Everyone knows the UXR hiring market is bad. How bad is it, though?

This post is a detour. Part 3 of the AI-native teams series was supposed to run in this slot, and it will come next instead. I keep seeing the same questions about the UX research job market, in research communities, in my inbox, under every post that touches hiring, and the answers being traded are mostly guesses attached to one statistic from 2023. So I did the deep dive.

The dedicated UX research job market is in bad shape. That is the short answer, and the data support it. The longer answer is more specific and, I think, more useful. Demand for research is rising at the same time as the number of people employed to do it as a dedicated job is being held down, and those two facts fit together once you look at who is being asked to do the work. Companies are authorizing fewer research seats, hiring at higher levels, filling more roles from inside, and spreading the work across product managers, designers, marketers, market researchers, data scientists, and the new category of people who evaluate how AI systems behave. If your title is UX Researcher and you are looking for the next job, you are competing for a smaller number of seats against a candidate pool that has not shrunk at all.

There is no official data on UX research employment

Nobody can tell you how many UX researchers there are, how many are unemployed, or how many UXR jobs have been eliminated since 2022. The U.S. government does not track the occupation. O*NET files conducting user research as a task inside the broader Web and Digital Interface Designers category, so there is no clean headcount, unemployment rate, or wage series to pull out. The titles do not help either. The same job gets posted as User Researcher, Product Researcher, Design Researcher, Experience Researcher, Consumer Insights Researcher, etc.

What exists instead is a set of partial sources that each see one corner of the problem: Indeed's job posting data, practitioner surveys from User Interviews and Maze, a salary dataset, a study of job descriptions, hiring-funnel data from an HR platform, and general technology-sector employment figures. Each has a known weakness. The practitioner surveys are convenience samples that overrepresent people still attached to the field. User Interviews says this about its own 2025 report: 485 qualified responses recruited through its newsletter, its product, LinkedIn, and Slack groups, with no clear route in for unemployed researchers, which the authors admit may make it read like a state of employed user research. Job postings are not hires, salary data only describe people who have jobs, and the tech-sector figures do not isolate researchers.

Several imperfect sources with different biases all pointing the same way is worth something. It is still a triangulation, and the numbers below should be read as a direction and a rough size. Anyone quoting a precise unemployment rate for UX researchers is making it up.

UX research job postings fell 73% from 2022 to 2023, then went flat

The number everyone repeats is that UX research job postings fell 73%. It comes from Indeed's own data. With Q1 2018 indexed to 100, UXR postings climbed to 566 in Q1 2022 and fell to 153 by Q2 2023. UX design rose to 265 and fell to 77 over the same cycle.

Two things about that number get lost when it is passed around.

The first is the starting point. UX research did more than grow during the pandemic. It expanded to about 5.7 times its 2018 level in four years, while UX design peaked at about 2.7 times. The bigger bubble produced the bigger correction. A 73% drop from an index of 566 lands at 153, which is still above where the discipline started in 2018. Nobody laid off in 2023 will find that comforting, but it changes what the number means: it describes the deflation of an abnormal peak, and the trough sat above the pre-pandemic starting line.

The second is the date. The 73% is a peak-to-trough comparison that ended in mid-2023. It is not a current rate of decline. Indeed's data through Q4 2025, published in January 2026, show postings across UX research, UX and product design, product management, and software engineering flattening out from the second half of 2023 and holding at something like a baseline level since. Relative to its own 2018 starting point, user research has held up better than design or engineering, both of which dipped below their baselines in Q3 2023; the authors put that down to research starting from a much smaller base. They are also careful to say that stable postings tell you nothing about how many people are competing for them.

That is the lasting problem. Roles fell and then flattened, and the candidate pool did neither. The researchers hired during the expansion are still in the market, along with the ones laid off since 2022 and the career changers and graduates who trained for the 2021 version of the field. Supply stayed high while demand reset to a lower, flat line. The market researchers describe in 2026 is the result: steady, and too small for the number of people in it.

Tech job postings are still a third below 2020 and layoffs are rising

UX research lives inside the technology economy, and that economy is the part of the labor market that has not recovered. As of October 31, 2025, technology job postings on Indeed were almost one third below their early-2020 level, while total postings across the U.S. economy were slightly above the pre-pandemic baseline.

