At a glance
AcademicJobs.com's THE Network pairs an ad-free, specialty-based academic community with a contracting faculty job market. Here is what the evidence says about small rooms, who they help, and where the limits sit.
THE Network launched into an academic labour market that had stopped behaving like one. The need it addresses is narrow: researchers who must find the handful of people in the world working on their exact problem, and search committees who must reach those people without reading two hundred applications.
AcademicJobs.com, the Melbourne-founded job board that lists more than 72,000 academic posts worldwide, built the network as a layer above its listings. Members declare a specialty rather than a job title — Victorian literature, coastal sediment modelling, academic human resources — and the platform sorts them into rooms on that basis. There are no sponsored posts and no advertising. Membership is free for verified faculty, researchers, leaders and professional staff.
The scale claim on the site is 15 million higher education professionals worldwide. Treat that as an addressable market rather than a membership count; the company publishes no audited member figure. What can be checked is the design, and the design is a bet that a small room produces better answers than a large feed.
The people most exposed to the change are the ones with the least institutional memory. Nature's 2023 survey of postdoctoral researchers drew 3,838 responses from 93 countries. Sixty-one percent were working outside their country of origin, and satisfaction with the postdoc experience had fallen five points since 2020. A two-year contract in a new country is not a network. The corridor reaches the people in one building.
The second group runs searches. A chair filling a lectureship in a narrow field usually begins with a stack and works backwards, hoping a qualified stranger appears. A specialty-based pool lets her begin with names instead. That is a different task, and a faster one.
Then there is the group nobody puts on the panel: the research office administrator who maintains the induction file, the faculty HR coordinator who notices the new hire is the only person in the building working on that question. They decide whether a new starter ever gets into the room at all.
The demand side is thinner than it was. The U.S. Bureau of Labor Statistics projects employment of postsecondary teachers to grow 7 percent between 2025 and 2035, from 1,378,200 to 1,476,400, with about 103,300 openings a year — most of them replacements rather than new seats, according to the Occupational Outlook Handbook.
Inside that national figure, hiring has tightened sharply in specific fields. The Computing Research Association's Center for Evaluating the Research Pipeline surveyed its academic members in late 2025 and found a contraction of at least 27 percent: 147 responding units hired 439 full-time faculty in 2024-25 and expected 319 in 2025-26. Half reported fewer open positions, and the number of units with no full-time hires rose from 26 to 36, the CRA reported.
Craig Wills's long-running analysis of advertised computer science searches found 311 institutions seeking 588 tenure-track positions, roughly 25 percent fewer institutions and 33 percent fewer positions than in 2024. Nearly half of the advertised roles sat in artificial intelligence and security. Regional variation was brutal: a 65 percent drop in positions in the Pacific West and 60 percent in the Plains.
Economics has its own ledger. Job listings on the American Economic Association's JOE board fell to 1,767 between August and December 2025, against 2,261 a year earlier — down 21.8 percent. Full-time academic posts in the United States dropped 32.7 percent. A running chemistry faculty list put 2025-26 at 185 openings at PhD-granting institutions, 104 below the 2022-2025 average of 289.
A thinner market does not reduce the value of matching. It raises it. When there are fewer searches, a misfiled application costs more, and a department that advertises once cannot afford to reach the wrong pool.
Photo by Zulfugar Karimov on Unsplash
The case for small rooms has evidence behind it, though less than the pitch decks imply. Sen Chai and Richard Freeman studied 1,265 attendees at fifteen Gordon Research Conferences in the biological sciences between 1991 and 1995, matched against comparable researchers who did not attend. Attendees with no prior collaboration inside the conference produced about 9 percent more joint publications afterwards, and junior researchers drew the largest benefit, as the NBER working paper set out.
The mechanism is not mysterious. Take the meeting away and the ties mostly do not form. A PLOS ONE case study of the International Milk Genomics Consortium tracked 311 attendees and found that after a first symposium, 64.3 percent had more co-authors and 88.7 percent published at a higher rate.
Scepticism deserves equal time. An analysis of the DBLP computer science database, covering more than a million papers and over 640,000 authors, concluded that at best 4.61 percent of collaborations could have emerged from people meeting at a conference. Proximity alone does little. Structure and follow-through do the work, which is why a room is an entry point rather than an outcome.
For early-career researchers the arithmetic is unforgiving. Collaborators with complementary methods are needed before a tenure clock starts, not after. Our guide to thriving in a postdoctoral role makes the same point from the other direction: the network is pipeline work, not something that begins once the thesis is bound.
The hiring layer sits underneath the community. A.I. Candidate Connect reads profiles on Google Scholar, ResearchGate, Scopus and LinkedIn alongside the platform's own data, compares research topics and publication context against a job advertisement, and notifies up to 100 of the closest matching specialists — described by the company as roughly the top fifth for the role. Most outreach goes out within about 90 seconds of a posting going live.
Pricing is published: US$345 for a single U.S. listing, US$999 for a month of unlimited postings, US$9,000 for a year. Candidate Connect is included with the unlimited plans and has been sold separately at A$200 per role in Australia.
Speed of notification is not the same as a shortlist, and no algorithm sits on an interview panel. The part that decides an outcome is still the human sequence — a clearly scoped role, a reply inside a day, a process that does not swallow documents. Our breakdown of employer branding in higher education hiring covers the same ground: passive candidates respond to a role written in plain terms and a recruiting process that respects their time.
Two limits are worth stating plainly, because the company states them too. The network will not find anyone a job by itself, and it will not make a search committee move faster. That candour is the right kind; a directory sold as a deliverable is where these products usually disappoint.
The harder limit is density. Every niche network faces a cold start, and whether the computational linguists or the coastal modellers you need have already joined is a question only your own field can answer. A general platform wins on volume by default. A specialist platform has to win field by field, and does not get credit for fields it has not filled.
There is also the ownership question. AcademicJobs.com has said blue-chip private equity investors approached the company after THE Network's rollout. No firm has been named and no deal has been signed. Founder Jarrod Kanizay previously built UniJobs in Melbourne from 2006, sold it to TES Global, the parent of Times Higher Education, in 2015, and left in 2020 before launching AcademicJobs.com in 2023. Investors are buying a track record; users are buying whatever the next owner decides to ship.
Photo by Zulfugar Karimov on Unsplash
For anyone testing the thing this term, the useful moves are small and specific.
- Write the profile around a problem, a method and a current project, not around committee service. A stranger should be able to act on the first sentence.
- Join the room before the deadline, not the week a grant call closes. The value compounds from having answered somebody else's question in March.
- If you run a department, add one line to the induction pack naming the specialty room a new hire should join, in their own field's words. Give that line to the administrator who maintains the file.
- If you are hiring, ask the platform to demonstrate reach into passive candidates — people who are not applying but would move for the right lab. That is the claim worth testing before committing a listing budget.
The room costs nothing to enter. The price is the hour it takes to describe your work in terms a stranger on another continent can use.
- higher-education
- academic-jobs
- academic-networking
- research-collaboration
- faculty-hiring

