Getting App users engaged (Zalando)- Managing your KPIs
Hello everyone. I’m trying to imagine you a little bit naked. You know to take off the distress but I’m not sure I want…My boss is here so I’m not sure how good it’s gonna be. [laughter] Let me start with basically what we have done in the last couple of months. And maybe we can start with a little bit of introduction. So I’m Lior and I’m working for Zalando for 8 months. Originally coming from Israel. I have a Bachelor’s degree in Business Administration and Marketing and Masters in decision making and negotiation. And I was working a little bit with Lavu [phonetics] before that and I have some experience in Cisco News [phonetics] and some other companies.
Actually when I arrived to Zalando I think it was like ‘Oh you’re gonna join this big corporation that’s gonna be very hard to change stuff’. And in the past 5 months we’ve done a lot. It was a huge train of a start up environment that helped us really really to improve what we are doing. What I really really love to say is always when we have these app summits and all these cool stuff it’s to talk about an app that I think changes the way I’m using my mobile. Erm…I use and Bevino [phonetics]. I don’t know how many of you know but it’s like a huge encyclopaedia for wine and every time you wanna know like what wine to buy it’s good to use it and that’s my favourite app.
The time is a bit complicated just before lunch so I believe everybody is tired so I will try to avoid any too complicated formulas. You’re not gonna have any graphs. And were gonna talk about process that we have done in the last couple of weeks and months that changed basically the entire process of how Zalando works with engagement.
So we will start with a little bit of intro about Zalando and we’re gonna continue with some information of how we’ve done what we’ve done and how we created the change. And some facts and figures about Zalando. As I said I joined Zalando 8 months ago. 5 months ago we actually moved into a matrix structure which means that we move to work for the app as an app team. App commercial team. And it allows us basically to start and restructure a lot of stuff. One of the main things that we found very challenging was reporting. Before me some people mentioned actually the horrible tool and we agree so we decided we’re gonna do something a little bit different so we reconstructed our database as we tried to understand how to do it. And just to explain to you how big is the app let’s go for a little bit of numbers. I will skip most of them. The most important one is that we have 26 million downloads last quarter. App downloads. We have our 50 almost 50% or even a little bit more than that coming from the orders coming from basically mobile. And 65% of our traffic is mobile. So it’s a really really big subject for Zalando and we understand if you need to tuck our engagement in a very important way. And this was while on an app commercial team.
We work in 15 different countries which they still have the on views and KPIs and targets for each of the countries and you need to know the countries and you need to know how basically to communicate as a central team in different countries and yet make sure that you can have the same numbers.
I brought you a cake . I’m sorry but you won’t be able to taste it. The cake as I see it is basically you have the berries on top of the cake. This are our visitors. We have a lot of them. They’re coming in and they all basically cost us the same but some of them are covered by chocolate. The one that are covered by chocolate are basically the users that you want to bring more because they’re pressing orders. That’s how I see it. You have a cake. You have your world and inside this universe you have users that are gonna convert or not gonna convert. And the cake is basically symbolising the wallet of the user. He has x amount of money he can spend and you need to understand how can you bring him to spend it at your app and not at another parts of the cake. So you want to have as many chocolate berries on your side rather than others.
When we’re creating the story we’re basically trying first to figure out what is standing in front of us. We understood that the engagement is a very important part. You cannot always trust an organic to do the work for you and just create more and more engagement. You need to measure it and you need to understand how to measure engagement. We found out very very fast that one of [05:05] the key things to do it is to first pose what you’re doing and really think about how we want to swatch the databases how we want to deform the KPIs. What KPIs we want to form there and we try to stop really. And when I’m talking about stopping we didn’t report anything for a couple of weeks. Because we set down and said ‘OK what are the KPIs that we want to have, what is nice to have and what we don’t need to have’. We cleaned our KPIs as much as we could. And than we arrived to these main 5 KPIs that we wanted to show. Which goes to Wows [phonetics] quality index which are for example the CLV [phonetics]. Which I think is a very important one. The number of sessions we produced. The PC 2 and of course the orders. Zalando is a very mature app. And we know that our brand is strong but we still at the end want to have as many orders as we can and we need to produce as much PC2. To the ones of you that don’t know PC2. PC2 is basically the revenue after we clearing some of the cost. And we tried to figure out how with 5 personal KPIs we can present what the app is doing.