Layoff announcements say the same thing. Through July 2026, technology companies had announced 149,023 job cuts, 67% more than in the same period of 2025, and tech accounted for 31% of all announced U.S. cuts. Those are employer announcements rather than completed separations, and the reasons companies give for cuts are not reliable. None of it is specific to research, but it describes the sector where most research jobs sit: few teams expanding, a large stock of experienced candidates outside, and employers who know they can wait for an exact match.

One in five companies laid off researchers last year and 14% now have none

In the 2025 State of User Research survey, 21% of respondents said their company had laid off dedicated user or UX researchers in the past year, about the same as in 2024. Eleven percent had been laid off themselves. Nearly 14% worked somewhere with no dedicated researcher at all, up from 6% in 2022. The UXPA salary survey, which MeasuringU has tracked back to 2009, found that 35% of organizations reported losing UX staff in 2024, roughly double the rate for any year between 2009 and 2022.

Research was not singled out. Respondents reported similar layoff rates in product and UX design, engineering, and product management. The difference is scale. A company that cuts 10% of its engineers still has an engineering organization. A company that cuts three researchers from a team of six has lost half its research function. Small disciplines absorb general cost-cutting as something closer to elimination, which is why the market feels worse from inside research than the company-wide numbers suggest.

The other effect of layoffs that never stop is that companies learn to live without the seat. A researcher leaves, the work gets redistributed or dropped, nothing visibly breaks in the next quarter, and the requisition never reopens. I think that is the most plausible explanation for the rise from 6% to 14% of respondents with no dedicated researcher on staff.

Entry-level roles have almost disappeared and postings skew senior

The clearest sign that this is a structural change rather than a slow year is what has happened to junior roles.

The 2025 State of User Research sample had 8.5 percentage points fewer junior respondents than the year before. Researchers at senior, lead, staff, or principal level made up more than half the sample, nearly 10 points more than in 2024. Some of that could be sampling: juniors who lost jobs are less likely to be on the mailing list. But a separate study points the same way from the employer side. Drill Bit Labs analyzed 1,394 U.S. UX research job descriptions collected in late 2023 and early 2024. Entry-level positions were a tiny fraction of the sample, more than half of individual-contributor postings were mid-level, and more than 40% of the common mid-level and senior categories were contract roles. Nielsen Norman Group's 2026 review of the wider UX market reached the same conclusion: the stabilization since late 2024 has been uneven, with senior and generalist roles recovering faster and entry-level positions still scarce.

Read the postings and the pattern is obvious. Employers want someone who can take an ambiguous question from a product leader and turn it into a study, choose the method, run it without supervision, defend the rigor when the findings are inconvenient, understand the business well enough to know which findings matter, and get a decision changed. Those are fair expectations for a senior researcher. They are also, increasingly, what the mid-level and even the nominally junior postings ask for.

The mechanism that used to produce senior researchers was a junior job. You got hired to run the usability sessions someone else scoped, you took the notes, tagged the transcripts, did the desk research, and after a few years you had the judgment the senior postings now demand. Companies have cut that job and kept the demand for its output. The profession is asking people to arrive with experience it has stopped providing, and then treating the gap as the candidate's problem.

This is worse than a hard first-job market. A hard first-job market eventually clears. A discipline that removes its own training positions does not get its next generation back when budgets improve, because the people who would have become mid-level never got in.

Demand for research is rising, and product and marketing teams are doing it

Maze's 2026 survey found 66% of respondents reporting increased demand for research, up from roughly 55% the year before. Demand is up, and the job market for researchers is flat or worse. The explanation is in the same report. Product managers were doing research in 39% of organizations, market researchers in 35%, marketers in 23%. Fewer than half of organizations gave those non-researchers any dedicated specialist support.

User Interviews found the other end of the same movement. Twenty percent of its respondents had moved into a new function or company in the past year. Among respondents who no longer held a research title, 63% said they used to. The report raised the possibility that research is becoming a multidisciplinary skill rather than a distinct role.

Together, those two findings describe one thing from both ends. Research is being unbundled. Discovery interviews go to product managers, usability testing to designers, customer conversations to marketing or customer success, surveys and segmentation to market research, and behavioral analysis to data science, while evaluating how an AI system behaves with real people is turning into its own specialist job with its own titles. The company still gets research-shaped output from every one of those places. What it no longer needs, or believes it no longer needs, is a dedicated team whose only job is producing it.

Whether the unbundled version is any good is a separate question, and my answer is that it is usually worse. A product manager with an AI interview tool can produce something that looks like a research readout. Whether it would hold up if a researcher asked about the sample, the leading questions, or the gap between what participants said and what they did is a different matter. But research quality is mostly invisible to the people approving headcount. What they see is whether a decision got made with something that looked like evidence attached, and the cheaper version that looks like evidence wins that comparison most of the time.

