Glean Japan: How a $300M AI Company Is Building to Win Early

24 Jun

Written By Murray Clarke

starting ARR

$100M

early 2025

months to triple

early 2025 → May 2026

current ARR

$300M

May 2026

ARR (USD millions)

Source: Glean press releases / TechCrunch, May 2026

Japan Market Entry

Most market-entry stories take years to read clearly. Glean’s took months — $300M ARR, a founding leader landing in Tokyo, and a partner-led beachhead built two years before the country manager arrived. The order is the strategy.

Murray ClarkeFounding Partner, TalentHub PartnersTokyo

Most market-entry stories take years to read clearly. Glean’s took months. The company crossed $300 million in annual recurring revenue in late May 2026 — roughly three times where it sat fifteen months earlier — and it is now putting a founding leader on the ground in Tokyo. I ran that search. The country manager starts the day this post goes live.

That timing is not an accident, and it is the whole point. In enterprise AI, the Japan window is not a five-year runway. It is a short sprint to plant a flag before the category hardens around whoever moved first. Glean is moving.

Here is the thesis I keep coming back to. In AI, the companies that win Japan are not the ones with the cleverest demo or the deepest balance sheet. They are the ones that get in early, pick the right founding leader, and commit real money before the incumbents wake up. Glean is doing all three — and the Box story tells us why that works even when Microsoft and Google are in the room.

The Context: Japan Was Never An Afterthought For Glean

Glean was founded in 2019 by former Google search engineers to fix a boring, universal problem: knowledge scattered across every system a company runs. That heritage matters less than what the company did with Japan. It did not file Japan under “phase three, once we figure out the US.” It signed a Japanese channel partner in 2023 — more than two years before it hired a country manager.

That is the part most US AI companies get wrong. They wait until they have a leader, then ask that leader to build everything from a standing start. Glean ran it the other way: a partner-led beachhead first, then direct leadership layered on top once the market signal was clear.

The urgency around them is real. METI’s own “2025 Digital Cliff” warning put the cost of Japan’s aging legacy systems at up to ¥12 trillion a year if firms fail to modernise — a number the government has reaffirmed as the deadline arrived. Add a shrinking working-age population and a labour market that cannot hire its way out of knowledge work, and you have the conditions where a tool that gives every employee back hours is not a nice-to-have. It is survival. Japanese enterprises are not asking whether to adopt enterprise AI. They are asking who to trust.

Why The Leadership Choice Mattered

This is where the case turns, and it is the part I lived.

Glean first ran the country manager search itself. That instinct was reasonable. Arvind Jain co-founded Rubrik before Glean — a company that went public in April 2024 at a $5.6 billion valuation — and founder-led searches had worked for him before. When something has worked, you reach for it again. But a self-run search for a founding Japan builder is a different animal from filling a role you understand cold. It cost the company four to five months. In a market moving this fast, that is the most expensive line item nobody puts on a budget.

After an introduction from one of Glean’s investors, I took the search on a retained basis. The mandate was narrow and correct: a proven builder from a tier-1 success case. Not a caretaker, not a quota-carrier with a bigger title — a 0-to-1 operator who has actually stood up a Japan business and made the lighthouse accounts say yes. That brief eliminates most of the market in a sentence. So I built the market map and shared it with the client, because the only way to defend a narrow mandate is to show you have looked at everyone it could possibly be.

Then the useful, slightly humbling part. Glean’s own recruiters had already messaged the person who eventually took the job. He never replied. The best builders almost never answer inbound — their inbox is a graveyard of templated outreach. I had known this person for more than fifteen years, and even he did not reply to me straight away. He came back only after he had done his own diligence on Glean, quietly, through another of his VC contacts. That is the detail every hiring company should sit with: at this level, trust runs candidate-to-company at least as hard as it runs the other way. He was vetting Glean before Glean ever got to vet him. Our landing page did real work here — it laid out the opportunity, the challenges to overcome, and the growth plan in one place, so by the time he looked seriously, the case was already made.

The Public Numbers Show Real Momentum

A founding leader does not say yes to a flat line. The momentum behind this search is on the record.

Glean reached $300 million in ARR in late May 2026 — about three times the $100 million it had reached only fifteen months before. The staircase is steep: roughly $100M in early 2025, $200M by December, $250M by February and $300M by the end of May. The company was last valued at $7.2 billion when it raised a $150 million Series F in mid-2025. Its customer list now includes Databricks, Reddit, Pinterest and Samsung.

The competitive read matters more than any single figure. For its first few years Glean had the category largely to itself. Now the heavyweights are building Glean-like tools — Google, Microsoft, OpenAI, Anthropic, Salesforce and Atlassian among them. That is exactly why the Japan clock is ticking. First-mover revenue is the permission slip to spend ahead of the competition in a new market. Glean has earned that permission, and it is using it.

