Hakuhodo DY Holdings - Company History

In October 1895 Seki Hironao opened an advertising brokerage in a room in Kanda and called it Hakuhodo, from his own phrase about serving customers broadly. What he brokered was not newspapers but education magazines — publications read by schools and teachers. Their circulations were small, but their readership had a sharp outline, which made them precisely targetable for an advertiser. Choosing media by the shape of the audience rather than the size of the print run was already the company’s idea at its first day of trading, and refusing to chase the leader on volume was already its posture.

Hakuhodo was in fact early: Dentsu, which became its permanent rival, traces itself to Mitsunaga Hoshiro’s company of 1901, six years later. Being first bought no advantage. Dentsu ran a news agency alongside its ad business, which bound it tightly to the newspapers, and it led the four mass media in billings for decades after the war. What the early start gave Hakuhodo was a different asset entirely — deep relationships with advertisers and publishers in education and publishing. The structural question of the trade — whether you stand on the advertiser’s side securing space, or on the publisher’s side selling it — Hakuhodo answered by going deep into one category’s media rather than broad across all of them.

The firm moved to Nishikicho in 1914, incorporated in February 1924 with Seki as its first president, and put up its own building in 1930 — though as a legal entity it traded for decades under the name of a news agency, with “Hakuhodo” as the advertising division. Wartime paper rationing and the forced consolidation of media shrank the business to little. What it rebuilt first, in 1948, was a house journal arguing for the usefulness of advertising itself, later retitled Kokoku. Only in April 1955 — sixty years after its founding — did the corporate name and the trading name finally become the same thing: Hakuhodo Inc.

In June 1957 Hakuhodo reorganized its sales force around clients rather than media and imported the American concept of the account executive. Under the old arrangement, staff sat by medium and an advertiser faced a scattered set of contacts; under the new one, a single person owned the client’s problem and pulled creative and media people to it from inside the firm. It was an attempt to differentiate on the design of the sales organization itself, since it could not differentiate on billings — and it is the foundation everything the company later sold, up to and including its 2020s consulting business, was built on: organize the firm around the client’s problem, not the ad slot.

That posture was then repeatedly declared in words. The Hakuhodo Declaration of 1960 staked out a position other than volume at the very moment the growth years were inflating advertising demand; “Grand Design Partner” followed in 1991 under Isobe Ritsuo, and “Power Brand Partner” in 2002. All three say the same thing — that the company intends to be involved in the client’s business design rather than merely execute its advertising — and none of them sets a billings target.

The most durable of these investments was the Hakuhodo Institute of Life and Living, founded in 1981 to look at people as seikatsusha — people living whole lives — rather than as consumers, and running a biennial attitude survey since 1992. Research of that kind earns nothing directly, but the longer it accumulates the less it can be copied; in 2026 the company’s president still named it first when asked what separates Hakuhodo from its rival. Creative reputation ran the same way: Cannes Grand Prix wins for Matsushita in 1982 and Nissin in 1993, a special award at the festival’s 50th in 2003 — and BMW moving its account from Dentsu to Hakuhodo in 1992. The company stayed second in size and first, often, in regard. Meanwhile a founder’s bequest hardened into capital structure: the education foundation set up on the 75th anniversary in 1970 is today the group’s largest shareholder, holding 19.77% as of March 2026, ahead of an employee-related association at 5.03%, Asahi Shimbun at 3.12% and Nippon TV at 2.40%.

Hakuhodo turned 100 in October 1995, and the market turned against it almost immediately. Total advertising spending peaked in 2000 and then fell as the internet bubble burst and the one-off demand from financial-sector restructuring disappeared. The damage was not evenly spread: Dentsu held both share and gross margin, Hakuhodo gained share while losing margin, and Daiko lost share outright. Everyone below the leader was in trouble at once, and the profit gap had widened well beyond what moving up one rank could fix.