AI is reducing how many researchers a team can justify

The timeline matters. UXR postings peaked in Q1 2022 and were already falling through that year, before ChatGPT launched at the end of November 2022. Pandemic overhiring, the end of near-zero interest rates, the collapse in venture funding that followed, and company-wide cost reduction explain the original drop without any help from generative AI. Blaming the 2022 collapse on AI gets the order of events backwards.

AI is in the workflow now. Eighty percent of respondents to the 2025 User Interviews survey reported using AI in their research, up 24 points in one year. Maze put adoption at 69% across its 2026 sample, concentrated on transcription, synthesis, and generating research questions, with respondents reporting faster turnaround.

Nobody has credible evidence on how many researchers AI has directly replaced, and direct replacement is the wrong test anyway. For a seat to disappear, AI only has to raise the amount of work one researcher, designer, or product manager can plausibly cover. The argument for the next hire gets harder from there.

Take a hypothetical that most research leads will recognize: a team that used to justify six researchers gets told that four with AI tooling should manage. A junior role built on transcription, tagging, desk research, study setup, and first-pass synthesis is now mostly a list of things the tools do. I think the tools do them worse than a good junior researcher, and nobody checks, because what the tools are being compared against is the cost of the seat rather than the quality of the work.

What AI has done is reprice the two halves of the job. Execution, the part junior roles were built on, is worth less. Judgment, meaning knowing what to study, whether the evidence deserves belief, and what to do when the numbers contradict the interviews, is worth more, and it is also the part you cannot hire at entry level. The premium on judgment is real, and judgment used to be learned by doing the execution work that the premium is now devaluing.

Applications per posting have doubled and hires are down 20%

Candidates describe applying to dozens of posted roles and hearing nothing, and the funnel data say they are describing the market accurately. Across BambooHR's five-year dataset of its customers' hiring, applicants per posting roughly doubled, from about 46 in 2021 to 95 in 2025. The hiring rate fell from 4.5% to 2.8%. Completed hires dropped more than 20% between 2022 and 2025 while the number of monthly postings stayed about flat. Internal mobility rose from 51% of fills in 2021 to 62% in 2025. Mentions of remote work in job titles fell from 4.2% in 2022 to 1.6% in 2025.

None of that is specific to research, and it stops short of proving that UXR listings are ghost jobs. What it does show is that a posting converts to an external hire less often than it did, and that every posting attracts about twice the competition. A company can open a requisition, collect the hundred applications the average posting now draws, interview eight people, pause, re-level the role, lose the budget, or give it to someone already on the payroll. Every one of those shows up as an opening on a job board. Only the hire shows up in the hiring rate, which is how postings stay flat while completed hires fall by a fifth.

The remote number is worth a sentence of its own. Remote roles draw applicants from everywhere, and the share of roles labeled remote is down by more than half since 2022. If you can only work remotely, you are competing for the smallest slice of the market against the largest pool of applicants.

Two thirds of employed researchers still make over $100,000

The 2026 UX Salary Report, built from 1,906 User Interviews records and 17,736 Levels.fyi records, found that 67% of 2,062 U.S. UX researchers earned more than $100,000. On its own that reads like a healthy profession.

Salary data describe people who have jobs. They say nothing about the researcher who has been on contracts for two years, the one who has been searching for nine months, or the one who left for a product role and no longer appears in the sample at all. Good pay and few seats is exactly what you would expect from an occupation that has become rationed. When a company decides it needs a senior researcher, it still pays for one. It is making that decision less often, at a higher level, and with a longer list of things the hire is expected to cover.

How bad it is depends on your level and specialty

It is one market, and it is not equally bad for everyone in it.

Entry-level researchers and career changers have it worst, and it is not close. The playbook that worked in 2021, a short course, three portfolio studies, and a wide net of applications, no longer gets interviews. The roles it was aimed at mostly do not exist, and the ones that do carry mid-level expectations. I do not have a good answer for someone in this position, and I have not seen a convincing one from anyone else. The least bad route I can see is an adjacent job that does research without the title, in product, marketing, customer success, or research operations, and building the record from there.

