เกี่ยวกับตอนนี้
Shownotes:
Transcripts:
Michael: Hello folks, and welcome to Ecommerce QA. This is a podcast where store owners, directors of e-comm and e-commerce managers can stay up to date on the latest tools and tech in e-commerce. I'm your host, Michael Bauer, and my guest today is Louis Grenier from hotjar.com. Louis has three jobs, he posts a digital marketing podcast called Everyone Hates Marketers.com, it's a show for digital marketers sick of shady, manipulative marketing, like everyone who's listening here.
Nobody on this show would ever do any of the things that you talk about on that show, right?
Louis: Yeah, for sure. And then you put them under pressure and the magic happens. It's not easy to not be sleazy sometimes when you have [crosstalk 00:00:43].
Michael: Oh, that's wonderful. I'm gonna write that down. It's not easy to not be sleazy.
Louis: Yeah. That should be amazed, actually.
Michael: That's what my wife, love that kinda quote. You're also the content strategist at HotJar, which I want to talk a lot about, but most importantly, you haven't said the third item but I think that's because it's the most important. You are the fiance to the beautiful Jennifer, and you guys are getting married in two weeks.
Louis: That's the biggest job, right?
Michael: Yeah. Cool. Well congrats on that, first of all. Why are you talking to me on the show, you should be go, wedding planning or whatever people do.
Louis: Yeah. It's already much, pretty much everything is done. Took us two years to organize it, it's not gonna be a big one, but we took it slow, one thing at a time. It's all ready.
Michael: Nice. Well, two years to plan a wedding, that's what we should have done.
Feel like we, it's hard to follow that one. Tell me this, Louis, why are you at HotJar?
Louis: Because I want to make the internet a better place. Simple as. HotJar is really the best kind of company to join to do that, the solution we are building is really a good way for anybody to improve Internet as a whole, because you can make any website much better by listening to people, by understanding how people behave on your website, by understanding what you should improve on, and stop guessing, right? This is why I join.
Michael: You're making me feel bad, because I thought my company was the best one at making the Internet a better place.
Louis: Okay.
Michael: So listen, I actually probably agree that your company is definitely way cooler than ours. We got hooked on HotJar back in, I think, '14 when a client of mine heard about your product when it was still in beta, and we were like oh my gosh, this is great. It's got all these useful things right in one place and they're really easy to use and onboarding is incredibly fast.
We haven't even talked about what it is. So tell us in a nutshell what HotJar does and is.
Louis: So, it's an all-in-one solution to understand how people behave on your website or on your app. What they do, what they don't do, what they like, what they don't like, and using that, you are able then to identify what you should improve on and where you can grow. So that's really it in a nutshell. Yes, as you've mentioned, HotJar was launched as a beta in our first strategy, so they really got a lot of early adopters, like yourself, to test the product and improve on, so the way we use HotJar is the same way we expect people to use it, which is constantly using it, making sure that we listen to people and we get back on board.
Michael: So you're using HotJar with HotJar, obviously, right? That sounds-
Louis: Yeah it's pretty, pretty metal, right? So that's exactly what we do.
Michael: Nice. So let's talk about this a little bit more. When we're talking about HotJar, we're talking about a few different things. The way we've used it is for funnel evaluation, so you tie in so you can evaluate your funnel. Check out funnel is usually what we're doing, as well as you can do user recordings, you can do user polls, you can do, oh, heat maps, obviously. What else can you do?
Louis: You also have a new feature called incoming feedback, which is a way to understand what people like and don't like about specific elements of your page or the page on its own. So it brings a bit more emotion to the user experience. It's brand new and we just literally launched it yesterday.
Michael: Okay, well I need to turn this on for a client. We went live with a client site last week, and we ran all the other stuff in HotJar but we should have waited til we got that, because this was the main thing we were trying to figure out is what are people reacting negatively to and positively to within the experience in general, and in specific areas. So I need to turn that on, I'm making a note about that right now.
But you do content at HotJar, tell me more. What's your vision for how the internet needs to change in healthy ways?
