Why I'm building another Tippspiel

Every August the same thing happens in German offices, family WhatsApp groups and Vereinsheime. Somebody sets up a Tippspiel, everyone predicts the Bundesliga scores, and a leaderboard runs quietly in the background until May. It is a small ritual and people are weirdly attached to it.

Most of those rounds run on Kicktipp. It was started in 1995 by Janning Vygen, who wanted to stay in touch with friends after they scattered around the country, and it is still run by two people. Roughly two million people use it every weekend. During the EURO in 2024 that went up to about six million. As far as I can tell from the reporting, it earns somewhere around €1.5 million a year on ads and a light premium tier, with no marketing budget at all.

That is a very good business, run by two people, in a category I now want to compete in. So the question I get, reasonably, is why bother.

The market is smaller than it looks, and that’s the point

Let me be honest about the size first, because it changes everything downstream.

If Tipperoo became the default Tippspiel in Germany, replacing the incumbent outright, the prize is roughly that same €1.5 million a year. Total dominance of this category gets you a lean, profitable niche business and nothing like a unicorn, and Kicktipp has already proved you can run it with two people.

Most people would read that and stop. I read it and found it clarifying, for two reasons.

The first is that it rules out the strategy that kills most challengers here. You cannot raise money, buy growth and out-spend anyone, because the revenue at the end doesn’t justify it. So the only viable posture is the one Kicktipp already demonstrates: stay tiny, keep costs near zero, and let the product’s own invite loop do the work. Being small is the configuration that wins here, which is a strange thing to have to plan for.

The second is that it points at where the money actually is. Consumer willingness to pay for a prediction game is close to nothing, and always will be. Workplace willingness to pay is much higher, and it’s the lane Kicktipp monetises weakly and the modern consumer apps ignore entirely.

Kicktipp wins on habit, not on quality

This part matters, because it would be easy to be arrogant about it and arrogance here would be wrong. Kicktipp’s feature set is genuinely deep. Freely configurable scoring, the Quotenregel, jokers, bonus questions, mixed-league calendars, team rankings for company events, multi-sport, livescores. If I tried to out-configure them I would lose, and it would take years to lose.

What Kicktipp is not is modern. The web UI is dated and doesn’t respond properly on a phone, which is where most people now tip. There are ads. There’s betting-odds adjacency in the wider ecosystem. The navigation gets called unübersichtlich often enough that it’s basically consensus, and Kicktipp itself concedes some of the admin screens “leider sind einige Masken etwas komplizierter geworden.”

So it wins on incumbency and habit rather than on quality. That’s a real thing to win on. Thirty years of group-invite loops and accumulated SEO is a genuine moat, and I’m not pretending otherwise.

There’s also a cautionary tale sitting right there. HALBZEIT was a better-built version of roughly this idea, and it died in 2024. Not because the product was bad. It died because there was no growth engine and the runway ran out. That’s the failure mode I’m most likely to repeat, and I think about it more than I think about any technical problem.

How you actually beat something like that

The framework I keep coming back to is counterpositioning, from Hamilton Helmer’s 7 Powers. The idea is that a challenger adopts a model the incumbent can see perfectly clearly, understands completely, and still won’t copy, because copying it would damage its own business. It’s one of the few durable advantages available to someone with no capital, because it rewards timing and position rather than spend.

Tipperoo is pointed at three places where I think Kicktipp is stuck:

No ads and no betting odds. Kicktipp’s web revenue is ads plus betting adjacency. That’s a meaningful part of how two people net €1.5 million. Going ad-free and betting-free would cannibalise their own income, so they won’t. I can, because I never built that revenue line in the first place. They are not asleep here, they are stuck, and stuck is a much better thing to bet against than lazy.

A modern, mobile-first rebuild. Kicktipp’s founder says the product “lives from its simplicity,” and there’s thirty years of refined code and user muscle memory behind it. A ground-up rebuild risks both. Incumbents defend with incremental micro-innovations, not resets, so that ground stays open more or less indefinitely.

Workplace-native. Kicktipp’s pro package is a website-integration product sold to marketers. Rebuilding around bottom-up, self-serve, Slack and Teams adoption is a different business, and there’s no reason for them to disrupt themselves for it.

The framework also predicts how the incumbent responds, which is to ignore the small niche player for as long as possible. That happens to be exactly the cover a challenger wants.

What I’m trying to get right

No ads, no odds, no entry fees. Permanently. Not as a launch promise I quietly walk back once there’s traffic, since the whole counterposition rests on it. The moment I run ads I’ve become a worse-funded Kicktipp and the reason for anyone to switch evaporates. The Tippkasse, if a group runs one, is tracked and never touched by us.

The simple core stays sacred. This is the failure I’m most worried about, because it’s the one that comes from doing my own idea too enthusiastically. The risk of counterpositioning hard on “modern” is that you end up with something so featureful that a casual player can’t just type two numbers and get on with their day. Kicktipp’s simplicity is the reason the category works at all. Depth in Tipperoo is opt-in, never imposed. If your uncle can’t use it, it’s broken.

Being right, publicly, on a Saturday afternoon. A wrong table destroys trust instantly and permanently. Six matches finish at once and everyone is refreshing. So a score stays provisional until two independent sources agree, results live in an append-only log so a correction recomputes the standings transparently instead of quietly rewriting them, and the UI is allowed to say “we don’t know yet” rather than show a confident wrong number. That’s a whole post of its own, and I’ll write it once the beta has had a chance to embarrass me.

Hitting the one window. Points don’t transfer between platforms and groups resist switching mid-season, so realistically there is exactly one moment a year when a whole round will tolerate moving: the start of the season. That’s why the beta rides on 2. Bundesliga from 7 August, and why GA is targeted at 18 August rather than whenever it feels ready. Miss it and the answer is “ask us again next August.”

Where this could go wrong

The likeliest way this ends badly has nothing to do with Kicktipp crushing me. I build something good, nobody ever hears about it, and it quietly folds, which is precisely what happened to HALBZEIT. Distribution decides this rather than the code, and the code is the part I’m naturally better at, which is roughly where I’d expect the whole thing to come apart if it does.

Second most likely: the season window opens, the product isn’t ready, and I ship it anyway because the date is emotionally load-bearing. I’d rather move the launch than launch something that gets a table wrong in front of somebody’s colleagues.

Ask me in September how much of this survived contact with reality.