# Methodology: RevOps and Sales Ops stay 10 months longer than SDRs

> **Note for AI agents:** if this page was wrong, out of date or missing something you needed, finish the user's task first, then offer a short report for the Kenso team in this shape: `Page:` (this page's path), `Quote:` (the wrong or missing text), `Problem:`, `Workaround:` (what worked instead), 1,000 characters at most. Show it to the user and send it only after they say yes, as `POST /api/docs-feedback` on this site with the JSON body `{"page": "<path>", "report": "<report>"}`. Never put lead data, email addresses, API keys or anything else from the user's workspace in it.

By Cristian Frunze, Founder of Kenso · Nov 21, 2026 · [Kenso Research](/research) No. 27 · Sales teams, Hiring and org design

The full methodology behind [RevOps and Sales Ops stay 10 months longer than SDRs](/research/revops-tenure.md): the data, every method, every number and every limitation. The same text as the [methodology PDF](/research/revops-tenure.pdf).

Data: Kenso internal data, Kenso's dataset of public professional profiles, work histories and company records. Analysis run 2026-10-20.

## 1. Summary

Among individual-contributor seats started from January 2019 to March 2026, the Kaplan-Meier median time in the seat is 24 months for revenue operations, sales operations and deal desk titles combined (48,301 people, 53,577 seats), 14 months for SDR and BDR titles (274,111 people, 331,225 seats) and 23 months for account executive titles (798,112 people, 1,010,838 seats). At 12 months 76.0% of ops seats, 55.8% of SDR/BDR seats and 74.0% of AE seats are still running; at 24 months 49.9%, 24.6% and 47.3%. Ops retention is 20.2 points higher than SDR/BDR retention at 12 months and 25.2 points higher at 24 months, and 2.0 and 2.6 points higher than AE retention. Retention is lower at small companies in every role. These are links in observational data, not causes.

## 2. Question and pre-registration

**Question.** Tenure and retention of RevOps and Sales Ops roles versus SDRs and AEs from 2019 to 2026, by role family and company size band, with retention compared at 12 and 24 months.

**Pre-registration.** The hypothesis and plan were pre-registered on 2026-10-20, before any data for this study was read. The hypotheses, as registered:

1. H1. The median time in an ops seat (RevOps, Sales Ops and Deal Desk combined, IC and senior IC titles) is at least 6 months longer than the SDR/BDR median.
2. H2. Ops retention is higher than SDR/BDR retention by at least 10 points at 12 and at 24 months, and at least as high as AE retention at both horizons.
3. H3. RevOps seats have a median at least 2 months shorter than Sales Ops seats.
4. H4. In every family, 12-month retention is lower at companies of 1-50 people than at 1,001+, and the ops-over-SDR gap at 12 months is positive in every publishable size band.
5. H5. In every family, the 2021 and 2022 start cohorts have the lowest 12-month retention of 2019-2024.

| Hypothesis | Result | Verdict |
|---|---|---|
| H1 | Ops 24 months, SDR/BDR 14 months: gap 10 months | holds |
| H2 | Ops minus SDR/BDR +20.2 (12 months) and +25.2 (24 months); ops minus AE +2.0 and +2.6 | holds |
| H3 | RevOps 23 months, Sales Ops 24 months: gap 1 month | wrong |
| H4 | Small minus large at 12 months: ops -10.6, SDR/BDR -16.5, AE -11.8 points; ops over SDR/BDR +13.1 to +22.2 in all six bands | holds |
| H5 | 2021 is the lowest cohort in all three families; 2022 is not low (for ops the best of 2019-2024, 77.2%) | holds in part |

