On-Demand Webinar

HOW TO BUILD A TOP-RATED AIR QUALITY APP

Download the Presentation Slides
Video Transcription

Ronit: Welcome to today’s webinar, from data provider to habit changer, a product manager’s perspective. Before we get started I wanted to go over a couple of housekeeping items. Today's session will be recorded and emailed afterward. There will be a Q&A session at the end of the webinar. In order to ask questions, please use the Q&A panel that can be found when you click on the Q&A button at the bottom of your zoom screen. Feel free to ask these questions during the webinar as well as at the end when we start the Q&A session. 

For those of you who this is their first time at one of our webinar events, my name is Ronit Margulies and I'm the digital marketing manager here at BreezoMeter and I will be your moderator today. On today's agenda, the importance of research and experimentation, plus tips from our expert identifying a unique consumer and what the data means to them, changing long-term behavior definitions and applications, and of course, as always your questions.

Today's webinar will be hosted by Yair Giwnewer, the product manager of consumer products here at BreezoMeter. Yair has a master's degree in industrial psychology and 11 years of experience as a UX designer and product manager. He heads our consumer products initiative here at BreezoMeter and we are lucky to have the opportunity to pick his brain. Take it away, Yair.

B2C - Data as a Service?

Yair: Cool, thank you for the kind introduction. Thank you. I'm actually very grateful for the opportunity to talk to you guys. We had a very very special year. We started a b2c journey in a b2b company, and the challenges that we felt we were facing, and this unique journey, I thought it would be a good opportunity to stop for a second and share some of our special, not only our experience and tips and things that we've done, but also you'll see some of the things that we didn't even develop yet. Things from our roadmap, and I hope nobody will get mad at me here. So I hope you really enjoy this. 

Part of the reason that this is special is that BreezoMeter as a DAAS company, we're selling data, right? We're in the business of selling data to our customers and they're improving their service to their consumers using our data. But in a B2C journey that really doesn't make sense. That assumption that the more data we were showing the better the experience would be, the more engaged our users would be. So we're really not in the business of data. We're in the business of improving people's lives and reducing exposure to poor air quality. You can phrase it in a few different ways. So thank you. I hope it will be interesting. 

Ronit: Looking forward to it.

Yair: Let’s do this. A word or two on BreezoMeter for those of you who don't know us. As I said, we're a B2B company. We create data about air quality and pollen, we also have weather, and we sell it to companies that integrate our data in their products in order to improve their customers’ experience. This is what the company has been doing for years.  Right now we are creating the most accurate actionable air quality data available in 91 countries. About a year ago we started a new journey. I joined the company and we're now aiming for our consumers as the new strategic initiative in the company, and I'm lucky to be part of it. 

Our B2C activities, we have several things that I'm in charge of and activities that we're doing. One is a website, of course. The company website. But we also have a dedicated B2C website. For every person who wants to know the air quality or level of pollen in their location, we have applications for iOS and Android. They were developed using google's framework called Flutter. That's a whole, it's really interesting actually, developing with a cross-platform framework. A lot of interesting stuff happened there. If it interests you if you wanna have someone from our team talk about Flutter, just give us a shout. It will be really really interesting I’m sure. We’re pioneering there.

We also have other things coming up. We‘re planning for widgets and wearables. We have very interesting opportunities across the industry, the real estate market. So yeah, we're doing a lot of interesting stuff for B2C, and relying on the same data that we were providing for the B2B business, so air quality data, pollen,  the same things.

Research & Experimentation

Before we dive deeper into changing behavior and the psychology behind it, in order to maximize the chances of anyone online who wants to incorporate some of our experience into their products, some of our processes, and the learnings that we have, I feel it's really important to start a bit with the basics. I'm not gonna even try, in such a short webinar, to cover all the aspects or even or even the majority of aspects regarding user research and experimentation. I'm pretty sure most of you are acquainted with all the different best practices in that field. 

But we had a very nice experience with a couple of things that I think, if you do them, you will benefit a lot. And also, if you're not doing anything of what I'm about to tell you and show you, any chance or any effort that you will do to change people's habits, people's behavior, will probably fail. I'm sure you can understand what I'm talking about in just a moment.

Reverse Engineering Personas Through Research & Analysis

With research, as I said, the most basic thing, right? If you want to change the user's behavior, you want to know who the user is, right? Books were written about persona creation, I can point you towards some very good resources on YouTube. But there is something that we did here that actually works really well. We called it ‘reverse engineering your personas’. That means you start with the competitive analysis. We did that by taking a very close look at all the different websites, applications that are in our industry. Some were only weather, but others had air quality. We did a couple of things.

