metrics and Compensation Research. https://blog.bridgegroupinc.com/hubfs/resources/TBG%202023%20Sales%20Development%20Report.pdf
- Shared sales careers extract: built 2026-10-15 for the Ken Research study *Who are GTM engineers, and where do they come from?*