Our pricing in EU countries where we've launched (UK + Ireland) starts at 2.4% + 20p (we've revised pricing for each country when we launch out of beta). We regularly give volume pricing well below 1.8% - so at both ends, considerably cheaper than the range quoted above.

Also, which country are you in? Stripe is available in full production-ready beta in France, Germany, Spain, Luxembourg, The Netherlands, Sweden and Finland, with more countries in the coming weeks - keep an eye on https://stripe.com/global for updates.

You can sign up and start using Stripe to process live transactions instantly in all our beta countries - this is a key requirement for us as we expand globally.

Will merchants be instantly approved, or does it take several days/weeks?

Great write-up - I hadn't seen that graphic before but it's great - full credit to the original source http://startingandsustaining.com/.

As a Stripe-r, I can't complain about the OP's conclusion in any way :) As stated they are based in Denmark, and Stripe is not currently available for Danish businesses - that's on us to fix, and we're certainly working on it.

However, re: the title "How to choose a payment provider for your Europe-based SaaS startup" - the situation in Denmark is not representative of Europe as a whole.

Stripe has now launched here in the UK[1], and we now have betas available in Ireland, France, the Netherlands and Belgium - with the latest released in the past few weeks. So if you're based on one of those countries, please do give us a look. And if you're based elsewhere, well - I guess we need to keep up the pace and hurry to your country ;)

[1] https://stripe.com/blog/introducing-stripe-uk

As other commenters have mentioned, it's really easy to make mistakes comparing complex pricing across different providers. With Stripe and PAYMILL, the fees quoted are all you pay. With Braintree's interchange+ pricing, they actually state themselves that "Total costs are typically 1.8% to 2.6% of the transaction. There is a minimum cost of €100 per month"[2]

[2] https://www.braintreepayments.com/faq#pricing-question

And that means it can't be as low as 1.8% of your transaction unless you're consistently taking in the region of €5,500+ per month -- low revenues for an established business, but not a trivial amount for a bootstrapped start-up looking for a payment service to launch with.

I've wondered whether this is a deliberate business decision by Braintree to discourage applications from brand new (and I assume on average more risky) start-ups. Then again, established businesses with serious revenues would presumably be considering a more traditional set-up where they can negotiate much lower rates with heavyweight payment services, and would perhaps care less about the hassle of setting those things up compared to the ease-of-use for developers of modern payment services. I'm not sure which part of the market Braintree are really trying to own at the moment: they do seem to have a USP among the "developer-friendly services" in the number of different payment methods they support through a common API, but this doesn't even seem to merit a mention on the front page of their site.

(I bootstrapped my last startup and 100 euros is roughly what I was paying SEOMoz, my accountant, etc. on a monthly basis)

It's way more important to just get something working and out-the-door and grow the number of customers then to spend a lot of time over what comes down to a relative small amount of money, you can always re-negotiate the fees when you grow and the absolute amount becomes meaningful.

Right, so if you're in that position, why would you put up with the onerous application process for a service like Braintree if you have a simpler alternative like Stripe available and it provides the functionality you need? You want a payment service that takes as little time as possible to set up and then just works, because you have a million more important things to be doing. Any advantages Braintree might have had if their pricing did work out more favourable than Stripe's flat rates is instantly lost because they can't give straight answers to too many basic questions (like "What does it cost me to collect a payment?") and their API and documentation are significantly more complicated.

http://www.thecompleteuniversityguide.co.uk/league-tables/ra...

That first term consisted of multiple choice exams in C and HTML, where one of the questions didn't have a correct answer.

In short, I sadly don't think that any lines can be drawn between 17% unemployment of CS graduates and a supply of great available talent.

Speaking as someone who's processed 200+ applications and hired 7 engineers in London the past 2 years, sharing stories with my colleagues at other companies - great talent is very hard to find.

It takes a certain mindset to write code. Not everyone has it or is interested in it. My children couldn't care less about it and that's up to them. They'd rather be digging holes in the garden and covering things in paint. I'm not here to indoctrinate them with my own interests but to nurture their interests.

Education is fundamentally flawed on the basis that it mandates knowledge on political whim and percieved societal need rather than nurturing and developing interest. Consequentially society is filled with people doing what they are bad at and hate.

And yet I'm glad that, when I was a kid, people encouraged me to plant things and paint things.

Just because learning to code is societally approved doesn't mean that kids shouldn't be encouraged to practice it for fun. Trust me, if the kids decide they hate it they'll figure out how to avoid it as much as they can in the future. They're good at that.

Mind you, your attitude that you shouldn't expect your kids to take after you is a healthy one, and I don't want to discourage it. But don't take it too far. It's good that people are trying to figure out better ways to teach kids about programming.

It's very possible that I may not be a programmer today if my best friend's dad hadn't pulled out an old Apple ][gs and had us hack away at it for an afternoon.

We found a demo floppy that had a BASIC programming tutorial on it, and from there I got hooked.

It takes a certain talent to write or play music, or paint, or do science. We encourage kids to try all these things, most won't pursuit them, some will. The kids choose without coercion.

How many parents encourage kids to learn how to code compared to the number of parents that encourage their children to do the above mentioned things? It is the children who decide what they want to do, it's about offering them a broader spectrum of things to choose from.

It takes a certain mindset to make a living as an author, and my kids might not have that. But I'm damned if I'll let them grow up not knowing how to read and write.

Indeed it has many problems. The first problem is that the Mathematics-Informatics profile in high school is considered elite, and every parent sends its children there. The effect is that the class will have the best, most intelligent students, but not necessarily the students interested in the curriculum. Neither the children or the parents know anything about the curriculum, all they know is that all the smart kids go there.

I've been in this class in one of the top-5 colleges, nation-wide. Some poll in the final year yielded 80% of the students hating both math and programming. Most preferred literature. Because of the inherent bias in Romanian education, the highest graded students will always be the ones who like liberal arts, and not the ones with a mathematical background and because of the way enlistment to high school works, these will be the ones that end in this class.

I mentioned I was in one of the top 5 classes nationwide to illustrate the second problem. Even though I was in one of the best classes, professors that taught programming were execrable, both in their talent as educators and in their talent as programmers. Incompetent is a word that's too mild to describe them. Reasons are easy to guess, high school teachers are paid poorly and anyone with any talent in CompSci will find a better job elsewhere. It's funny that the only two skilled professors that taught programming were two guys that did something completely different and only taught at high school because they enjoyed teaching and working with young students.

The third problem is that the curriculum is very abstract, way too abstract and far from reality, and very old, nothing feels like today, students are still required to write DOS applications with a DOS editor.

I know I would have loved programming in my younger years, just didn't have too much access at the time. That's at least my two cents :-)

Is it something that most people really can't do? If so, that scares the hell out of me. And not because they can't write computer programs, either. My concern is, how sloppy is their thinking in general? How easily do they fall for circular arguments, or succumb to "lost purposes", where what they do no longer has any connection to their goals?

I'm all for providing exposure to a variety of things in school and at home, coding included, but we need to let our children be themselves. Give them exposure, let them taste what the world has to offer, but don't compel them to be the über-coder that you wanted to be.

Include group work and gentle supervision.