Interview: Todd Green talks about Candy Crush Saga over the last decade and the upcoming 15,000th level

Candy Crush Saga is one of the most recognisable games in the world, let alone on mobile, so when King started celebrating its 20th anniversary, it only made sense we'd spend some time speaking about the company's most iconic game.

Soon, the casual puzzler will add its 15,000th level, which has been created by King's latest design recruits, a long-held tradition. But it's not a case of simply making a level and moving on. The team is constantly tweaking the existing lineup in a bid to get each stage as close to perfect as possible.

During our recent trip to King's London office, I spoke to Todd Green, General Manager of Candy Crush about this approach, alongside discussing how the game has remained popular over the last decade.

My name's Todd Green. I'm the General Manager of the Candy Crush franchise. That means I'm spending a lot of time working on the Candy games, but most of all Candy Crush Saga and Candy Crush Soda Saga. I've been at King for nearly 11 years, and am working about half and half between London and Stockholm.

That's a good question, actually. I don't think anyone's ever asked me that. I suppose if we would choose to have any competitors, we'd rather they were games that we make as well. So, do we see them as competitors?

In the end, we're trying to build King as a long-standing success and to do that, we have multiple different games, and sometimes people move between the games. We spend a lot more time thinking about how we attract players or retain players who might be interested in other people's games.

We've tried to—I don't know if one can say specialised because the games are deliberately very mass appeal, but I think where we've got most of our existing games in this space of casual free-to-play matching type games. And what we've found over the years is that there are some overlaps, but they're not as big as one might think.

You can have players who played Pet Rescue, Bubble Witch or Farm for years and years and years. And they probably tried Candy at some point, but a different mixture of components works better for them in those games. So I think we see them as friendly add-ons and kind of compliments to one another.

Well, there were a lot of games released around the same time that probably haven't sustained quite so strongly. So, timing matters, for sure. But, I think that, as each year goes by, that becomes less helpful and relevant for us. It's much more the work that we've done since then. I would say there are probably a couple of things.

First is that we feel like every day we can improve the game. So, each day that goes past is an opportunity for us to make the game better somehow – to meet the needs of the players. That's true not only in product but also in marketing, how we think about talking about the games. Also, how do we build the organisation – who do we hire and train?

I've been here 11 years. I try to embody the philosophy that this is a marathon, not a sprint. You take one step forward and don't stop - you're going to get to the end eventually. And that's how we think about it. We can take one step forward every day, and that's great. The games get better.

• Steve Collins, Chief Technology Officer at King, discusses how the company is implementing AI tech

The other thing is we try really hard to hire people who are really interested in what we're doing, who care about the long term, care about the players. One of the earliest experiences I had at King when I joined at the beginning of 2013 was we were just on the cusp of scaling up the team. Within a couple of years, we scaled up to around 2000, and I think I did 200 interviews in my first year. The ones towards the end of that year were better, definitely.

The point being, though, that we were very strict about how we did the hiring. I think that's kind of stayed with us over time – trying to hire as broad as we can. All the different range of skills that we need but investing a lot of time in that hiring process. And that pays back over and over. You get somebody who's really good or understands how to work with other people then that helps you today and next month, next year and beyond.

There are two types of improvements. One is to move the ball forward very slightly, and then we try to do these leaps as well – introducing a new system or a radical change to the game. And we're trying to do both of those at the same time and hope that those add up.

The second one is stuff that's very clear and visible to the player – here's a new level that wasn't in the game before. And then there are tiny details. I agree with you. I don't think that most players would be able to say, 'Oh, I love the way the candy bounces when it falls down from the top.' But it sums up the experience somehow.

I believe, and I think many of us working on the games at King believe, that the compounding effect of very many of those small improvements is really significant. We try to improve things that players would maybe have a hard time describing but can somehow feel the quality. That's what we're aiming for.

So, it will be the same on the day that we launch it as what we present to everyone in the game. We're proud of what we've made. We also know that if you're on level 7500 unless you're going through the levels very fast, it might be some time before you get to level 15,000. And so, while it might recognisable—we hope it's memorable—we kind of feel like it's okay because it might be some time before a lot of players reach level 15,000.

The other thing, though, is that we do change and try to improve the levels at scale on a very regular basis. So that level today probably will be different by the time you get to it because we will have learned from both talking to players and observing a huge amount of anonymised data that we get when we watch results of how our players play – we'll learn something that helps us make it better.

And when we're creating a new level, we do have a combination of historical data and our own intuition. We're starting to be able to predict better and better - by building kind of synthetic players, as it were – how they might respond. In reality, of course, it has to meet the players, and that means that over time we're changing every level in the game a bit to try and make it better.

I don't know, actually, I'd quite like to know. It's a bit patchy, I would say. If you imagine a couple of different approaches - one approach, which would mean all the levels get refreshed, would mean that we essentially go either manually or automatically through the whole game and then do a series of progressive improvements.

That's not really how it works. Because the levels are different, naturally, the way that players experience them is different. So what we try to do is to hunt down those levels which are a bit too difficult, a bit too easy, or they don't seem to be very fun, or players tell us they're a bit confused about what to do.

So, instead of every level getting refreshed on a regular basis, we take specific levels and iterate on those until they work. That still has the same effect – bringing up the average level of quality and enjoyment for the game, but it's a different approach.

We're trying to make those events for three groups of players – people who aren't playing the game today, people who used to, and people who are already playing the game. But it depends a bit on how novel they are or how marketable we think they are.

With the Barbie one, we did some activity to try to make a noise in the player community and beyond because it was widely understood there was a movie going on. We felt it was interesting we had this unique tie-up with them. And that we could present something to players they wouldn't be able to get elsewhere.

There are other events—like the Halloween season—where maybe what we're doing is an interesting combination of different elements inside of the game. Perhaps it doesn't have a hook with a celebrity or brand, and for those, maybe we would expect more response from players who are already playing the game.

Download now!

Candy Crush Saga