Louis: Right, so my role is really to try to help inspire as many people as we can using content. Guide, blog posts, anything. Podcasts, anything like this, right? Division we have is really trying to, as I said earlier on, to make Internet a better place by really trying to make as many people understand that you should listen to people. If you understand how they behave, if you truly ... if you truly give a damn about your users, you will have an unfair advantage, and you will be able to go faster than anybody else. You will be able, in the long run, to have a system and a business.
Michael: Talk to me about the fact that the tooling that you provide in HotJar is a kind of a mixture of qualitative and quantitative, and why have you guys decided to go with the exact mix that you use right now.
Louis: That's the question, right, why we've decided this way. Initially it was really about putting, trying to get all of those tools together. So all of the known tools together, as an all-in-one solution, right? So heat maps is not new, polls are not new, recordings are not new, all of the features we had weren't new on their own, but we put them together in a way that really enabled you to get the full picture, the full funnel, the full journey.
Now we are in the phase where we are moving on to just providing stuff to move on to things that are new, brand new, like true innovation such as the incoming feedback, which is not something that is being seen anywhere else. We are moving on to that next phase, right.
Michael: Yeah, you know, it seems like a super big pain when you're trying to make any kind of substitute change or even kinda just understand what you have on a website, because you know, there's user testing but that seems to take forever to get good quality feedback, and I just think that it's great that you're making that that much simpler, and things like that.
Now, let's say that somebody wants to be really, really quantitative, and you know, maybe they've got mixed panel or they've got Periscope or they've got Giraffe or some kind of super duper quantitative metrics tool. How do these tools play together? How would HotJar work in tandem with these other tooling?
Louis: Those tools that you mentioned, I would also include in analytics. Those are, those give you the what. So they give you the key pages that you should focus on. They give you the number of people going there and leaving. They give you the big picture and the what's, right, and HotJar gives you the why. So this is how you should connect them together. HotJar is not intended to compete against [inaudible 00:06:55] on this panel. HotJar is really intended to compliment them, to give you more context around why certain people are doing certain things.
If, for example, you have an e-commerce shop called Fat Pillows, because you said those pillows are actually better than anybody else, right? And you have this check out funnel, and you don't understand why you have such a drop off. Maybe like 90% of people landing on checkout end up leaving. Well, you can use HotJar there to understand why. You set up a heat map, understand where they click. You set up a poll, asking them why they are leaving or about to leave. You set up recordings to understand how people behave on the page, and after a day or two of traffic you will know the answer. You will know exactly why they are leaving.
Michael: So can you lay this out step by step? Let's say that somebody wants to improve let's say the check out. So we've got a check out, what are the steps that you're gonna use, and don't just talk about HotJar, maybe talk about what the other tooling is that you'd need to set up to do this properly.
Louis: So the first thing is, whatever tool you're using for that analytics, you need to understand how many people are coming to this check out. It can be done via Google Analytics, fairly simply. So you need to understand what type of product they are adding to their cart and the last step of the funnel, which is the check out. So, you can use any type of free tool to do that.
The main thing here is really understand ... I think the check out might be something that we can talk about in the next two minutes, but I think the best example to give to start with as a rating and action plan would be maybe at the top of the funnel. So when people land on your website in the first place, because the check out is kinda the last thing. But actually, if you improve, maybe if you improve the first step of the funnel, you might have a much bigger lift than if you put the check out, right?
You figure out that when people land on your home page or on a specific landing page where you're paying a lot of money for people to land on, if you find out that many, many people drop off, maybe 80%, maybe that's where you need to look first, right?
You need to look at the number of people landing on this specific page and then the amount of people leaving, the amount of money you're potentially losing, and therefore the opportunity that you have in front of you to improve.
Let's say people come from an ad and land on this landing page. The first thing you should do is understand why are those people on this page, so it might seem like they all want to buy your product, in your head. That's not necessarily the case, and that's usually not the case, especially if they have never heard of your brand before.