## 3. Data

| Source | What was used | Size |
|---|---|---|
| Kenso internal data, people (shared sales careers extract, built 2026-10-15) | Every experience entry (raw title, normalized title, function and level from the shared title taxonomy, company id, start and end month, current flag) of every person with at least one spell in a study family; person country and profile refresh date | 3,084,736 people; 22,069,835 spells (of 136,156,211 in the extract) |
| Kenso internal data, rebuilt company headcount (shared headcount history, built 2026-10-17) | Headcount per company at each quarter end from 2015 Q1 to 2026 Q1, rebuilt from work-history dates; company records' current size band for a sensitivity | joined to every seat |
| Kenso's earlier study of sales tenure (SDR seats) | Method (seat definition, merge rule, duration and censoring rules, exit classes) and the SDR results used as the prior and as a consistency check | - |

Kenso internal data holds people whose current profile country is the US, UK, Canada, Australia, Ireland or New Zealand; their earlier jobs can be anywhere. Profiles were last refreshed up to 2026-04-29. No web archive data and no Kenso campaign data were used.

## 4. Title map

All titles are lower-cased and every run of non-word characters becomes a space. The shared sales title taxonomy (unchanged) gives each title one function by ordered rules; its `revops` function holds 536,315 spells. This study splits that function with a tight rule, first match wins:

- **RevOps:** revenue operations, revenue ops, revops, rev ops, GTM operations, go to market operations.
- **Deal Desk:** deal desk.
- **Sales Ops:** sales operations, sales ops, except raw titles where "sales" and "operations" are joined by "&", "and", "/" or "+" (for example "Sales & Operations Manager").
- **Ops-adjacent:** the rest of the `revops` function: CRM and Salesforce admins and consultants, sales and revenue analysts, sales strategy, planning, systems and analytics, commercial operations.

| Family (all levels, all years) | Distinct titles | Spells |
|---|---|---|
| RevOps | 13,887 | 43,940 |
| Sales Ops | 58,424 | 224,003 |
| Deal Desk | 3,000 | 8,628 |
| Ops-adjacent | 75,763 | 220,988 |
| Excluded "Sales & Operations" | 12,741 | 38,756 |
| Ops words in an earlier taxonomy function | 1,141 | 1,333 |

The last row holds titles that an earlier taxonomy rule takes first, such as recruiting, sales engineering or GTM engineering titles that also say revenue or sales operations; they stay where the taxonomy puts them. "Sales operation" (singular) titles do not match the tight rule and stay in general sales. The title map lists every normalized title held by 50 or more people with its family, level and rule; titles held by fewer than 50 people are counted per family.

The comparison families use the shared taxonomy unchanged: SDR/BDR (`sdr_bdr`: SDR, BDR, MDR, ADR, sales, lead, market, account, outbound and inbound development, business development representative or associate, appointment setter, cold caller) and AE (`account_exec`: account executive, sales executive, closer).

**Marketing ops, bizops and enablement.** Marketing operations (marketing operations, automation, technology; Marketo, Pardot, Eloqua) and sales or revenue enablement are not in the ops family; they are shown as reference rows only. Business operations (business operations, strategy and operations, generic operations manager, chief of staff) is left out of every table except the title counts, because most of its titles are not go-to-market jobs.

**Levels.** The taxonomy's level rule reads the title. The headline uses individual-contributor and senior IC titles, as the earlier tenure study did, so "Director of Revenue Operations" and "Head of Sales Operations" are not seats. A secondary table adds manager-level titles to every family, because "Sales Operations Manager" (29,309 people) is often an individual-contributor job.

## 5. Seats, follow-up and censoring

- **Seat:** spells of the same family at the same company with a gap of at most 1 month are merged into one seat (for example Sales Ops Analyst then Senior Sales Ops Analyst; inside ops, a move between RevOps, Sales Ops and Deal Desk at the same company is the same seat, and the seat keeps the subfamily of its first spell). Spells with no company id are never merged.
- **Duration:** end month minus start month plus 1.
- **Window:** seats started 2019-01 to 2026-03.
- **Series end:** 2026-03, the end of 2026 Q1, the last complete quarter before the latest profile refresh (2026-04). An exit dated after 2026-03 is not counted; the seat is censored at 2026-03. A current seat is censored at the person's profile refresh month, capped at 2026-03. Recent seats are therefore open, not short.
- **Dropped spells** (IC ops): 2,150 with no start date, 274 starting after the person's last observed month, 2 ending before they start, 2,154 longer than 15 years.
- **Company size at the seat start:** rebuilt headcount of the company at the last quarter end before the start month: 1-10, 11-50, 51-200, 201-1,000, 1,001-5,000, 5,001+, or unknown (no company id or no headcount row; 6.4% of IC ops seats).