How can we align? That’s the best question everybody is asking themselves because at the end you know our goal is orders and you have different themes inside it. And everyone has different opinions. Developers for example will always look about ‘I don’t care about orders I care about the users. For me Wows is the most important indicator if I did good or not. The more users coming to visit the better it’s gonna be’. One of the main goals of the producing for example is to create more trafficking inside the app. As far as they care they want to see the user every second day. That’s a very good point when you come into engagement. You need to understand what the product is trying to achieve and than based on it you can also line with the marketing afterwards.
When we’re going basically to the marketing channel managers they care about the end what the management goals are. They want to understand that they sat and arrived to the goals. And they achieved that. And they want to optimise based on KPIs that were sat in advance.
The finance people also looking at very very rough view about budget about the spending. What basically was reduced so what revenues we drove into the company and based on it they can optimise it. And of course the product managers as we spoke before so they basically looking at traffic and trying to understand ‘OK is the app working or not working’.
What we tried to figure out here was ‘OK so we have these 3 different types of users that are gonna buy our product at the end which is the reporting. How can we basically sell it for them?’ We tried to sit down and to better understand ‘OK we have now 3 different stake holders that we need to supply data but we need them all to align in one aiming. At the end it’s producing more orders’. And more orders means that we need basically to create more traffic. Creating more traffic means more quality of traffic. So we tried to define how we’re gonna do that. When we came to an engagement. And as you know we always have the finance guys that are coming by and saying like ‘OK guys increasing spending doesn’t necessarily mean you’re gonna bring more revenue or you’re gonna basically create more profit.’ So you need to sit down now and see how with these 3 guys you can actually come and actually arrive to a moment that you can actually speak about what KPIs you’re interested in. And that’s where we arrived to a very very simple conclusion. Some of it I will go to this presentation.
So a user is not coming once a quarter that’s for sure. We know it. And if the user is coming once a quarter it means that we’ve done a very very bad job. So we need to improve it. We need to make sure that it’s more engaging and coming more often to see us and use our app. We made engagement measurements so simple that our developers can actually use it. Which means basically that we simplify it. We took the numbers and the KPIs that we used to have and we reduced them. We made sure that there is one number that everybody aligned on it and understand what it means. So they have one KPI that is coming in the morning and I said ‘oh we did well yesterday’ or ‘oh we did so awful’. And that’s basically what we try to figure out here. We try to see what KPI we’re gonna move all of these different function into one purpose producing more orders. And than to create a more…a better engagement that we don’t only trust organic to come back. When we’re bringing them we know that we’re giving them the best experience that they’re wanting to have.
Sorry that’s just to make sure that everybody are still awake. [laughter] I ask my friend to send her cat. It’s very grumpy [10:04] one. I couldn’t use the grumpy cat because of copyright. And than basically we…when we did this process we understood one thing that there is a certain amount of time that we arrive to work everyday. So we have 8 hour of working and we have a certain amount of budget that we need to spend. We needed to figure out ‘OK guys when we come to these event of checking our KPIs in the morning how do we focus what do we look at? What is the most important stuff?’ And we started to…based on it…say for 8 hours let’s try and see what the guys from the chain managers…We need basically to see the first time coming in the morning. What data they can optimise based on. And we try to reduce it. So we try to focus their basic campaigns into top and bottom. ‘OK these is the best ones these is the worst ones and this is how you need to look at it’. We created basically one number that based on it the know they created a profit or they didn’t create a profit and based on it they know how to optimise their campaign. We went to the product guys and we basically pitched them the idea that at the end they need to create more orders and if the app is correct you’re gonna create more orders so based on it you need to know exactly what is your performance. And we went to the finance guy and told them ‘OK guys guys we spending money. There is 2 teams in here that are working a lot to create it. Let’s see if we’re actually doing it’. So than we arrived to a moment that we said ‘OK nets revenues is the most important KPI could be but not necessarily’. And than we needed to figure how we’re gonna do it.
