We spoke to co-founder John Hancock about the project and discovered some of the key benefits of working in an accelerator, the problem with current payment systems and how to build something huge in a matter of days.
Idealog: So what is the problem you are looking to solve here? Why do we need a National Account Registry?
John Hancock: What we are trying to do is look at real problems in business. Our focus is very much one of trying to make life easier - for small businesses in particular.
There were two problems we were focused on. One of them is this really banal problem that when you're trying to pay someone over the internet you have absolutely no idea - even if you get the account number right - whether the money is actually going to go to the person you want it to. That results in people checking, rechecking and re-re-checking the number quite unnecessarily, just to make absolutely sure they haven't made a mistake when they're transferring money.
Then the more subtle banking problem, one that many people aren't even aware of, is this problem of pay e-fraud - the phenomenon where some malicious person intercepts an emailed invoice, manipulates the account number in the invoice, then forwards the email on as if it had come from the original person. It’s quite a well-established form of fraud in some countries. Those were our two problem scenarios.
Paying people is a pain in the backside and it shouldn't be. It should actually be really easy. That was the hypothesis we started with.
How long did you spend on research?
Not ages, but we did quite a lot of validation.
I suppose the advantage we had was that we had access to Xero people who obviously do heaps of research on the needs and wants of small businesses. We had access to a lot of the internal people at Kiwibank who were very, very good at helping us understand the detailed mechanics of how business payments are actually supported in new Zealand.
All the people we spoke to, their main experience was much more based on their personal experience doing internet banking as an individual rather than as a business. So if you're trying to pay for something expensive with internet banking a lot of people almost get into a rabbit-in-the-headlights paranoia about getting the account number wrong. You type in the number and you really don't have any idea if the number is the right one for the person you're trying to pay, but you're potentially sending thousands of dollars. That's not a very comfortable experience.
That represents a pretty big opportunity for you
There are millions of small business payments happening around the country every year. Even when you're only talking about the scenarios where someone is trying to pay someone new, the productivity benefit, even if you only saved a minute every time a new payee was set up, you're talking thousands of man-days a year. That’s absolutely staggering for such a trivial friction in the process.
Rather serendipitously, the New Zealand Police have recently released a report looking at this type of fraud in New Zealand and it's very difficult for them to estimate how much of this fraud goes on because a lot of cyber crime is never reported. In some cases people are completely unaware they've been defrauded and in other cases they are slightly embarrassed - it's not a great look to report a crime like that. The police estimate that in the last six months New Zealand businesses have been defrauded by between five and ten million dollars. And it's completely unnecessary. Banking should be really easy.
So once you had your hypothesis, the actual development was really rapid, right?
Extraordinarily so. And that's really due to the power of the public cloud these days.
One of our supporters in the project was Amazon Web Services. They have a very, very large, secure, scalable and powerful public cloud deployed globally, so we simply used orthodox technical architectures and orthodox tools that we subscribe to from them. The actual development time was less than two weeks...
...and you can see the scale and sophistication of the solution we built - it's absolutely mind blowing.
This is the sort of thing that used to take years to produce and used to cost millions of dollars. When you did this in the past you used to solve exactly the same problems from scratch every time. What you’re doing now is avoiding all that by exploiting the power of public cloud resources.
Given that these resources are available now, what does that mean for a small country at the bottom of the world like New Zealand? Has that tyranny of distance truly been overcome now?
That's absolutely our proposition.
So what have you got from going through the accelerator process that you couldn't have got from elsewhere?
Creative HQ can support small businesses from the point of being a good idea to the point of having a minimum viable product that can be pitched to investors for seed funding.
What we were doing was not exactly a perfect fit for that, but Creative HQ did a very good job of adapting their resources to suit the kind of problem we were trying to solve.
The things that worked very, very well for us were having a structured environment around achieving a certain outcome within a ridiculously short amount of time, having business analysts to tap into and having access to mentors from the accelerator program - senior people with a lot of experience with open financial services of the sort we were experimenting with.
You’ve certainly achieved a lot in a short time.
Even Xero and Kiwibank, who are both quite nimble, agile startup-y-type organisations, even they acknowledged that it would have been quite difficult to do an internal project that achieved what we achieved in the time we did it.
But it makes sense. If you take people out of the office, give them the resources they need, support them and give them a really clear target - and you give them all the information and support they need to be successful - you can actually get quite a lot done in quite a short period of time.