So they combined. A 2001 business alliance became a merger within two years: Hakuhodo, Daiko and Yomiko — the industry’s second, fifth and sixth — announced a joint holding company in December 2002 and established Hakuhodo DY Holdings by share transfer on 1 October 2003. Their combined standalone revenue for the year to March 2003 was ¥985.4 billion against Dentsu’s ¥1,367.6 billion; Hakuhodo alone had been ¥709.7 billion. In one step the gap to the leader went from roughly half to just over seven-tenths.

The design of the merger was unusual, and it is the point of the whole episode: in December 2003 the three firms carved out only their media and content organizations into a wholly owned Hakuhodo DY Media Partners. Media buying and development were pooled; sales and creative stayed separate in all three companies. This preserved the international convention that one agency does not handle competing advertisers in the same industry, and kept the three competing internally. Contemporary critics read it as defensive rather than expansionist — with integration limited to media, the cost that could be taken out was limited from the start, and four-media procurement of about ¥700 billion still ran at only around seven-tenths of Dentsu’s. But the choice protected the client base it might otherwise have destroyed, and it stood for twenty-two years.

The listed company delivered on its stated numbers. It joined the Tokyo Stock Exchange first section in February 2005, passed ¥1 trillion in revenue that same year and reached ¥1,118.7 billion by the year to March 2008 — beating the ¥1.1 trillion target it had set at the time of the merger. Then the advertising market broke: under Toda Hirokazu, who became president in 2006, the year to March 2009 produced a net loss of $35.3M (¥3bn), the first since the merger, as a head-office relocation coincided with a collapse in television and newspaper demand. Toda’s response was to point past advertising altogether, at the roughly ¥2 trillion of sales-promotion money sitting in a second corporate pocket.

Digital was acquired rather than grown. In February 2009 Hakuhodo took a placement in Digital Advertising Consortium and made it a subsidiary — an admission that its own sales force would never prioritize internet advertising, where a job might be worth a fraction of a television campaign and the individual salesperson’s incentives pointed elsewhere. Rather than fight that logic, the group bought a specialist and let it run. The consolidation then proceeded at a deliberate crawl: a joint holding company with Irep in 2016, and only in October 2018 a tender offer taking it to 100%, at a 40% premium that cost $1.0B (¥113bn), most of it funded by long-term debt. Nine years from first stake to full ownership; by April 2025 the DAC name was gone, merged into Hakuhodo DY ONE.

The same instinct shaped the overseas build. Instead of exporting its own brand, in 2014 the group formed kyu, a strategic collective, and began buying: two American firms at the outset, 30% of the design consultancy IDEO in 2016 with an option over a majority, and by 2019 ten companies and more than 2,000 people, mostly in the West. What it assembled abroad was not advertising billings but design and business-transformation capability. At home it did the same by joint venture, pairing with NTT Data on consulting and with an app developer in 2025 — filling in the systems and software side where, for all its accumulated data on how people live, it could not match the global consultancies. Dentsu was making the same move at the same time.

Underneath, results improved and then were spent on repair. The year to March 2022 produced record ordinary profit of ¥75.7 billion; by the year to March 2026, gross profit — now the group’s headline measure — reached ¥406.0 billion at a 25.7% margin, up 1.1 points, while the year to March 2025 fell to ¥10.8 billion of net profit under the cost of early retirements and back-office consolidation. The structure was rebuilt in parallel. In April 2025 the media company carved out in 2003 was absorbed back into Hakuhodo, folding the merger’s defining design in the opposite direction after twenty-two years. In December 2025 the group won control of Digital Holdings — losing the price contest at ¥2,015 a share against a ¥2,450 rival bid, but winning anyway by cutting its minimum acceptance threshold and pre-arranging the founder side’s stake.

In June 2025 Nishiyama Yasuo became president, adding the CEO title in April 2026 as Hakuhodo and Hakuhodo DY ONE moved to unified operation. He concedes that generative AI’s automation and clients’ in-housing both question what an agency is for, and argues that drawing out what a client or a broadcaster actually needs is the part machines cannot take. Then he said the thing the company had avoided saying for a century: that it was no longer in a position to be described as “a great number two.”