So we found that the really impact is coming basically when you’re creating something that is very very simple and people can relate to it. So it’s one number. One number that they come in the morning and they see ‘OK we did well yesterday or we didn’t do well’. And this is where we try to do orders over wells for example. We said ‘OK that can be a very interesting KPI. Is it really creating any change?’ Than we went to networking and we tried to see if this was creating a change. And we went into a lot of KPI trying to really figure out how can we create any change inside what we’re showing.
What is our quality index? Basically what we’re trying to do in here is to create some index that can expand to us ‘OK where are we standing? What are we doing? What’s going on with us? And are we doing well?’ We used a lot of factors inside it. We took a lot of data sources and we combined them together to make sure that we can track everything in one place.
And than we came to some conclusions after this that basically explained to us what we need to do and what we’re not supposed to do. For example when we sat the KPIs we understood that we could not go and force people to optimise based on certain KPIs. We need to come to the teams and understand exactly what is the aiming and than define the KPIs together with them to make sure we give them the best experience they can get out of our reports. Because if they’re gonna enjoy they’re gonna use it. They’re gonna optimise and they’re gonna improve because they understood what is going on or what is going well and what they’re doing bad. And based on it they know what to change. We questioned everything. We basically decided that we’re doing one main question that everybody will ask in the morning and than from there they can have sub-questions to really understand how they’re doing. We really decided to define the question as a very important one but it’s not only the…it’s not meaning that it’s the purpose. Don’t say ‘OK my purpose is writing more orders’. Different departments, different teams has different ways to look at it. So we tried to define as best as we could.
We took the data structure that we had when we used to use Excel and we used our to run stuff and we moved it into the data warehouse. So we started to process our data ourselves. We created KPIs. We connected as many sources as we could and we made sure that all these sources can connect together. So we made sure that we know how to take the data and combine them into something that is meaningful for everybody.
Our users as I said before are not the same. We understood that we need to do something that will change the rule of the game. We need to walk in…At Zalando we need to walk in your client’s shoes. So this is what we’ve done. We went to each and every team and we tried to understand ‘OK what are you interested in? What your optimise is based on? How we will be able to change the way you’re doing stuff? What will we basically move in the morning? And we tried as much as we could to adapt to this needs. When we build dashboards we done it today very very fast way. We are developing and we keep changing. We never stop because the users needs never stop. You always need to optimise it. You always need to make sure that you are bringing the latest information and the most up to date one and you need to make sure that your dashboards are always with the right KPIs on the first look.
We aligned our KPIs between the department so we made sure that for example product guys will look at the end commercial KPIs. Finance guys also using commercial KPIs. At the end we need to optimise it at the same KPIs and we have one main goal. So we needed to make sure how we align all this data in one simple dashboard [15:11] and not be creating too many dashboards that people will get lost. People are getting lost when they’re coming to checking their staff and especially when they’re coming to engagement speech. I see a very very sensitive subject. We’re trying to see ‘OK this specific client what is his quality? What is he worth?’ So we need to optimise it based on it and we need to make sure that we have it in a very easy and safe place.
We check our data relevancy so we keep checking ‘OK am I actually bringing better traffic or worse traffic? I’m trying to understand now where I’m standing in the middle of my acquisition’. Because than we have organic traffic and we have paid traffic. And when it’s coming to paid traffic it’s a very big challenge.
So our learnings, our main learning for that is don’t let Excel rule you. Connect as many APIs as you can and than go as ow as you can with the data so the most granual level of data that I will get it up. And keep control over your data. Don’t let it control you. That’s all.