First, we noted what were the main features that they were providing. That helped us identify opportunities, not only for specific features but as a general kind of strategic direction. That helped us see with clarity where we were going in the long run, and where we could do better. That was really really helpful for us in doing the roadmap.

The second thing is, when you look at the website, for example, they chose only educational context, related to air quality, of course. Then you know that the persona that this specific website was aiming for is maybe a teacher, right? Or somebody may be in informal education. But that's definitely what they were looking at. If you're looking at a different website where you see a lot of information about legislation and things related to government rules, then you know it's a whole different player, right? They’re looking at a different persona. 

By doing that, we could reverse engineer our own personas. It's not that we didn't do any other type of research, but it definitely validated our assumption and helped us be confident with the persona that we were looking at from the beginning. So I really recommend you guys do that, no matter in what state you are in your venture. 

The third thing is using social networks to learn. I cannot express the importance of getting intimate acquaintance with your users and hearing them speak. For us, it meant opening and being very active in content creation. On Twitter, Facebook, Linkedin. We have several accounts. For example, you publish a piece of content, it can be a little tweet, it doesn't have to be a whole blog that you just grow, and you see how people react to it. You start following the experts and you participate in the discussion.

To give you an example that was useful for us, one of our more important products: we developed a real estate air quality report, and it was born out of a discussion we saw mainly in the UK around air quality on the social networks, mainly on Twitter. So that was screaming out how much people want to know about air quality when they’re purchasing a house or other real estate. It really helped us. So I strongly recommend you get involved, dive into the social networks, and learn as much as you can.

Ronit: And there you actually are learning their pains, their motivation, the things that drive them.

Yair: Absolutely. You will hear anything from an expert publishing a very important article, and you need to be the expert, you need to be up today, and also somebody complaining that his neighbor just lit his wood stove and it’s causing a lot of smoke in the neighborhood, and he’s pissed off. So it's the terminology, it’s the nuances. It's wonderful. I strongly recommended it.

Establishing MVP & Proof of Concept Through Experimentation

The next thing is experimentation. Again, lots of opportunities to learn out there about the importance of an MVP mindset but here at BreezoMeter, this is how we do it. We first give you a proof-of-concept inside the company and we see whether something is feasible technically, that's the very first step, that's kind of obvious. 

Then, we do not release any product out there without doing some kind of MVP with it. On the two screens on the right you can see on the last example see a button named Stay Tuned. It’s a feature called ‘Family Hub’. It’s a feature that we’re developing right now actually. You see an example, right? You see a nice graphic of our intention and what we think is the right direction for it. And then if you click it we tell you, ok that's under development, we can send you an email when it's ready.

Just by doing this simple graphic, without any substantial code behind it, we just gather hundreds of emails of people who are interested in this feature and we validate it without doing anything. So if you're not doing this type of work you'll probably be working much harder than you have to, and then from what you hear next about changing behavior, which really requires a lot of experimentation, then it's really vital. So I strongly recommend it.

The second one is A/B testing. We are doing it using Firebase, really basic stuff. What you see on the right, our cards on the bottom, is the main navigation in our application. Some cards are out of sight. Theory shows and tells us that what's out of sight will be used less frequently. So what we do with A/B testing is just change the order of the card to see how much impact we do with that test. And being completely honest, we changed the order of the cards, Ronit, and we didn't see any change.  That kind of surprised us, but that's part of the important thing. We're ready to fail, we come as naive scientists and we're ready to be surprised. In this case, we were actually surprised. So this is the basics of it.

Is Data What Users Need?

Ronit: So does this method validate how our data is presented to the end-user?

Yair: This validates the user interface, sure. But it's not only that. It helps us understand deeper meanings of what we are doing. This data is what helps us validate a more substantial concept than what we were going for. Let me show you an example. I think this will make it clear. This is Garmin's application. Garmin is a wearable, one of the most common fitness watches out there. Their application allows you, when you wear the watch at night, which I have, it allows you to track your sleep. Now you see on the screenshot of this guy, it's not me, I did not sleep --

Ronit: I was going to say, are you telling us you sleep eight hours a night?

