Below promise, above expectation

How did the original Denník N investors, who include five of the twenty richest people in Slovakia, react when the company started paying them six-figure dividends?

“I don’t think they even know about it,” Tomáš Bella says. “Whether they get 100,000 or 200,000, I don’t think they notice, to be honest.” One of the more successful ventures in this study distributes money its capital investors may not even notice receiving.

Everything so far in Capital Stacks has been about money going in: who supplied it, in what form, for what stake and at what cost. But money coming out matters just as much, to the people who built these ventures, who have mortgages like everyone else, and to investors, for whom a sector that capital can never leave may not be very appealing.

So this chapter follows the money out, through dividends and exits. To be upfront: remarkably little came out through the channels a conventional investor would expect. Dividends went largely to journalists or into the companies’ own protective structures. Exits happened surprisingly often, but were overwhelmingly financed by the companies themselves, the remaining founders or incoming employees.

The dividend that almost nobody paid

When we asked our cohort if “founders or investors ever take dividends out of the company?” eighteen of twenty-one answered: never.

Where dividends were paid, the purpose was primarily to reward the people producing the journalism or to protect the journalism itself. Financial success mattered because it could give those people, and their newsrooms, greater independence, not because extracting the largest possible return was the point.

Denník N is the sample’s one substantial payer: somewhere between €2 million and €5 million over its lifetime, by Bella’s estimate. Recall from Chapter 3 that the newsroom and its founders started with 49 percent of Denník N; today they own about 80 percent, and the company started paying dividends, in Bella’s words, primarily to pay journalists. The study’s largest dividend stream is mostly deferred newsroom compensation, with the investors’ checks riding along likely unopened.

elDiario.es itself reports that it paid its first dividend from its 2015 profits: a gross €37,556 distributed among around twenty shareholders. Since then, the company has distributed a minority of its cumulative profits to shareholders while retaining most of them as reserves or reinvesting them in the newspaper. By the end of 2024, elDiario.es reported €8.3 million in cumulative net profits since its founding, of which €2.2 million (26 percent) had been distributed to shareholders and €6.1 million (74 percent) retained or reinvested. Founder Ignacio Escolar has explained that dividends serve both to compensate the shareholders who invested their savings to launch the newspaper and to comply with legal limits on indefinitely retaining profits. Capital and labor also substantially overlap: more than 70 percent of the shares in elDiario.es are held by people who work at the newspaper.

Then there is Mediapart, a highly profitable publisher whose corporate design assumes dividends flow upward, as reserves in a holding company whose sole purpose is safeguarding the outlet’s future and a smaller part toward its own press freedom NGO for re-granting.

Dividends went to the people producing the journalism or into structures designed to protect it. Even when money came out, profit remained subordinate to independence.

Three ways of not paying a dividend

There are three good reasons for this near-universal abstinence.

The legal form forbids, or limits, it

Some European jurisdictions offer hybrid structures for organizations that trade commercially but are not supposed to behave like conventional profit-maximizing companies. Germany’s gGmbH is the charitable version of a limited company: it may earn revenue and accumulate surpluses, but those surpluses cannot be distributed to its shareholders. Britain’s Community Interest Company is slightly less restrictive: its asset lock protects the company’s resources for its stated community purpose, while a CIC limited by shares may still distribute up to 35 percent of its profits.

These are general-purpose structures, available across sectors. France has gone one step further and created one specifically for journalism: the entreprise solidaire de presse d’information. A qualifying publisher must devote at least 70 percent of its annual profit to reserves and retained earnings, leaving no more than 30 percent available for distribution; in return, investment in it receives preferential tax treatment. Contexte adopted the status, although remarkably few French publishers have followed it.

Four outlets in the sample are anchored in structures where distributions are entirely off the table: CORRECTIV’s gGmbH and the foundations or associations behind OKO.press, Dossier and Recorder.

The for-profits imposed the restriction on themselves

Across markets with no common legal template, founders operating ordinary commercial companies repeatedly recreated the essential features of these hybrid forms in their own paperwork.

