Maybe the french stuff was more useful in practice, the german always seemed scammy, somehow.
It was highly pushed by the national telco. The telco lent terminals for free so they were easily available, and there were tons of useful services, especially but not only from national companies: starting from the 80s you had stocks, online shopping, travel reservations, information services, message boards, databases[0], games, dating sites, ...
Monetisation / payment was integrated from the start, through the phone bill: just like premium-rate phone numbers, minitel had premium-rate services (both first and third party), so you could make money out of valuable services. It was nothing compared to the modern web (~25000 services and about $1bn revenue at its height), but for the mid-80s to mid-90s and out of a population of 60m it was quite massive.
[0] trying to get rid of phonebooks was a big reason for minitel in the first place
De Gaulle had neglected the country's phone network and it was in a piteous state by the 1960s. A popular comedy sketch, "Le 22 à Asnières" by Fernand Raynaud had the hapless protagonist trying to call number 22 at Asnières and it was such an ordeal he only managed to get connected by a telephone operator in New York...
During the 70s, specially under the technocratic president Valéry Giscard d'Estaing, the French government invested massively to catch up and for several years the Directorate-General of Telecommunications was the single largest government department by budget. They bet massively on digital switching and created huge telco gear vendors in Alcatel. Towards the end of the 70s that massive surge of investment had started to dwindle and the government was trying to figure out what could be the next wave to foster the telecommunications equipment industry, and Minitel was the answer.
It didn't export well. There was a trial in the US with US West, but it didn't pan out. My former boss Jean Lebrun was one of the key people on the electronic phone directory project, and at the time it was the largest real-time database in the world. It was essentially a distributed in-memory database build on 1980s computer technology, and as you can imagine very expensive. Fun story: Oracle tried to get them to evaluate their RDBMS for the project, and even gave them access to the source code, but it was found to be inadequate.