Mid-level qualitative researchers have a market, and it is crowded. Interviews and usability tests are still useful, and they no longer differentiate anyone, because every candidate can run them and, increasingly, so can the product manager. What separates candidates at this level is domain knowledge, evidence that a decision changed because of their work, enough quantitative competence to be trusted with a survey that matters, and the ability to handle a stakeholder who does not want to hear the finding.

Senior mixed-methods and quantitative researchers still have a viable market, and it is selective. In 2026 postings, mixed methods means survey design, experiment literacy, comfort with product metrics, and the judgment to combine those with interviews. Tacking a five-question survey onto the end of an interview study does not count.

AI research and evaluation is the one segment where demand looks to be growing, and it is harder to get into than the demand suggests. The work involves probabilistic systems that behave differently every time, evaluation design, trust and safety questions, longitudinal effects, and products that change between study design and fieldwork. Researchers with a quantitative or behavioral science background move into it more easily than researchers whose experience is mostly moderated qualitative work. The opportunity is real and the bar is high.

Research managers face a thinner market than senior individual contributors. Leaner teams need fewer managers, and the senior leadership roles that do open tend to be filled through networks or executive search and often never reach a job board.

Remote-only candidates are working with the narrowest slice of the market, for the reasons above. Geographic flexibility is now a competitive advantage, which it was not in 2022.

The work survives, with fewer and more senior people doing it

Research survives because companies keep making expensive decisions under uncertainty, and some of them remember what it cost the last time they got one wrong. That demand is durable. The question is which people get paid to meet it.

The researcher in the most danger is the one whose job is to receive a request, run the requested study, deliver the findings, and wait for the next request. That job can be consolidated into a smaller team, distributed to the product manager, partly automated, contracted out, or cut, and companies are doing all five.

The researcher who is harder to replace changes the question before the study starts. That researcher can tell a team that the five people who loved the prototype were recruited from the company's own power users, or that the behavioral data contradict what participants said, or that the problem being treated as a usability issue is a market issue, or that the product team is about to spend six months building something nobody asked for. The role borrows from product strategy, market research, behavioral science, data analysis, experimentation, and AI evaluation, and it is less purely UX than the title suggests. I think it produces more influential researchers. It also produces fewer of them.

The answer to the question in the title has two parts. Companies still value research, and nothing in the data says otherwise. What they have stopped paying for is the dedicated researcher role in the shape it had in 2022: a team of specialists, a junior pipeline feeding it, and a monopoly on producing the evidence. They want customer understanding without adding headcount, and they want senior judgment without funding the junior roles that create it.

For someone with eight years of experience, a quantitative skill set, and a network, the market is selective and maybe survivable. For someone trying to get in, it is about as cooked as a job market gets, and the standard advice about portfolios and persistence will not fix a market with no entry level. The work will keep getting done. The test of whether this model holds is whether the companies that cut their junior pipelines can still find senior researchers in three or four years, and the evidence on that does not exist yet.

The advice that follows from the data: build the network early, and do not be an asshole

I have kept advice out of this piece because most of what circulates is guessing. But the data points at one conclusion worth acting on. Internal candidates take 62% of fills, leadership roles get passed around through networks before they are posted, and the public channel converts at under 3%. The paths that still work run through people who already know you.

And UX research is a small field. Small as fuck, honestly. The same names cycle through the same companies, Slack groups, and conferences, and the junior you dismissed in 2022 is a hiring manager in 2026. In a community this size, being someone people want to work with again is not a soft skill. It is the strongest predictor of how your next search goes. So:

• Build the network while you are employed. A network assembled the week after a layoff is a list of strangers receiving asks. One built over years is people who forward your name before the role is posted.
• Give before you ask. Answer questions in communities, share what you learn, review a portfolio, make an intro. It compounds, and it costs you almost nothing while you have a job.
• Stay in touch when you need nothing. Sudden warmth after three years of silence reads exactly how it sounds.
• Skip the weird version. No cold messages that open with flattery and close with an ask, no coffee chats that are really referral requests. Network like a person or not at all.
• Assume the reference check is already happening. In a field this size, it usually is.
• Managers and directors: this is mostly for you, and I mean that kindly. Equivalent openings are the rarest in the field, leaner teams need fewer leaders, and the roles that appear get filled through someone who already knew the person. The title protects you while you hold it. The relationships are what you keep when it goes.

None of this fixes the market. It decides how the market treats you when your turn comes, and it is the only part you fully control.

🎯 Part 3 of UXR for AI-native teams is next. Subscribe at www.thevoiceofuser.com and it arrives when it publishes, no algorithm required.