De Correspondent’s founders passed a unanimous shareholder resolution at the outset capping any return on investment at five percent, a cap they have never come close to needing, having distributed nothing at all. Follow the Money, operating as a for-profit in a country without a suitable social-enterprise form, invented its own version: by signed commitment, any profit above five percent remains in the company. Magyar Jeti also maintained a formal no-dividend policy for several years, despite being a plc. The legal wrapper remained commercial; the distribution rules did not.

Sometimes the restriction came not from the founders but from their funders. Grantmakers willing to support for-profit newsrooms face an obvious problem: a non-repayable grant should finance journalism, not subsidize a distribution to shareholders. Some therefore make support conditional on a temporary no-dividend undertaking. Civitates, one of Europe’s most important philanthropic supporters of independent journalism, requires for-profit recipients to reinvest their surplus and has used grant agreements under which no dividends may be paid during the funding period. This is not a permanent mission lock—the restriction expires with the grant—but while it operates, the financial effect is much the same.

Even where no law, statute or grant agreement prevents a distribution, the optics and internal logic of combining philanthropic money with shareholder payouts often do. Jan-Willem Sanders of Follow the Money told us: “If we get grants, it’s a little bit strange to give yourself the dividend.”

Growth logic

Finally, some outlets faced no prohibition at all; their owners simply regarded dividends as an inferior use of scarce capital while the company was still growing. As Zetland’s Tav Klitgaard said: “Dividends are a technical financial instrument … our investors have always thought it’s better to keep the money—let’s invest it in Zetland instead of getting it out.” Even during its recent ownership transaction, becoming a dividend-paying company was considered and rejected as belonging to a different stage of life.

These mechanisms are fundamentally different. A permanent asset lock, a shareholder resolution, a temporary grant covenant and an annual decision to reinvest do not create the same rights or protections. But throughout the period covered by this study, they produced much the same result: profit mostly remained inside the newsrooms.

“Give us money, and trust us”

Almost nobody promised their investors a number. In the entire sample, three outlets have any projected rate of return on record. Jan-Willem Sanders of Follow the Money described its actual pitch: “give us money, and trust us, because the work is important.”

Zetland was one outlet that did have an annual IRR projected. When we asked its CEO, Tav Klitgaard, whether they delivered on those promises, he told us it was below promise, still above the expectations of the investors. When we repeated the phrase to Telegram CEO Miran Pavić weeks later, he adopted it on the spot.

Denník N’s investors did not initially assume they would recover their capital. Follow the Money’s investors were somewhat surprised not to lose theirs. Tsüri’s early backers hoped, in the best case, to get back what they put in. Some Republik supporters deliberately waived both financial return and voting rights.

The fact that some of these investors thought financial success was an unlikely scenario created its own set of difficulties later.

Shareholder agreements were often designed to protect the outlet from failure or interference, not to govern a profitable company ten years later. When investors eventually wanted to leave, there was no obvious buyer, valuation mechanism or exit process. The capital had been patient partly because nobody had seriously planned for its return.

Exits: The market with one buyer

Sixteen of the twenty-one outlets report at least one founder or investor exit, a somewhat surprising number for a sector nobody associates with liquidity. But in almost every case, the buyer was the company itself, the remaining founders, or incoming employees. With one exception, no outsider has ever bought in through an exit.

The buyback. Chapter 3 ended with what owning Denník N cost its investors: the calls, the press conferences, the pressure on a software company over an investment it never quite made. Here is how that story ended: three exhausted investors asked to leave, and the company bought them out from cash flow, paying over roughly three years. Investors originally suggested donating their shares to the newsroom, but in the end everyone agreed to go with an independent valuation to reduce the risk of the government challenging the transaction. It ran to hundreds of thousands of euros, and a return in the 3–5× band. So a few months after launch, the investors held 51 percent and the newsroom 49; today it is about 20 percent investors, 80 percent newsroom.

Contexte has turned exits into part of its ownership system. It has completed roughly €2 million in cumulative share sales, at an average multiple Boulanger puts at four to five times. The record was 9×, achieved by an angel investor who now regrets putting in only €10,000.