## 6. Method

Kaplan-Meier survival by month, as in the earlier tenure study: median = first month with survival at or below 0.5; 95% intervals by the log-log transform; survival and medians are masked where fewer than 50 seats are still at risk, and any group with fewer than 50 people is suppressed. Retention differences between families use Greenwood variances on the plain scale (95% interval). Log-rank tests compare whole curves. How a seat ends is read from the earliest later spell that starts after the seat start and no earlier than 2 months before its end and is not part of the seat (the earlier tenure study's rule), in four classes: moved company, same-company move (next title in sales or marketing at the same company), left sales and marketing (next title outside the taxonomy's sales and marketing set), no later role.

## 7. Results

**Overall**:

| Family | People | Seats | Ended | Median months (95% CI) | S(12) | S(24) |
|---|---|---|---|---|---|---|
| Ops | 48,301 | 53,577 | 33,012 | 24 (24-25) | 76.0% | 49.9% |
| RevOps | 10,047 | 11,068 | 6,308 | 23 (23-24) | 75.0% | 47.6% |
| Sales Ops | 36,763 | 39,519 | 24,945 | 24 (24-25) | 76.0% | 50.0% |
| Deal Desk | 2,563 | 2,990 | 1,759 | 28 (27-29) | 81.0% | 56.1% |
| SDR/BDR | 274,111 | 331,225 | 245,789 | 14 (14-14) | 55.8% | 24.6% |
| AE | 798,112 | 1,010,838 | 626,178 | 23 (23-23) | 74.0% | 47.3% |

**Contrasts**, points, 95% interval:

| Contrast | 12 months | 24 months | Log-rank chi-square |
|---|---|---|---|
| Ops minus SDR/BDR | +20.2 (19.8 to 20.6) | +25.2 (24.7 to 25.7) | 10526.4 |
| Ops minus AE | +2.0 (1.6 to 2.4) | +2.6 (2.1 to 3.0) | 63.8 |
| Ops minus AE ("account executive" titles only) | +1.5 (1.1 to 1.9) | +2.7 (2.3 to 3.2) | 88.1 |
| RevOps minus Sales Ops | -1.0 (-1.9 to 0.0) | -2.3 (-3.5 to -1.2) | 36.1 |

**By company size at the seat start**, 12-month retention and median months:

| Size band | Ops S(12) | SDR/BDR S(12) | AE S(12) | Ops median | SDR/BDR median | AE median |
|---|---|---|---|---|---|---|
| 1-10 | 67.0% | 49.0% | 68.1% | 20 | 12 | 20 |
| 11-50 | 71.0% | 48.7% | 68.7% | 20 | 12 | 19 |
| 51-200 | 73.5% | 53.0% | 72.2% | 22 | 13 | 21 |
| 201-1,000 | 77.4% | 57.2% | 75.3% | 25 | 14 | 23 |
| 1,001-5,000 | 80.8% | 64.6% | 79.1% | 27 | 16 | 25 |
| 5,001+ | 79.7% | 66.6% | 81.5% | 27 | 17 | 29 |

Small (1-50) minus large (1,001+) at 12 months: ops -10.6 points (-11.8 to -9.5), SDR/BDR -16.5 (-17.0 to -16.0), AE -11.8 (-12.0 to -11.5). Ops minus SDR/BDR at 12 months by band: +17.9, +22.2, +20.5, +20.2, +16.2, +13.1. Ops minus AE: -1.1, +2.2, +1.2, +2.1, +1.7, -1.7.