Yair: No chance. I did not sleep for seven hours and fifty-six minutes in a long time. But what you see here is data, right? This is what I meant. When I bought the watch I wore it for the first time I went to sleep, I woke up in the morning, I checked my sleep, I was really excited to see what happened at night, and it showed me the data and it was fun. I admit it was really fun, the first night it worked great. The second night I checked it again, I did it maybe for a week, and then it started fading.

Ronit: So curiosity and novelty did the trick for about a week but it wasn't sustainable.

Yair: Yeah it wasn’t. I mean if you wore the same watch, okay, let's do this. If you wore a Garmin on your hand because you wanted to know something about your sleep, what questions would you ask yourself?

Ronit: I would want to know how well I slept this week, and if I didn't sleep well I would maybe want to have some recommendations on how to sleep better. Maybe it could ask me some questions about my behavior during the day.

Yair: Exactly. My thoughts exactly. My question was how can I improve my sleep, and they're giving me data about how my sleep was. So in a way, they missed my intention when I started using this feature. They did not manage to create a habit for me that sustained over time, just because of pure motivation, which I talked about, later I kept using it a bit, but that's it. It was over really quick, and that's one example.

Another example here is from Strava. We’ll be talking a bit about fitness apps today, guys so those are the examples we’re looking at. So Strava, look how much data they're giving you and what, we're trying to find on the screen, Ronit, where is the raw data here?

Ronit: I see very little -- oh I found it, but only because you pointed me in the right direction.

Yair: Yeah I know that was easy to find, with the big arrow and the raw data thing. But this is my point. Fitness apps, a couple of years ago, started with just showing you raw data. Now it's such a small part of the entire experience because psychological needs are achieved with the achievements in different gamification and game mechanics that they have here.

Ronit: Social sharing.

Yair: Yeah absolutely. Social is a huge part of fitness apps today.

Ronit: Personalization.

Yair: It's not about the data. This is my point exactly. Still, I think there's a way to go even here, because over time… let me ask you a question: did you ever download a fitness app? Be honest.

Ronit: I downloaded a fitness app or two. Erased. And then downloaded another fitness app or two, yes.

Yair: Ok. And did your fitness app include a safety button?

Ronit: Not one of my fitness apps includes a safety button, and at times that have tripped and fallen in my run, it might have been handy.

Yair: Okay. My point exactly. If you want to ‘catch habit’, catch engagement, pinpoint behavior in the right way, you need to look at it in a more holistic way. And again, now you understand why experimentation and research are so important. How else will we know that a safety button might be something important for our users? 

Ronit: Or if you offer them a high-impact video and then ask them if they were short of breath, and they say yes, the next time you offer them a lower impact video, within their interests.

Look Beyond Task Completion - Data Isn’t Enough

Yair: Exactly so my point here is: look beyond just task completion. It's not about the data anymore. On the right, you see Maslow's pyramid of needs. It's a motivational theory that assumes that once you have the basic needs fulfilled you can go up the pyramid until you're in self-actualization, which is the highest you can go. Motivation works its way from the bottom up.

 I'm showing this to you for some inspiration. Where you can look for the more interesting features that are not just about the task itself, the broader perspective, which is really important when you're about to change people's lives, maybe create new habits. That's the point I’m making.

Ronit: So really data is not enough.

Yair: Data is not enough for us. What we know now at BreezoMeter is we can have the best air quality data, the most accurate air quality data in the world, but we will not have the most engaged users, we will not have the impact that we want to have, we will not complete our mission, we want to improve people's lives, if we don't change people's behavior and create a new habit. And as an example on the right, you can see our recommendations for sensitive populations, such as children, pregnant women - take the raw data and brings it to the next level. These are personalized recommendations that help people change their behavior, and this is just the start.

Ronit: Which is really important if you have an audience as wide as ours, because you're going to be touching every kind of population, every sensitivity, and any type of, you know, asthma, elderly, pregnant, athlete. These are things you have to take into consideration beyond the data.

Yair: For us, yes. Maybe people online, they have a different crowd, they have different types of users, and their opportunity is to twist the data and improve it and digest it. That's going to lie elsewhere.

Changing User Behavior

So what does it mean to change behavior for us? Changing behavior, again, that's the next step that you guys need to do for your own domain. Changing behavior for us is becoming air quality aware. That means you're aware of air quality in every step you take during the day, with small decisions, such as where do you walk, what street do you walk to get to work, up to where do you buy your next home, okay? 

So everywhere you’re air quality aware. You check air quality regularly. If you're not checking it, we want people to have the habit: oh, I wake up in the morning and I check the air quality to know what it’s gonna be today so I can do some planning. Doing that regularly is really important. 