So, the first thing I would do is try to understand what are those people looking for, and might only be two people just saying pillows but actually, if I search for pillows or good pillows on Google, it might not mean that I want to buy one. It might just mean that I'm doing research right now, and therefore if I land on a page, kinda forcing me to buy something, I'm gonna leave because that doesn't give me the answer I'm looking for, right?
So, you could set up a poll there, on this page, straightaway after somebody lands on it or after a few seconds, asking what's the purpose of your visit today? What are you actually trying to achieve, right? The second question I would ask, which is actually really interesting especially in e-commerce because Cempathyion is really fierce, is have you heard of our brand before? And asking these questions means that if, let's say, the vast majority of people say no, never, well it's unlikely that they're actually gonna take the decision to buy from you straightaway, because they only said they trust you right now, right?
There actually have been a lot of studies about that. The more people know your brand, the more they are familiar with it, the more likely they are to go through the purchase, to click on an ad, and all these kind of things, right? That's called brand affinity, and this is incredibly important, especially at the top of your funnel. So that's what I would do first, understand what are people looking for, and then if you find out that people are looking for, like have a lot of questions, don't necessarily want to buy right now, perhaps you need to switch your ad into something more educational about the guide to find the best pillows and get a reader this way, educate those people and then lead your brand this way so that they have heard of you and they are much more likely to buy in the future.
That's what I would look at first, you know? The type of people landing on the page and what they are looking for.
Michael: Yeah, yeah. You know what my favorite thing to do with HotJar is? Is to stick up a poll after check out and say what almost caused you to abandon today?
Louis: Absolutely. That's ... sorry to cut you, yeah, yeah. It's just, I'm passionate about this and I tend to cut people off.
Michael: Right.
Louis: So exactly. So let's say now you've moved into the middle of the funnel, where you're more likely to drop off for other reasons. This is exactly what you can ask, you can say "What almost prevented you from buying today?" As you mentioned in the check out. But you can before that-
Michael: Yeah, even before that, exactly.
Louis: You can ask what's preventing you, actually, from doing what I want you to do today. So you can set up something that's a poll that asks this particular question when they are about to leave the page, and honestly, it's not a [inaudible 00:11:59] hard sell, you can, the mini tool's enabling you to do that, right?
Michael: So you would just set up an exit intent poll?
Louis: Yeah, yeah. So you can set up a poll that's on Test Doctor that says when they're about to leave the play, display the poll.
Michael: Why are you leaving?
Louis: Yeah. So you need to have a decent amount of visitors in order to get enough answers, but it's a quite careful one. But let's not forget one thing, though, we are talking about technology and HotJar and all of the other tools like analytics. Do you know, there is a tool that we all have that we don't use too much, do you know what that is?
Michael: It's not my screwdriver. I don't know.
Louis: It's our mouth, our ears, basically our senses, right? So, let's not forget that we are human beings and talking to other human beings is still our most valuable thing ever, which means that instead of investing a lot of money in ads and in user testing, which you can do, it's just bringing a few friends in the same room as you, showing them your website, letting them use it and just see what they do, how they react to it, what they say, the question they ask, and that's gonna be probably the most effective thing you can do if you have a low budget and don't have a lot of visitors right now.
Michael: I'm gonna push back just a tiny bit on that and say that works really well if those people are somewhat close to your target market or at least can ...
Louis: Of course.
Michael: You and I probably aren't that similar to the people that use our tools, in a sense. I mean, it's a good idea, I'm not saying not to do it, I'm just saying that's one thing we have to watch out for, is making sure that whoever's giving input is a member of the audience that we're trying to hit. So good input there.
Okay, so we hit top of the funnel, middle of the funnel, conclusion of the funnel. We hit asking basic questions. Can you back up a little bit and do some kind of more high level discussion around one really important method: clickthrough rate.
Louis: Mm-hmm (affirmative).
Michael: What are some really good tactics that you've seen people using HotJar for to make a big dent in things like click through?
Louis: So do you mean click through rate from an ad to a landing page?
