How DoorDash Pays

Technical ladder — E3 through E7

Engineering management ladder — M1 and M2

Disclosed executive layer

Track divergence

Hierarchy qualifications and legacy structures

• The E and M codes come from two DoorDash engineering blog posts that define scope and impact, not time served, not salary and not headcount. There is no published companywide grade catalogue that covers non-engineering roles.
• Three job architectures coexist. DoorDash uses E, M and D codes; Wolt, acquired in 2022 and the source of the Helsinki, Berlin and Tokyo evidence, retains its own structure; and Deliveroo, acquired on 2 October 2025, retains a London-centred legacy structure. No public document confirms a mapping between them.
• The commercial and operations population is mapped by title alone and is deliberately kept out of the band table. Its posted ranges are base salary, sometimes plus on-target commission, and never include an equity value, so scaling them by a city factor calibrated on equity-bearing engineering packages would produce cells where total compensation sits below base. Those ranges are carried in the variable pay and career sections instead.
• The frontline and support population, including Customer Experience agents and DashMart associates, is deliberately excluded from the band table. No public pay data exists for it, and the SEC median-employee figure of $36,373 suggests it dominates the global headcount.
• Dashers are independent contractors, not employees. Their earnings, guarantees and incentives sit outside the employment relationship these bands describe and are excluded from every figure on this page, including the median-employee population used for the pay ratio.
• Executive rows are carried in the executive section rather than the band table, so that no single worldwide chief-executive figure drags the location factors toward parity at the top of the ladder.

Peer-level mapping

Office and market catalogue — calibration factors versus San Francisco

London and Auckland are verified DoorDash-group corporate locations but are absent from this table because no defensible band-by-band pay evidence exists for either. London is primarily the Deliveroo legacy organisation acquired on 2 October 2025, with harmonisation of bands, benefits and equity not publicly disclosed; Auckland is verified through current DoorDash Technologies New Zealand postings that carry no pay ranges. Inventing a factor for either would have been worse than leaving them out.

San Francisco anchor medians — the basis of every modeled cell

• San Francisco is the anchor. The engineering rows are Levels.fyi national medians multiplied by the bundle's own 1.04 San Francisco location factor, so a factor of 1.0 here means San Francisco, not the United States average.
• The seven United States markets carry a single factor applied to both base and total, because the bundle calibrates them on one city multiplier: San Francisco and New York at 1.04, Sunnyvale 1.05, Seattle 1.01, Los Angeles 0.96, Chicago 0.91 and Phoenix or Tempe 0.86 against the national median.
• The band table is the engineering and engineering-management ladder only. That is the population the city factors are calibrated on, and mixing the base-only commercial and director postings into it would misstate every non-United States market.
• India carries separate base and total factors because base compresses far less against the anchor than total compensation does: United States packages carry much larger equity components. The Hyderabad column sits about 15 percent above the India national column on the strength of a Hyderabad E5 city submission.
• Australia, Canada, Finland, Germany and Japan all show total compensation compressing faster than base as level rises, for the same reason in reverse: the United States equity component grows much faster up the ladder than the local one does.

Executive incentive targets — percent of salary

• The 23.7 percent figure is a capacity ratio, calculated as awards outstanding divided by awards outstanding plus shares available. It measures plan headroom, not how much of the pool employees actually hold.
• Total awards outstanding across the approved and inducement plans were 26,449,344 shares at 2025 year-end, against 85,030,374 available. Roughly 9.34m of the outstanding total is the chief executive's single 2020 performance award.
• Unvested restricted stock units stood at 23.961m shares with an intrinsic value of $5.427bn at the filing's year-end price. That value is price-sensitive and not all of it is near-term realisable.
• Unrecognised restricted stock unit compensation was $1.900bn at 2025 year-end, expected to be recognised over 2.31 years, and is subject to forfeitures.
• A large share reserve is legal and accounting capacity, not a promise of individual grant size. Award decisions rest with the compensation committee and the plan administrator under dilution limits and hiring needs.

2025 year-end equity position

Vesting — common reported employee schedule

Four years is the general shape, not a published policy for every award. Submitted employee offers describe several profiles, including even quarterly vesting and front-loaded 40/30/20/10 economics, and those are individual offer reports rather than a corporate schedule. Legacy options under the 2014 plan generally vested over four years with terms of up to ten years, which is why fully vested options from that era are still being exercised in 2026.

Eligibility by hierarchy level

• There is no published minimum level for equity eligibility. Stock appears in submitted United States E3 packages and in India and Australia E4 and E5 packages, while some Wolt roles carry none.
• Director-level postings mention an equity opportunity without quantifying it, so the largest component of a director package is the one number DoorDash never publishes.
• The inducement plan proves that special new-hire equity exists outside the approved pool, but nothing in the record quantifies a systematic lateral-hire premium against internal promotees at the same level.
• Frontline and support roles have no public equity eligibility rule at all. Eligibility is role, entity, location and offer dependent, and the Wolt and Deliveroo entities have not been shown to be harmonised with DoorDash practice.
• Non-employee directors receive a $300,000 initial restricted stock unit grant and a $300,000 annual grant from 2026, inside a $750,000 normal annual limit and a $1,000,000 first-year limit.

Indicative annual grant value by band — San Francisco

Annualized planning value (±25%), not the face value of every new-hire grant. Share equivalents use the $237 reference price at 26 August 2026 market snapshot and ignore plan valuation rules and PSU performance. Highlighted rows are disclosed grant-date stock awards.