And the last part: minimizing exposure to poor air quality. 

Ronit: This is the bottom line.

Yair: This is the bottom line. If we manage to do this for people we won. They won, we won, everybody’s happy. 

On the right, you can see an example of how we impact people's lives with our fire alerts, which again, obviously have a huge effect on the quality of air and that's something that we're, again, giving to users, not just the data, but also what can you do with it. We're just scratching the surface, guys. It's a journey. I'm not gonna lie to you. It's something we're starting and it's a journey and I'm sharing with you just the start. 

Captology - Computers as Persuasive Technology

You mentioned my background, so I had to bring some theory to the table. 

Ronit: Remember when I said that Yair has a background in psychology?

Yair: Yeah I'm sorry guys. I like going back a bit and bringing some theories. But I only do it when the theory is actionable. So, Captology, for those of you who don't know it, is, in essence, how you use technology in order to change people's behavior. I know ‘persuasion’ sounds bad, I know…

Ronit: In the business of improving lives…

Yair: Yeah. So BJ Fogg is a doctor from Stanford University. He’s been researching Captology for many years now, and it's very very interesting.  I really recommend you dive deeper into that if you're in the same business of changing behavior. On the left, you see a classic example of it. It's daily exposure. I told you one of the things we're doing with our research is understanding what people need, right? So people want to understand what the quality of air has been during the day. How many hours was the air better? We're experimenting, so this is one of our experiments. Once you look at it, it impacts your motivation, impacts your perception, impacts your attitude, and obviously behavior. 

So this is what Captology is about. Technologies to change behavior. Sorry I have to show an equation. Everybody likes equations, but this time it's friendly I promise, and it's really useful. So how do we change behavior? Ronit, how do we change behavior?

Ronit: How do we change behavior?

Yair: Okay I'll do it. To change behavior we need three ingredients. We need to have the motivation, we need to have the ability, and we need a trigger. It's as simple as it sounds. If I wanna change somebody's behavior, I need to build motivation, motivation can be built in many many different ways. Then I need him to feel that he is able to change his behavior, no matter what it is. I need to have some sense of self-efficacy. And then the trigger - what makes him change that specific behavior? We need clues, and using it in an application, using technology, we can do all of that. That's a great opportunity for us.

Ronit: Can you give us an example? A real-world example?

Yair: Yeah, of course. By the way, this will work on any kind of behavior, it can be a small one, but the catch here is it will work only for the short term and not for the long term. Let me explain, and we will do an example as well:

Fogg’s Tiny Habits

If I want to run a marathon, okay? I am currently a couch potato, I decide I'm really motivated, and I'm deciding that tomorrow I'm gonna run a marathon. I'm not able to do that because I don't have the ability, right? The ability is missing. No matter what the trigger is and what the motivation is, it will not work, right? Tiny habits are the only way, not only to change large goals and larger-scale behavior but also to create long-term change in behavior, and this is critical for all of us.

If I’m a couch potato, I want to train every day for an hour. I'm not gonna start from the couch, go out and do it for an hour, right? My willpower will be exhausted really quickly. It won't work, it's not enough. So this is what we do:

First, we said we need a trigger. My trigger? When my kids are in bed I have a trigger. Every time, every day, they go to bed and I go out for a walk. But it starts with a really small one, just to the end of the street and back. This is how it starts. The next day, when I come back I need to celebrate it, right? That's part of my motivation. Woo-hoo! I walked down the street, that's awesome, I'm great. The next day, I do it again. The trigger is there, motivation is there, the ability is there, I go down the street. And then I can build on it. Build on it, build on it, and I have a habit. 

Once I have a habit for the long term, I’m done. If you guys are looking, I know some of us are looking at people who jog in the morning and say, how do they do that? How do they have the willpower to wake up in the morning and go out and run 10k? Well, guess what? They don't have the willpower that you imagine they have. They just have the habit. They did it enough and they built it step by step, and now they're ready to do that without any special effort. That's amazing.

Ronit: So how does this apply to consumers?

Yair: That's a great question. For us, building habits are small things. I want people to, I mean the big message will be, from now on you will be air quality aware. Well, what are the tiny habits, right? So the tiny habit is: walk on the far side of the sidewalk, as opposed to walking right next to the road. We know evolution is much much higher next to the road. 

