What3words is a simple and genius idea with one of the worst business models of all time

In the midst of the US-Iran war, a clip went viral as a CNBC anchor wrestled with President Trump’s latest threats to wipe out an entire civilization — did his comments present more downside risk or upside risk for investors?

Where’s the humanity?! Late-stage capitalism won’t ever catch me out like that, I thought, as I swiped away to watch a day-in-the-life video of a protein-maxxing 32-year-old man who works in a flavor of finance I hadn’t seen on TikTok before.

Anyway, What3words, a completely free-to-use platform that’s literally saved lives, is really bad because it loses money.

The app, which maps every 3-by-3-meter square in the world onto a unique address made up of three random words is just one of those very good, very simple, and therefore very rare ideas. If you’ve fallen into a hole and need rescuing, you can communicate a precise location, avoiding conversations like “I was on this road somewhere in the middle before I fell.” Finding your friends at a huge stadium, directing a delivery to a specific access road, or sending funny triptychs to your friends all become a breeze with the 57 trillion squares the company has mapped.

Since its founding in 2013, the company’s balance sheet shows cumulative losses of 146 million pounds ($197 million). In 2024, its most recent full year for which we have information, the company did just 2.1 million pounds in revenue... which is a little under $3 million... which is a little under what the typical US McDonald’s restaurant brings in each year (~$4 million).

The McDonald’s restaurant also probably doesn’t spend 8x its revenue on administrative expenses — an outlay that meant an operating loss in 2024 of some 14 million pounds for W3W. In 2022, that was a 46 million-pound loss, for what it’s worth, so it’s progress of a kind.

We don't know exactly how the company did in 2025, as public accounts haven't yet been posted on Companies House — although they shouldn't be far away, with last year's filing appearing on April 17th.

need.more.cash

Now, the company is running another crowdfunding campaign with a post on the popular Crowdcube website showing that What3words is raising at a pre-money valuation just shy of 50 million pounds, with 1,591 investors already taking.the.plunge. That is a huge haircut to the 250 million-pound valuation that The Times reported the company had reached in 2020. It’s also the second time the company has tapped retail investors, the brilliant Louis Ashworth wrote up the first one for FT Alphaville two years ago.

The raise is set to be a mix of primary and secondary equity, giving some employees or investors an exit route from the company’s cap table. In fairness to What3words, revenue did double in 2024, as the company pushed its enterprise product and API access.

Nevertheless, an 80% haircut to its peak valuation, secondary sellers, a lack of institutional buy-in, and a second go at the crowdfunding table, especially for a company that spent 8x its revenue on operating expenses, is a lengthy string of red flags — red bunting, perhaps.

For what it’s worth, I hope I’m wrong. There are genuine reports that the app has saved lives and, though there are plenty of flaws in the algorithm — one user on Reddit pointed out that reworked.sheets.lions and reworked.sheet.lions are within 2 kilometers of each other, but separated by a mountain valley — it would be a shame if the company collapsed altogether.

It maybe just doesn’t make sense as a profit-seeking enterprise.

One thing I do respect is that there are only two very short mentions of AI on the fundraising page — it hasn’t yet gone full Allbirds.