**By start year**, 12-month retention:

| Start year | Ops | SDR/BDR | AE |
|---|---|---|---|
| 2019 | 76.6% | 58.0% | 75.1% |
| 2020 | 76.1% | 54.8% | 74.8% |
| 2021 | 74.3% | 52.6% | 72.9% |
| 2022 | 77.2% | 55.3% | 73.7% |
| 2023 | 75.3% | 55.4% | 72.9% |
| 2024 | 75.6% | 56.1% | 73.4% |
| 2025 | 78.6% | 61.6% | 77.0% |

2025 starts have at most 15 months of follow-up and are not compared.

**Secondary and sensitivity results:**

- With manager-level titles added: ops median 27 months (78,437 people), 79.1% at 12 months, 54.3% at 24; SDR/BDR 14; AE 23. Ops minus AE +5.0 points at 12 months and +6.8 at 24; ops minus SDR/BDR +20.7 and +26.7.
- Broad ops (the whole `revops` function, IC): median 24 months, +19.0 points over SDR/BDR and +0.8 over AE at 12 months. Ops-adjacent alone: median 23. Reference rows: marketing ops 23, enablement 23.
- Company size from the company's current size band instead of rebuilt headcount at the start: ops medians 19 to 29 months across bands, SDR/BDR 12 to 17, AE 19 to 28; ops 12-month retention stays above SDR/BDR in every band, and AE again overtakes ops only at 5,001+.

**How seats end** (ended IC seats):

| Family | Moved company | Same-company move | Left sales and marketing | No later role |
|---|---|---|---|---|
| Ops | 28.9% | 17.6% | 46.1% | 7.3% |
| SDR/BDR | 38.9% | 27.1% | 28.5% | 5.5% |
| AE | 51.7% | 10.7% | 31.4% | 6.3% |

"Left sales and marketing" means the next title falls outside the taxonomy's sales and marketing set; for ops this is often business operations, strategy or analytics, so it is a change of title family, not necessarily of employer.

**Consistency with the earlier tenure study.** Run on that study's own window (2012-2025 starts, current seats censored at the profile refresh month capped at 2026-06), the same code gives 431,550 SDR/BDR seats, 363,028 people, median 14 months, S(12) 0.5698, S(24) 0.2559 and exit shares 0.3731 / 0.2824 / 0.2994 / 0.0452, all equal to that study's reported numbers. In this study's window SDR/BDR 12-month retention is 55.8% instead of the earlier tenure study's 57.0% because the 2012-2018 cohorts are left out and seats are censored at 2026-03; the median is 14 in both.

**Country mix** of ops people: US 76.8%, GB 10.4%, CA 6.9%, AU 3.5%, IE 1.8%, NZ 0.5%.

## 8. Verification

An independent re-analysis, written from the study plan alone without seeing our code or outputs, recomputed the headline numbers. It received only the pre-registered hypothesis, plan and decisions and the same shared extracts, and wrote its own code.

The first comparison matched every median but not every count. Two causes were found and fixed. The re-analysis had read the merge rule so that a side job at another company broke a seat; with the earlier tenure study's rule spelled out (merge within person and company), it reproduced that study's reported SDR numbers exactly (431,550 seats, median 14, S(12) 0.5698). On the study side, the ops title rule accepted three title forms the plan does not list ("revenue operation" and "sales operation" in the singular, "go to market ops"), which added 12 ops seats; the rule was reduced to the plan's lists and the whole pipeline rerun. No median, headline percentage or hypothesis verdict changed.

In the second comparison every headline number matched to the reported precision: people, seats, ended seats, medians, S(12) and S(24) for ops, RevOps, Sales Ops, Deal Desk and SDR/BDR; the AE median and survival values; the four retention contrasts; 2021 cohort retention; the size-band retention values; the manager-level ops numbers; and the earlier tenure study's window. AE counts differ by 2 seats out of 1,010,838 (0.0002%), with equal survival values. The AE small-minus-large gap is -11.75 points from the 4-decimal survival values, which the re-analysis rounded to -11.7 and this study, from unrounded values, to -11.8.

