Staff UX Researcher (L6/IC6): Research Strategy at FAANG-Tier

What staff UXR means at FAANG-tier

Staff UX Researcher (L6/IC6) is the senior individual-contributor rung that runs parallel to research management at FAANG-tier companies. Where a senior UXR (L5) owns research for a single product surface and partners with one or two PM/design counterparts, a staff UXR sets research strategy across multiple surfaces, authors decisions about how the research organization itself is structured, and owns a research roadmap whose outputs influence company-level product strategy rather than feature-level shipping decisions.

The defining shift from senior to staff is not depth of method, but breadth of leverage. A staff UXR rarely runs more usability tests than a senior; they run fewer studies of larger scope, and they spend a meaningful share of their time on what Kate Towsey calls the "research operations layer" in Research That Scales — tooling, repositories, participant infrastructure, and democratization boundaries that determine how much research the rest of the organization can absorb.

Will Larson's framework for staff engineering archetypes applies cleanly here. The four staff-engineer archetypes Larson identifies on lethain.com — Tech Lead, Architect, Solver, and Right Hand — map onto staff UXR almost without modification. A "Solver" staff UXR parachutes into the hardest foundational question on the org's plate. An "Architect" staff UXR designs the repository taxonomy that ten teams will live inside for three years. A "Tech Lead" staff UXR carries the research strategy for a major product line, with several senior researchers operating inside that frame. A "Right Hand" staff UXR partners directly with a VP of Product or Design, translating organizational ambiguity into research-shaped decisions. The shape of the role is recognizably the same.

Two scope expectations distinguish staff from senior in practice. First, a staff UXR is expected to influence the work of researchers who do not report to them — through artifacts, frameworks, and example-setting rather than authority. Second, a staff UXR is expected to make the research function more capable than they found it: a better repository, a sharper participant pipeline, a more honest standard for what counts as evidence. If the researcher leaves and nothing is structurally different, the role was not staff-level.

Staff-level interview bar

The staff UXR loop has three rounds that dominate the outcome. Tactical method rounds still appear, but rarely decide the offer.

Research-strategy round (the dominant filter)

You are given a product area and asked to author a 12-to-18-month research strategy. The bar is not a clever study design; it is whether your strategy ladders up to a business question worth answering, sequences sub-projects so that early findings de-risk later ones, and explicitly names what you are not going to research and why. Interviewers are listening for the second-order question behind the first-order question. Tomer Sharon's writing at tomersharon.com on framing research questions is the closest public reference to what is being evaluated.

Executive-presence behavioral

You will be asked about a moment you changed an executive's mind, a moment you held a research finding against pressure to soften it, and a moment you decided not to do a piece of research that was requested. The signal is whether you treat research as a service function or as a peer discipline. Erika Hall's writing at muleshq.com on research as a strategic act, not a deliverable, captures the posture being assessed.

Foundational-research deep-dive

Walk through a multi-quarter foundational study you led. The interviewer is checking three things: did the study reshape the roadmap, did it produce reusable artifacts other teams still cite, and could you articulate what you would do differently with a year of hindsight. Roadmap-shaping impact is the load-bearing word here.

Comp at staff (L6/IC6)

Total compensation at FAANG-tier companies for staff UXR (L6/IC6) generally lands between $470,000 and $700,000+ per year, with the upper band concentrated at Meta E6 and Google L6 in high-cost metros. The composition is roughly 35-45% base salary, 35-45% equity (vesting four years, often front-loaded), and 10-20% target bonus.

The most reliable public source is levels.fyi/t/ux-researcher, which aggregates self-reported offers and lets you filter by company and level. Read the median, not the maximum — the maximum is usually a refresh-heavy senior staff offer that has been mislabeled.

Two structural notes. First, staff UXR comp at FAANG-tier is competitive with staff software engineering comp at the same level, which was not true a decade ago and reflects the maturation of research as a discipline. Second, the gap between L5 senior and L6 staff is larger than the gap between L4 and L5; the staff promotion is the largest single comp jump in the IC ladder, and it is gated on demonstrated company-level influence, not tenure.

Outside FAANG, the picture changes. Staff-equivalent UXR titles at later-stage startups and mid-sized public software companies typically pay $300,000-$450,000 total compensation, with a higher base-to-equity ratio and a wider variance driven by equity quality. Public foundational research output, of the kind catalogued at research.google, is concentrated at companies that pay at the top of the band; the correlation between visible research investment and staff-UXR comp is not coincidental.

One negotiation note: at staff level, the most consequential lever is rarely base salary. It is initial equity grant size and refresh expectation. A 10% increase in base is worth a fraction of a 25% increase in initial equity, and recruiters know this. Treat the equity grant as the primary number.

Worked scenario: 12-month research-strategy redesign

Consider a staff UXR at a FAANG-tier company who owns research strategy for a productivity-software surface used by tens of millions of knowledge workers. The product's annual planning cycle has named "collaboration" as a top-three bet, but the organization disagrees about what collaboration means, who the users are, and which workflows are broken. The staff UXR is asked to author a 12-month foundational research program that resolves this.

The program is decomposed into four-to-six sub-projects, sequenced so that each later study uses the population segmentation produced by the earlier one:

1. Q1, ethnographic foundational study — in-context observation of collaboration moments across twenty teams, output is a published taxonomy of collaboration archetypes.
2. Q2, longitudinal diary study — six-week diary across the archetypes from Q1, output is a friction map keyed to archetype.
3. Q2-Q3, large-N quantitative validation — survey instrument calibrated against the diary findings, output is prevalence estimates for each friction.
4. Q3, jobs-to-be-done synthesis — cross-reads the prior three studies, output is a JTBD framework that engineering and design organize their roadmaps around.
5. Q4, evaluative pre-mortems — small evaluative studies on the highest-leverage product bets the JTBD framework surfaces, output is a kill-or-keep recommendation per bet.

Two staff-level moves are visible in this design. First, the research artifacts are infrastructure: the archetype taxonomy and the JTBD framework will be cited in roadmap documents for years, which is the leverage Kate Towsey describes in Research That Scales. Second, the program is sequenced so that the staff UXR can hand later sub-projects to senior UXRs once the foundational frame is set, freeing them to work on the next strategy. Senior researchers run studies; staff researchers build the frame the studies fit inside.

The harder, less-discussed staff move is what is excluded. A program of this scope generates dozens of incoming requests for tactical evaluative research, each individually reasonable. The staff UXR's job is to refuse most of them in a way that does not damage relationships, by pointing requesters toward the upcoming JTBD framework and the prevalence data the survey will produce. Erika Hall's framing — that research is a strategic act, not a service function — is what makes those refusals defensible.

Outputs at the end of twelve months: a published archetype taxonomy adopted in three roadmap documents, a friction map cited by two engineering directors in their planning, prevalence estimates that killed one previously-funded bet and accelerated another, a JTBD framework picked up by a sister product organization, and four senior researchers who can now run their own multi-quarter programs inside the frame. The staff UXR is, by the end of the year, partly redundant inside their own program — which is the correct outcome and the signal that the next staff-scope problem is theirs to define.