Michael: More like once you hit the landing page and then you're gonna go beyond there and actually click all the way through the site. You could also use the term abandonment, so what are some really great ways people have been using your tools to identify problems with their abandonment and how have they been solving those problems?
Louis: Just think, first of all. In HotJar we're not a big fan of best practices, right, so it's very difficult to tell you this is the thing you should try today that definitely works. That's a word of caution, every business is different and therefore everything you can do, your customers are different so you should be careful of so-called best practices, generally speaking.
However, what we like to talk about more is kind of what we call first principles, so what are the things that will never change about people? And let's focus on that, so there's one good thing that we mention in outlining to prepare for this show is the choice of a lot, so data, like behavioral scientists have proven many times over that the more choice you're giving someone in front of them, the less likely they are to take a decision and the more likely they are to be frustrated about it.
Michael: So ...
Louis: Think about any time of you're going to the restaurant and you have this menu with literally fifty items for starter, 50 items for mains, 50 items for dessert, then you have the daily special, you have the middle of the day, you have so many options. You feel overwhelmed, you don't know what to choose, you get frustrated. Now, I show you just the early bird menu, two items for each. All of a sudden, you just feel much more relaxed and you're able to take a decision. So it's actually extremely important, so that's one of the first principles you can use. Try to simplify the experience you're offering users. Try to remove things that are not being clicked on. Try to remove things that people don't look at, because chances are it might overwhelm them to take a decision, so that will definitely be one of the first principles you should be looking at.
Michael: Mm-hmm (affirmative). Yeah, that's a super important thing to me. It's like people feel nervous when they're removing things, but in reality it's like you're reducing friction, you're making it easier for people to make decisions. So if you know anything about your product at all, prioritize the content that you want people to take action on and you will find that they will be inclined to actually do that.
Louis: Absolutely. Another thing I can say is, so we love to talk to our own users and customers in HotJar, so we like to spend time interviewing them, understanding how they talk and how they describe HotJar and how to describe the features we have. So, another thing that works really well, because once again it's based on first principles, it's based on the fact that you are using your customer for your marketing, is that you should use the language they use.
That sounds cliché or that sounds simplistic, but it's not. So, there's something happening in the marketing world or even in the e-commerce world is that when we write copy or when we try to explain something to customers on a page, we tend to put this hard, which is the hardcore salesman person, trying to sell you something and we use words we will never use face to face with anybody, right? So we tend to overpromise, overdeliver, we tend to use words that don't make any sense, that are like overly generalizing and we are not specific enough.
A good thing to do is to literally use exactly the language your customer are using to describe your products and this will have a dramatic effect. It's likely to have a dramatic effect on your commercial, on the experience because that's how people talk, right?
Michael: You have a good way of figuring out how ... that's one thing if you're just talking to someone in person, how would you do that? Maybe there's a way to use HotJar to determine how customers are talking about your product?
Louis: Yeah, so you know, if you send surveys and ask them the question "How would you describe your product to a friend?" And asking them to be as literal as possible, you recollect a lot of data around how people actually describe that to a friend, right? So surveys is a good way, if you don't have a lot of time to spend talking to people, but I would say the number one thing, even if you can't meet people face to face, is to have a 15 minute conversation with them and just make them talk. So you're not a salesman in this conversation, you are a journalist. Your aim is to understand them better so you must make them talk. And once again, the question could be as simple as, let's say you are describing what you do to a friend. What would you say exactly? And let them talk. And if you recall the conversation, you can use transcript as a way to really improve your marketing, to something that really talks to people much better.
Michael: That's great, and I think that would apply to product descriptions, that would apply to landing pages, that would apply even to your motifs and your logos and your, it's a very useful thing to do. How have you guys used that approach with HotJar to change your messaging?
