So let me give you all an update on where we’re at with this.
Yes, the merchant feedback system sucks. It’s time-consuming, intensive, and basically hasn’t changed in 2+ years. Stuff goes wrong, merchants get tagged wrong, they get placed in the wrong groups, there’s a bunch of stuff that’s wrong with it. There’s really no way and no point of me sugar-coating it - it’s not up to our standards. Danny isn’t wrong for pointing this out.
There’s two ways we can approach this. One would be to rewrite the merchant feedback system from scratch. In an ideal world that’s what we would do. Come up with better ways for users to help us with this through peer-reviewed changes or a set of “super-users” who earn the privilege to be able to help edit and correct the data (similar to Foursquare/Swarm).
The issue with that right now is that we simply don’t have the available resources right now to do it. We’re in the midst of a HUGE rewrite of our live chat system to bring all of the tooling in-house so that we don’t need to rely on third-party tools. It’s a massive project. Given that there’s some common ground there, it might be possible to have some of the people working on this project look into scoping out a merchant feedback re-write once the grunt work of the current project is done - but obviously that’s not my call to make - there are lots of upcoming engineering projects that I have very little visibility on, or context for the importance of. Other than making it clear that this is something that is steadily causing frustrations for customers, that’s about as much as I’m able to say.
The second thing we can do is simply throw more manpower at merchant feedback. This is something that we are actively looking into at the moment. Now, our hiring tends to happen in big waves. Late last year, we “overhired” - by that I mean, there wasn’t enough work on front-line customer support for all of our customer support team. That was great! Because it meant that folks got time to work on all different things. Some of those were things that are usually low priority - like merchant feedback tasks! Some of them were starting new projects that could turn into something really great. Some people used the time to learn specialisms and have the opportunity to work with different teams. Here on the forum, things were looking pretty rosy and we were able to respond to flags pretty quickly.
Then… because of the “overhiring” we went on a hiring freeze. Lo and behold, with our customer growth, we pretty quickly ended up back to the point where there was almost too much frontline support queries for us to handle, and so all hands went back on deck on being “interactive” (as in, dealing with support, rather than working on other projects).
Now, this is of course our fault as a business. We made incorrect projections about growth and hiring. We underestimated the impact of some new variables and overestimated the impact of others. These are hard things! We also had new people come into the business and start to assess some of these failings so they wouldn’t happen again - notably our COO, Tom Foster-Carter is someone that has made a HUGE impact on the company in the time since he started. So we’ve shuffled, built new teams, built new projections.
Right now, we’re back in the position where we need to hire - and we are doing. I believe we’re hiring maybe 20 new COps this month, or thereabouts. We also have what appear to be pretty robust predictions of how many COps we need to hire for next month, and the month after that, etc etc. So we’re going to be in a position where we simply have more people that can be working on these sort of tasks during some of their work hours.
The other thing that we are actively exploring is whether we can hire people specifically to handle these sort of tasks, who wouldn’t necessarily ever talk to customers on the frontline. With that, there’s lots of different ways we could approach it - we’d need to look at the risks, and the feasibility of that as a role - it might be a less permanent role, done on a flexible basis - perhaps suitable for students who can only work a few hours per week, or something like that? It’s something we’re looking into - do we have them working on anything that has access to any customer data, and if not, is it easier to create such a situation since we may not have to do as rigorous background checks etc?
Essentially - we know, and we’re exploring the best way to combat the issue, and taking all your thoughts on board as well. We’ll get there