By the way, if you're taking anything from this presentation, from this webinar, take this: when you're walking with your kids down the street, not only should you choose streets that are less busy with less traffic, but also, since kids are about the height of the car's exhaust, they’re breathing much quicker and they're taking a lot of pollution in, so make them walk away from the road, that's really important. And that's a tiny habit. 

So using the application, we're not doing it yet. I told you we're just starting. We're gonna tell people: do it one day, a second day, build the habit, and in the end, you’re gonna have it. That's our plan. This is what we're doing.

Takeaways and Q&A

So our takeaways: 

  • Before anything else, we talked about the importance of research and you've seen how important it is in order to apply it. So whatever you do, learn everything you can about your users, social networks, work rate. Experimentation is basic. We're not gonna, not us, not you, nobody succeeds on the first try. Be open to failure. Be open to feedback. And keep trying. Try different ways, A/B testing, and you'll get it. 

  • Driving engagement is something you need to define for your own product. We know what will work for us, but you need to put in the effort in order to understand what is the basic core need? It doesn't matter what. Maslow’s pyramid. Try and find what's the important thing for users to drive engagement. 

  • Changing behavior: you need motivation, trigger, and ability. And for long-term behavior change, you need a step-by-step plan. That's the whole message. That's it.

Ronit: Thanks, Yair. That was super interesting. I learned a lot, I hope that you guys did too.

Yair: I know it was short so if anybody is still interested I'm very passionate about it, so if you want to have further discussion, please let me know.

Ronit: Feel free to reach out. We’re always happy to bring on more topics related to products and BreezoMeter in general. We are going to take one question. Thank you for bearing with us even though we're running overtime. If you want to send in your questions now please feel free. 

They're coming in. There's actually a couple of interesting ones. If we don't get to them we are going to follow up with you afterward. I recommend that if you're sending questions as ‘anonymous’ to please send them maybe directly to me or to Yair. We'd be happy to answer all of these questions, they’re all very helpful. The first one is actually related to the previous slide and I'll ask it in that way. 

We saw that the father taking his son to the park decides with his wife's help, she obviously has the habit, to go to the cleaner park maybe not the closest part. So the question is, from this attendee: 

How do we know that consumers are actually building the habit? Where do we get this feedback that we know that we're being successful?

Yair: Well, that's a great question, and really, it's about understanding what's happening with the engagement, right? For us, we see we can track, using our tools, how often people open the app. We want to see time in the app, we want to see daily engagement. For us, we want to see how often people are coming back. So right now let’s just throw a number: we know people are, on average, coming back once every 2-3 days for the app. If I'm creating a habit for you, I want to see daily engagement, right? That's for us. I want people to brush their teeth. To use the app. That’s daily engagement. 

So the answer is in the numbers. You have to keep track. Don't do anything in your app that doesn't have an event that you can track, and you see the habit formation happening over time.

Ronit: Great, thank you. That's really useful. I'm just gonna broach one more, and feel free to answer as much as you can, I know you touched on it previously. The question is: what recommendations do you have on incorporating personalization? 

Our attendee asks in regards to our categories of personal recommendations. I guess he's asking from a general standpoint but also from our experience, how did we decide, elderly, pregnant, etc? Keep in mind that we're coming from a place of people's health, so we have to touch on these subjects. Maybe it's something that you can sort of explain.

Yair: What you said is correct. For us, we have a very strong science team on board here in BreezoMeter, so we know what kinds of pollutants sensitive populations are more susceptible to than others. We know athletes are different and pregnant women, the elderly, even just the name of the relevant pollutants are different, right? I don't want to go into too much detail. This is how we know that you need to give personalized recommendations there. 

I guess maybe for others it's part of the process of identifying your personas, right? If your application, if you're doing sleep tracking, let's go back to the sleep topic, maybe you want to know what type of people are interested in their sleeping habits? Maybe it's people that have certain types of insomnia, maybe, or some other reason that they want to know, to track their sleep. 

So let's see who they are and see how we can personalize that for them. Maybe it's by age, maybe by profession. I don't know, there can be many different ways. That's case by case. That was a very interesting question, I can follow up with the person who submitted it.

Ronit: Please follow up with us directly with any more related questions. Again, if we didn't get to your questions we will follow up with you directly after the webinar. We will be sending a recording of the webinar tomorrow. I want to take Yair again for doing this...

Yair: Sure. Thanks for having me.

Ronit: ...and for providing the interesting content, and we will see you guys on our next webinar. 

Yair: Thank you, guys!