Louis: So we are doing it right now. We don't see that as a one-off project, obviously it's a continuous project. So at the minute in the content team we are talking to at least one customer a week, we do that over 30 minutes and we have a customer delotment process. So if you don't know about customer delotment, it has been kind of developed by, in the start-up world as a way to understand what people value about your tool or your product or you're selling, what they don't value, how to describe it to a friend, et cetera, et cetera. So we have a few questions like this that we go through and we let people talk. We compare them every month and we try to improve things one thing at a time, one page at a time, one piece of copy at a time.
At the minute, we are actually undergoing a big project around the jobs to be done, [inaudible 00:19:36], which is a very powerful concept. It sounds also really simple when you explain it, but it's actually very powerful. So instead of trying to sell features or even benefits, what you are trying to do is consider your solution as the way somebody uses to perform a job. Such as, you know, you don't buy a pillow just to be comfortable, you buy a pillow to sleep better, but even beyond that, maybe you buy a better pillow, to get on better with your wife because you've been snoring for the last 6 months.
Michael: You keep coming back to pillows, man. You must have been shopping for pillows recently.
Louis: Yeah. I'm trying to stick to one example or else it gets tricky, doesn't it? So, this is kinda the job to be done of the pillow. If you understand exactly why those customers hiring this pillow for or your product for, you really then understand that you can explain what you do and what you said much better. Because it's never about the pillow.
Michael: Yeah. You know, it's interesting with the jobs to be done, the thing I find funny and I'm not saying not to use it, I'm a really really big endorser of this, but it's funny how it sounds when you write out "The job for this website to be done is blank," as though it's a person, but that's the whole point, right? We're trying to take the things we're creating and have them do things that cater to the needs of real people, rather than just roughly approximating.
How, for example, with HotJar, how would you describe the job to be done with point to the components of HotJar as you understand it now?
Louis: So you see, it's not about the component anymore, so that's what we've discovered. It's never about the component, it's more about the job. So for marketers in particular, let's take an example on them because we are interrogating many marketers at the minute. It seems like the best, the biggest job that they have is to understand people, understand how people behave on their website, right? That's the job, so they want to understand that. It means that they don't really care whether they have to use a poll, a heat map, a recording, or all of the above. They want the job to be performed.
So what they want to know is how to use those features together and in each other to perform the job, right?
Michael: Exactly.
Louis: The other thing we discover is there are adjacent jobs that they want to be done at the same time. The first one is they want to get stuff done. They don't want to lose time doing it, they want to spend six hours a day performing this job, they want to do it as fast as possible because they are so busy, they have so many things in their head. So therefore, not only do they want to understand how people behave on their website, but they want to understand 50 minutes a day.
That really gives us a huge insight into, we need to get better at giving them what they need faster in a better fashion, better designed way, so that they get the value straightaway and they can move on into the rest of their day.
Michael: Interesting. That almost sounds like a parameter in part of the job. Your audience wants to understand user behavior; they also want to do that quickly within under 50 minutes a day.
Louis: Exactly, right? And the second adjacent job we discovered is also, not only do they want to do it fast but they also want to be able to convince the decision makers, the managers, the C Suite, to change the experience based on that. So they want the tool to be able to convince them, right? Because they are convinced, but it's not the case of everybody that's here.
Michael: So to talk about ... I want to riff off of that for just a moment. That was a really great example of using the jobs to be done framework. What do you see HotJar going towards in the future, if you could share anything about that? Right now I see the tool is providing a lot of ways to understand your audience quickly. Anything you can share about the future of the company?
Louis: Sure. So the first thing to say is that we are very transparent and you can access the HotJar roadmap right now if you go to, let me check again the address. You should just Google HotJar roadmap, you'll find it straightaway. So we actually publish what we are working in the future. There are a few things that we are working on that are more like fixes rather than brand new features. There's a few things that we need to fix, especially regarding recordings, that we're improving but we are very excited about a few things coming next.
The first thing, as I said, is the incoming feedback feature, which is coming out of beta, which is a very good way to evaluate emotions rather than just things. We are also working on integrating Google Analytics with HotJar, which is something that people have been asking for quite a lot. Also integrating it with Xavier. We are also working always on recordings, so at the minute-
Michael: Oh, that'll be awesome.