These exits have fueled two parallel processes. First, they supply the annual internal stock market described in Chapter 7. Second, they have allowed Boulanger to increase his own stake from 52 to 77 percent, often by acting as the buyer of last resort when no employee was ready to take the shares.

But shares do not always move simply because everyone agrees that they should.

Several outlets described how surprisingly difficult it was to unwind stakes held by people who had joined at the beginning, contributing money, expertise or both, but no longer played an active role in the company. These were not hostile or extractive investors. They were friends, early supporters and people who remained strongly aligned with the mission. That could make the conversation harder rather than easier: there was no dispute forcing a resolution and no obvious reason, from the shareholder’s perspective, to leave.

For the outlets, however, the dormant stakes complicated what they wanted to become, and what they wanted to be able to say about themselves. A company cannot quite claim to be entirely owned by its workers, newsroom or current team while even a small piece remains with somebody outside it.

At two outlets, persuading such a shareholder to leave took six and seven years respectively. One involved an external early investor who was perfectly content to remain; the other, a co-founder who had stepped away for health reasons. Both relationships remained friendly, and one of the eventual transactions was worth less than €100,000. The obstacle was not primarily the money. It was the slow work of convincing someone to surrender a legitimate stake so that the people still building the company could clean up its cap table and finally say: we own ourselves.

Chapter 3 warned that a stake granted cheaply at the outset can prove expensive to unwind. Sometimes the largest cost is not cash, but years.

The nominal exit. At the bootstrap end, exits happen, people move on, life intervenes. Recorder bought out a departing co-founder for a few hundred euros, a price agreed over a beer and since forgotten by everyone involved. German nonprofit law fixed the price when a CORRECTIV partner left: shares transferred at nominal value, about €25. Krautreporter’s exiting coop-members practically get back what they put in.

The exit to end all exits. The sample’s largest transaction was engineered to make itself unrepeatable. In 2019 Mediapart bought out every investor at a €16.3 million company valuation: €1 million of shareholding donated by its owners, €4.4 million paid from accumulated profits, €5.5 million from a bank loan and €5.4 million from seller credits agreed by the founders and one of the historical shareholders. The entire company was transferred to a holding structure modeled on The Guardian’s Scott Trust, whose statutes forbid the stake ever being sold. Early backers realized returns in the 3–5× range; the company purchased its own permanence. Krautreporter performed a similar ritual in 2016 even if at a smaller scale: its three founders transferred their GmbH shares into the cooperative and got coop-shares in return.

The exit that wasn’t. In 2022, an unnamed international competitor offered €12 million to buy Contexte, then generating roughly €6 million in annual recurring revenue at the time: a clean 2× ARR valuation. The offer came with an edge. Sell, the message ran, or we will put that money into building or backing a competitor in your market. For Boulanger, who owned 65 percent of the company at the time, accepting would have meant more than €6 million on paper. He did not dismiss it lightly. He went away to think for a few days and then came back and said no. He did not even try to negotiate the price upward.

The one that got away. The sample’s single conventional exit is Zetland, whose existing investors sold their majority stake to Bonnier News, the privately held Swedish publishing giant. The price is undisclosed and under NDA. (Will Media’s combination with Chora is better read as a merger of two Italian podcast publishers than as an exit, though the original backers received a share of the purchase price.) One trade sale in twenty-one companies over fifteen years is not really a very busy market, if we can call it a market at all.

Exits by force majeure. Politics closed one more position: as Chapter 9 recounts, Magyar Jeti bought out MDIF in 2025 as a foreign-agent law loomed; ownership acquired partly as protection became a liability. Denník N’s buyback half belongs here too. In this sector, politics can write itself onto the cap table in both directions.

What it all returned

Pull the known multiples together and the picture is unglamorous and rather dignified: one at 1×, two each at 1.1–1.5× and above 5×, and a cluster of four at 3–5×, on small bases, over roughly a decade, paid almost entirely by the companies themselves out of profits.

Nobody got venture returns. But as Chapter 3 established, nobody was promised any; most of this capital was wired by people who assumed it was gone. Mission-driven money in European independent journalism did not set itself on fire. It mostly came back, with a multiple, from the company’s own till, years later, when everyone was ready.