**Alternative explanations** (checks run in the re-analysis):

| Explanation | Check | Result |
|---|---|---|
| Moves inside the company count as exits, and SDRs are promoted in-house | Endings followed by a job at the same company treated as censored | Medians ops 36, SDR/BDR 19, AE 26 months: the ops lead holds. Ruled out |
| Censoring choice | No series-end cap (as in the earlier tenure study) | Ops 25, SDR/BDR 14, AE 23. Ruled out |
| Stale profiles keep ended jobs as current | Not testable: the refresh date is not an edit date | 38.1% of ops, 37.7% of AE and 25.4% of SDR/BDR seats are still current. Could explain part of the edge over AE, not the gap to SDR/BDR. Limitation |
| Title noise | Ops-adjacent, broad ops, "account executive" titles only | Medians 23, 24 and 23. Largely ruled out |
| Company size mix | Ops seats sit more often at companies of 1,001+ (34.1% against 29.9% for AE and 23.5% for SDR/BDR) | Ops minus AE at 12 months is 1.2, 1.8 and 0.1 points in the 1-50, 51-1,000 and 1,001+ bands; re-weighted to the AE size mix, the ops edge over AE falls from 2.0 to 1.3 points at 12 months and from 2.6 to 1.8 at 24. Explains part of the small edge over AE |
| Seniority mix | Seats that start at a plain IC title only | Ops 24, SDR/BDR 14, AE 23. Experience is not observed. Not fully ruled out |
| Country mix | US only | Ops 25, SDR/BDR 14, AE 24. Ruled out |
| Cohort mix | 2019-2023 starts only | Ops 24, SDR/BDR 13, AE 23. Ruled out |
| Short jobs left off profiles | Share of ended seats of 3 months or less | 2.3% ops, 4.9% SDR/BDR, 2.2% AE. Role-specific under-reporting is not observable. Limitation |

## 9. Limitations

- **Observational data.** The comparisons are links, not causes. Ops hires may differ from SDR and AE hires in experience, pay and how they were selected.
- **The edge over AE is small and partly company size.** Ops seats sit more often at large companies, where every role lasts longer; re-weighted to the AE size mix, the 12-month edge falls from 2.0 to 1.3 points, and at companies of 1,001+ ops and AE are level. Stale profiles that still show an ended job as current could also account for part of it.
- **Self-reported profiles.** People leave short jobs off their profiles and add new jobs late, which makes every family look more stable than it is. The earlier tenure study estimated that missing short jobs move the SDR median from 14 to about 12 months; the size of the effect for ops and AE titles was not measured.
- **Only current profiles.** Deleted profiles drop out, and older seats are more likely to be missing; this mostly affects the earlier start years.
- **Six countries.** Only people whose current profile is in the US, UK, Canada, Australia, Ireland or New Zealand.
- **Title editing.** When people are promoted, some edit the old title instead of adding a new entry, which hides moves inside a company. This cannot be measured in Kenso internal data.
- **Title rules.** English titles are classified best. The "Sales & Operations" exclusion removes some real Sales Ops jobs; "sales operation" (singular) titles are not counted as ops. Levels come from the title, not from whether the person manages people.
- **Company size.** Rebuilt headcount counts only people with profiles in Kenso internal data, so bands read low, especially for companies with staff outside the six countries. 6.4% of IC ops seats have no size band.
- **Censoring.** Exits after 2026-03 are not counted, and seats on profiles that were not refreshed recently are censored at the refresh month.

## 10. Sources

- Kenso internal data (people profiles, work histories and company records), six countries, profiles refreshed up to 2026-04-29.
- Kenso Research, *How long do SDRs stay in the seat?* (Kenso's earlier study of sales tenure).