Considering my own experience with a wide range of technology vendors over the years, Glean is the first one that I came across where every single person I talked to who was using the product raved about how good it is. This kind of honest feedback gains attention when it is shared to senior builders across the market.

The Partner Strategy Was Already In Motion

By the time the country manager walks in, Japan is not a blank page.

Glean’s first domestic reseller is Ashisuto, which signed an agency agreement announced on 31 May 2023. The targeting was precise from day one: Japanese enterprises with several thousand employees already running Box and Microsoft 365, with annual subscriptions reported from around ¥39 million. An early adopter, Chuden CTI, put Glean to work as a company-wide search engine across its cloud systems. Tellingly, Ashisuto was a Glean customer before it became a partner — the most credible kind of reseller, because it sells what it already runs.

With first-mover advantage, Ashisuto has been both aggressive and successful in building the business for Glean, citing a wide range of customer success cases, including a 10,000-seat deal with Kansai Electric Power company. Other partnerships have followed, including with Tokyo Electron Device in 2025 and Cloud Ace, announced just recently in 2026.

So the founding leader inherits a channel, live references and a pipeline. The job is not to invent demand from zero; it is to convert a partner-built beachhead into a direct enterprise motion and a real Japan team. That is a far better starting hand than most first country managers are dealt.

What Box Taught Us About Coming In Hard

I will admit something. For years I expected Box to get flattened in Japan. The logic looked airtight: Microsoft owns the desktop with SharePoint and OneDrive, Google has Workspace, and a mid-size US content company was going to be ground between them. I was wrong.

Box came in hard and Japan became its largest market outside the United States. It did it channel-first — through Fujitsu and a deep ecosystem of local partners — and built an enterprise and public-sector install base while the incumbents owned the productivity suite. The company went on to pass $1 billion in annual revenue. Aaron Levie’s own framing of Japan is simply that it is “a very channel oriented environment,” and Box respected that instead of fighting it.

The lesson transfers almost cleanly to Glean. Incumbency at the operating-system and productivity layer does not decide who wins the knowledge layer in Japan. Commitment decides it. The right local leader decides it. A partner-led motion decides it. Glean’s setup rhymes with Box’s at every beat: a partner-first entry, enterprise targets, and a category — “Work AI” — that sits across Microsoft and Google rather than trying to beat them at their own desktop. Box proved the room can have Microsoft and Google in it and still belong to the company that took Japan seriously first.

Why This Matters For Other Companies Entering Japan

This is the most recent case I have to point to, and it is more condensed than the ones I usually write up — the market simply will not wait for AI entrants the way it waited for earlier waves of software. Many companies are about to run this exact play, so it is worth saying plainly what the pattern is.

Get in early through a credible partner. Glean signed Ashisuto in 2023, not 2026. The beachhead pre-dates the leader.

Hire a founding builder, not a caretaker. The mandate was a proven 0-to-1 operator from a tier-1 success. Narrow on purpose.

Run the search properly. Glean lost four to five months self-running it. The fee for a retained search is trivial against the cost of the months you cannot get back. 12–18 months from now, that 4–5 months could be worth millions of dollars in ARR.

Mind the mandate audit. A “country manager” title with no P&L and no decision rights is a trap, and proven builders smell it instantly. Glean’s role carries real authority — which is precisely why a tier-1 builder said yes.

Invest ahead of revenue. Momentum funds the next layer. The AEs, the solutions engineers and the field marketing come after the leader, not instead of one.

The sequencing trap is the one I see most. Companies hire an account executive to “test” Japan, or run the search in-house to save a fee, and then wonder why eighteen months later they have activity but no anchor. The order is the strategy.

The Bigger Lesson: Glean Will Invest To Win

Strip everything else away and the premise for every AI entrant is the same: get in early, take share while the category is still forming, and cement your place before the incumbents turn up in force. That is not a strategy you can do cheaply. It takes investment — in partners, in a founding leader with a real mandate, and in the patience to fund a market before it pays you back.

Glean is making that bet. A partner since 2023. A founding country manager starting today. A team to be built behind him. The companies that hedge Japan — that wait for proof, that under-fund the first leader, that treat the market as a line item — are the ones that lose it. The ones that come in hard, the way Box did, are the ones still standing when the category settles.

Glean will invest to win. The rest of the AI field would do well to watch how, because most of them are about to face the same choice — and the window is already closing.

If you are weighing a Japan market entry and want it sequenced properly the first time, that is the work we do every day — see our executive search practice and our Why Japan, Why Now brief for how we think about timing the build.

Sources

Murray Clarke