Louis: Yeah, right? A lot of people, lot of customers have been requesting that for ages. The ability to record visitors on an ongoing basis without to set up new recordings all the time, that's definitely something that has been ask a lot. More in the future, we are gonna move to what we call an event-based architecture. So, instead of just recording page views, we really move on to give you the ability to select the event you want to record. It could be anything, right? So stemming out and [inaudible 00:24:52] you can record specific triggers that are unique to you and really use HotJar to edit them. So it's really a move from basic function ADT's around page views and site visits to really integrating with your marketing stack and understanding how people truly behave on your website.
Michael: That's absolutely fabulous. You know what we're doing with the client right now is we're setting up MixPanel for all of the qualitative, the what, you know, quantitative, I always get those two mixed up. And what I want to do is we already have HotJar installed, but I want to get it fully aligned with the MixPanel data so we'll be able to see the what and see the why and just dig deep into that. It's gonna be fantastic.
Louis: Yeah, we are really looking forward to it.
Michael: So Louis, tell me this. Any final thoughts as we kinda start to wind down here?
Louis: Oh, that's a wide question. I would say to summarize, really, it's all about empathy. You know, we see HotJar's almost an empathy as an obvious kind of product, it's really about understanding kind of people. So, it can be difficult sometimes for you to see the forest from the tree when you are like in front of your computer all day and have to manage this e-commerce tour and looking at these Google Analytics numbers and those spreadsheets and all, but trust me, if you do try to speak more to your customers and understand who they are and what they are trying to achieve in their life, you will get better at what you do. You will increase your sales in the long run, because you'll have more empathy for them. So I would say that's probably the most single biggest improvement anybody can take in their business, can make in their business.
Michael: That's great, empathy, yeah. So cool. How can people get in touch with you if they've got questions about how to maybe set up a really effective HotJar, testing and recordings and all that, or maybe they've got some questions about how can we collect user data better in general, or some of the methodological or framework stuff that you shared?
Louis: Right, so there are two things you can look at. You can Google HotJar big picture, which is a big picture worksheet that allows you to basically do what I explained this episode, which is understanding why people are doing certain things, it's one page that really enables you to get the big picture of what's going on on their website.
Michael: Oh, I love this. I'm looking at it right now, this is fantastic.
Louis: The second thing is the HotJar action plan, which is a little bit more actionable items that tells you what you need to do. So as I mentioned, you set up your key pages, you set up your heat maps, you set up polls, all these kinds of stuff. So it's really a solution based action plan, it's not about choosing x feature or x feature, it's about using them in coordination in order for you to perform the job you have to do, right? So those are two step I think you can do next, right, and if you have any questions you can email me at louis@hotjar.com, so it's L-O-U-I-S@hotjar.com. Any question, really, you can email me and then I'll forward to the right person if I can't answer them directly.
We have an excellent support team as well, who are more than happy to help you if you need any help. They're also on the ball if you need them.
Michael: Let's make sure to mention your podcast, that's EveryoneHatesMarketers.com, because everyone does hate marketers. Do you think people hate sales people or marketers more?
Louis: They hate them both.
Michael: How about lawyers?
Louis: There's actually a study about, you know what, I saw it, I think it was on HotSpot, marketers and salesmen are only 2% of people who trust us. Even lawyers are above us, around 4 or 5% if I remember well.
Michael: I think I'm gonna have to tell my brother that. I'm always telling him lawyer jokes, I'm gonna have to switch to marketer jokes.
Alright, well let's change the industry. Let's start using real data to solve people's needs and then they'll be happy and like us instead of hating us, right?
Louis: Exactly.
Michael: Cool, Louis. Thanks so much for joining the show and congrats on your upcoming marriage. This has been fantastic. As you know everyone, you can find the show notes at ecommerceqa.com. If you have any questions regarding the show or suggestions for a future topic, podcast@sellry.com, S-E-L-L-R-Y.com and that's it for today. Thanks everyone. Thanks Louis.
Louis: Thank you.