Named executive target equity

The named-executive stock grants for 2025 were $15.26m for Prabir Adarkar, $13.22m for Ravi Inukonda, $6.61m for Tia Sherringham and $4.07m for Keith Yandell, all vesting quarterly over 16 quarters. Vested-equity value during 2025 was separately disclosed at $27.96m for Adarkar, $26.01m for Inukonda and $16.74m for Sherringham, and option-exercise value at $76.74m for Adarkar and $2.05m for Inukonda. Those realisation figures are not part of the Summary Compensation Table and should not be added to the grant values.

Board compensation framework

The board package is cash-light and equity-heavy: a $60,000 retainer with no meeting fees against a $300,000 annual restricted stock unit grant means roughly 83 percent of an ordinary director's pay is stock. Committee chair premiums are modest by large-cap standards, and the $5,000 nominating and governance premium is the smallest disclosed element of the whole policy.

Not publicly disclosed

Modeled promotion planning timeline

Equity timing is the one part of progression with a documented shape: service-based units generally vest over four years and named-executive units in 16 quarterly instalments, and the Australian careers page describes quarterly vesting for employees too. That quarterly cadence matters more than the review calendar, because at senior levels the refresh grant, not the merit increase, is what changes a package.

Dataset summary

City-level H-1B wage history

All-years aggregates. P25/P75 are rounded reconstruction values marked modeled; medians and record counts are the stronger fields.

Selected title / worksite records

Evidence classification

Explicit “Not publicly disclosed” index

Known gaps and diligence before relying on a cell

1. 1Executive rows are deliberately absent from the band table. A single worldwide chief-executive or named-officer figure interpolated across locations would force every offshore factor toward parity at the top of the ladder and produce absurd cells, so those figures are carried in the executive section only and the band ladder stops at Director of Engineering.
2. 2The frontline and support population, including Customer Experience agents and DashMart associates, is excluded from the bands. It moves in the opposite direction to the professional ladder across geographies and has no public pay data at all; the $36,373 disclosed median employee is the only figure that describes it.
3. 3Dashers are excluded entirely, as independent contractors rather than employees.
4. 4Only one research bundle exists for DoorDash, so there is no second source to reconcile against. Where the bundle itself carried conflicts they are preserved: FY2025 H-1B volume is 514 records at a $176,067 median in one dataset and 544 filings at a $207,205 average in another, and the chief executive's 2020 award is worth about $2.115bn at the filing's year-end price against about $2.21bn at this report's $237 reference price.
5. 5The M2 anchor total was rebuilt from its components. The submitted national median was published as $724.0K while base, stock and bonus sum to $723.5K; the components control here so that the row reconciles.
6. 6The commercial and director postings are excluded from the band table for a structural reason, not a sourcing one. Their disclosed pay is base, sometimes plus on-target commission, with equity unquantified, so their total-to-base ratio is close to one. A city factor pair calibrated on equity-bearing engineering packages carries a total factor well below its base factor in Australia, Japan and the Wolt markets, which would have rendered a Sydney or Tokyo director with a total below their own base. The posted midpoints are therefore carried in the career, variable pay and nuance sections instead, and the band table keeps to the seven engineering and engineering-management levels the factors actually fit.
7. 7That exclusion costs the report band count. Seven rows is fewer than the ten to eighteen these reports normally carry, and it is a direct consequence of DoorDash publishing level codes for engineering only.
8. 8No employer benefit load, superannuation, provident fund or payroll tax is included in any total. Several markets in the bundle would look richer if it were, but that is employer cost rather than pay.
9. 9India, Australia, Canada and the Wolt markets are evidenced at only two or three levels each. Their level-dependent factors are fitted to those observations and then extrapolated conservatively rather than by straight-line projection, which in the India case would have implied a junior total factor near 0.06 and in the Australian case a senior base factor near 0.12. Both would have been artefacts of a two-point sample. The residual misfits worth knowing are the Bangalore E4 total, which the curve puts about 10 percent above the submitted ₹54.00L, the Toronto E4 total, about 18 percent above the submitted C$204.0K because that single Canadian observation sits oddly low against both E3 and E5, and the Berlin engineering-manager total, about 11 percent above the Wolt aggregate.
10. 10Bangalore is retained as an India-market benchmark column at the India national medians because the research brief asked for it, but no major DoorDash corporate office there was verified. Pune, which the brief also listed, is omitted. Singapore and Dubai are omitted rather than shown with an invented factor.
11. 11London and Auckland are verified DoorDash-group locations that carry no band-by-band pay evidence and are therefore absent from the location table. London in particular is the Deliveroo legacy organisation, and treating it as a DoorDash market would be wrong on the current record.
12. 12The 23.7 percent equity figure is a capacity ratio, not a utilisation rate. It cannot be compared with an ESOP utilisation percentage from a company that reports one.

FX rates used — 1 USD equals, snapshot 2026-08-26

Single-date conversion layer for comparability; it ignores payroll-date FX, tax, purchasing power, benefits valuation and hedging. Compensation intelligence, not legal, tax, investment, immigration or employment advice.

Source register — 27 sources

DoorDash discloses no city-level or country-level headcount and no attrition rate, so office density and retention pressure cannot be measured from the outside. The 32 percent year-on-year increase should be read as a corporate action rather than a hiring signal: about 4,400 employees from four acquired groups were still outside the harmonised compensation population when the 2026 proxy was filed.