Japan startup expo corporate scouting: work the participant list

METI's Global Startup EXPO closed in Osaka on October 7. If your mandate touches Japan or Korea, the useful part of that event is still sitting on a public page, and you do not need a flight to use it. The harder question is what you do with a participant list between now and December, when the next two rooms open.

Quick answer: Japan startup expo corporate scouting pays off in the participant list, not the floor time. Government-vetted cohorts such as J-Startup, the National Startup Award winners, and the six Korean companies Busan selects for Startup JAPAN EXPO are pre-screened for activity, so a scout's job is to filter that cohort against a written mandate, confirm a Japan entity or named industrial partner, check whether each company is exhibiting with a proof of concept, and route the first contact through METI, KOTRA, or the regional innovation center instead of a cold email.

The participant list is the durable asset, not the floor time

Attendance is the least valuable part of an expo. Three days of booth conversations produce business cards and a triage backlog you will work through in January, which is exactly the failure the Future Scouting newsletter described in its October 5 issue: open innovation measured in connections rather than in speed of transformation. The participant list, by contrast, does not expire on Sunday. It is a snapshot of which companies a government agency judged ready to be shown to foreign buyers, and it stays useful for the next two quarters.

Global Startup EXPO 2026 was built around J-Startup companies, National Startup Award winners, CVC funds, and government-backed investors, with a16z, NEA, and Atomico on the speaker list. Those speaker names are the reason a scout books the trip and the reason the trip underperforms. You do not need to hear a growth-stage VC's Japan thesis. You need to know which of the exhibiting companies has a line trial running with a named Japanese manufacturer.

Government-vetted cohorts screen for something a directory cannot

A cohort list carries a screening signal that no database field reproduces. J-Startup selection runs through recommendations from investors and accelerators and then government endorsement, which means someone with their name attached judged the company real. The National Startup Award is a ministerial-level prize, so the winner list functions as an activity timestamp. Busan's selected six for Tokyo will have been chosen against criteria that include readiness for corporate and investor meetings and for POC project support.

Compare that to the standard commercial list. Tracxn's Japan energy page returns Mitsubishi Electric and Toshiba among its "startups," which tells you the list was assembled by sector tag and not by anyone asking whether the entity is an acquisition candidate. Directories are indexes, not diligence. A cohort is a short list someone already defended in a meeting.

The gap is also a reason cohort lists are worth building a habit around. Japan's GX Acceleration Agency is making equity investments in deep-tech companies under the ¥20 trillion GX program and disclosed its first battery investment without naming the amount, so GX-funded companies are active and capitalized while remaining close to invisible in standard screens. We wrote about that cohort separately in Japan GX startups: sourcing active companies METI funds. Korea runs the same pattern through Super-Gap, which is funding 120 startups across 12 deep-tech industries, mapped in Korea deep tech startups 2026.

Six checks that turn a cohort list into a mandate-ready short list

A cohort list arrives as 40 to 200 names and no ranking. These six checks cut it to something a two-person corp-dev team can actually work, and they run in order because each one is cheaper than the one after it.

Checks two, four, and five are the ones that consume a week of analyst time per cohort, because they mean entity lookups, Japanese-language newsroom reading, and figuring out who actually owns the relationship. That verification layer is the part Chibit runs for you: active, relevant, vetted matches against a stated mandate rather than a sector tag dump you then have to defend.

Floor-first sourcing versus list-first sourcing

The two approaches cost about the same and produce very different January pipelines. Here is what each one actually yields.

List-first does not mean never travel. It means you travel against a short list. GCV Asia Congress runs in Tokyo November 9 to 11, and COMEUP 2026 is at COEX Seoul December 9 to 11, where the 2025 edition logged 3,447 investment and business-matching sessions and the COMEUP Stars cohort routes 30 companies through country tracks including Japan and the US. Those are matching platforms. Walking into a matching platform without a pre-scored list is how teams end up with 3,447 sessions' worth of noise and four follow-ups.

Where this playbook usually goes wrong

Most of the damage happens in the first 48 hours after a scout downloads a cohort list. The mistakes repeat across teams.

Scouts treat the funding round as the activity signal. FounderNest's 2026 scouting report, built on more than 1,500 dealmakers, found that most corporate teams still run outdated playbooks that miss 40 to 60 percent of the market. Screening on last-round date is one of those playbooks, and in Japan it is worse than average, because GX and NEDO money often does not show up as a priced round at all.

The second mistake is cold-emailing the general inquiry address. A Japanese deep-tech company with a government-program affiliation has an introducer whose job description includes making that connection. If you cannot contact them through a named route, they are not on your short list yet; they are on your research list.

The third is assuming the English materials indicate readiness. An English deck means a company has done an overseas program. It does not mean the company has a Japan entity that can countersign a pilot agreement, and it does not mean anyone on the team can run a technical exchange with your plant engineers in Osaka. Check the entity, not the deck.

What to do this week

Two dates still matter. Busan's application window closes at 5:59pm on October 15, which means the selected six for Startup JAPAN EXPO 2026 at GYM-EX in Tokyo's Ariake district on December 16 will be known shortly after. A scout with a Korea or Japan mandate should put a calendar note on the announcement, request the cohort from BCCEI or KOTRA, and run the six checks before the Tokyo date rather than at it.

For the Osaka cohort, the window is open now and nobody is competing for it, because most teams who attended are still writing up notes. Pull the exhibitor list, filter against your mandate, and send six specific asks through named introducers this month. The broader Japan picture behind these cohorts is in Japan deep tech startups: sourcing the 2025–2026 rebound.

FAQ

Is it worth attending a Japanese startup expo if I already have the participant list?

Attendance is worth it once you have a scored short list and booked meetings, and close to worthless before that. A corporate scout should treat the expo as a venue for 8 to 12 pre-arranged conversations about specific pilots, not as a discovery exercise.

How do I find Japanese deep tech startups outside the expo cohorts?

The best sources outside expo cohorts are government program rosters: J-Startup, NEDO project participants, GX Acceleration Agency portfolio companies, and prefectural innovation center portfolios. These rosters are published in Japanese first and are rarely mirrored in commercial databases, which is why the companies on them look absent from standard screens.

What is J-Startup?

J-Startup is a METI-led program that endorses a selected group of Japanese startups for domestic and international support, with selection driven by recommendations from investors, accelerators, and corporates. For a corporate scout, J-Startup membership is a screening signal and also a routing signal, because each company has supporting organizations who can make an introduction.

Which Korea Japan open innovation program should a corp-dev team watch in Q4?

The Busan Center for Creative Economy and Innovation's Startup JAPAN EXPO track is the sharpest Korea Japan open innovation program on the Q4 calendar, because it selects six companies, places them in Tokyo on December 16, and includes POC project support rather than booth space alone. KOTRA's overseas matching programs and COMEUP Stars country tracks cover the same corridor with wider funnels and looser screening.

Do I need a local entity to run a pilot with a Japanese startup?

You do not need a Japanese entity yourself, but the startup usually does. A target without a KK or GK registration will typically need three to six months before it can countersign a procurement or pilot agreement with a Japanese subsidiary, which is why the entity check belongs early in the screen rather than in diligence.

If your Q4 mandate touches Japan or Korea and you would rather start from a scored short list than a cohort PDF, describe the mandate to Innovation Scout and see which companies come back active and relevant. That is the same screen described above, run for you.