Why are customers leaving your site? Why aren’t they converting? Analyzing your website’s metrics may give you some insight, but it’s probably not giving you the whole picture. Jared Spool is here to dive into the world of personal interaction with customers. Jared is the founder of User Interface Engineering and is considered the “godfather” of usability. He takes us beyond numbers and into the minds of your clients. We discuss what metrics to focus on, how to learn from the client experience, and more!
This is Stephan Spencer, you’re listening to Marketing Speak and today I am so excited to have Jared Spool on who is a guru on usability. He’s one of the top guys. And just a little bit about Jared before we dig in to this episode, Jared Spool is the founder of User Interface Engineering or UIE. He’s the co-founder of Center Centre and author of the book, Web Site Usability: Designer’s Guide. He’s also co-authored Web Anatomy: Interaction Design Frameworks that Work. He is the conference chair of the UI conference in the UX Immersion Conference. He’s been doing usability since 1978. He is the “godfather” of usability. It’s great to have you, Jared! It’s so great to have you, I’m really excited to dig in!
I’m very excited to be here!
Yeah, so let’s start with a more broad and hone in on some topics here but I want to understand a bit more about how you see marketing and usability interfacing or going hand-in-hand because of course they do but most people see them as separate disciplines.
They are all part of the experience of a customer. And I guess it depends on whether you consider marketing to be a pre-sales activity or a post-sales activity or a lifetime activity. I don’t know—what do you think it is?
I think it’s lifetime so if you are good at marketing, you’re going to get prospects coming into the funnel and then if you are good at marketing post-sale then you have a great customer lifetime value and they are going to continue to buy more and they are going to buy more with increased frequency and spend more per purchase and they will be with you for a lifetime, hopefully.
Yeah, so if you’re going to use that sort of thinking, the number one thing that is going to drive that lifetime value is the quality of the experience that they have with the product or service. If the service is buggy or if the service is frustrating in some sort of way, there is going to be a problem with the design and that is just going to make the marketing work that much harder whereas, if the design is delightful, if the experience is delightful, and if every opportunity people just want to have more then it basically markets itself and they tell other people so that not only increases word-of-mouth, which is going to reduce the marketing effort. Now, I guess if your job is to create marketing effort, reducing it is probably not a good thing so if you’re an agency, having a great product is probably not great for you because the agency won’t have to do much. But as the company that is running the operation, it probably makes a huge difference.
So I heard you speak recently as you know and you talked about being delightful in your talk of how to go beyond customer satisfaction because that is essentially-called-If you could describe your favorite restaurant as “edible” or “very edible” or “not very edible”—that is how we see our customers and clients in these surveys as “Well, that was very dissatisfying,” or “That was somewhat satisfying,” and “That was somewhat edible,” and I love that analogy. We tend to think of just getting satisfaction out of our clients so that they don’t run off and tell everybody and their brother online that this is a horrible company to do business with and ruin our reputation and then we have to a big ORM cleanup job. Apply this concept of customer delight into what could somebody tangibly do in order to take some action that will get more customer delight and take the focus off just mere satisfaction?
The basic element of delight is that it is the opposite of frustration. So think about flying on an airline and these days, to delight a customer, you just have to basically have a flight where nothing goes wrong. It’s a very low bar but one could imagine a flight where there was plenty of food, the aisles were nice and wide, and the plane was quiet, and the ride seems to pass very quickly, and the entertainment system worked really well, and there was never a line for the bathroom, and all of those types of things. It’s possible to create delightful experiences even from very frustrating ones today if you put some interesting thoughts into it and that’s what designers do. Designers look to seek out removing the frustration and to add delight into the process so the question then becomes how do we tell whether we’re frustrating people or delighting them and how do we keep that as the overall objective? As you mentioned, we spend a lot of time measuring satisfaction but saying something is satisfactory is a very neutral statement, right? Satisfactory means “It met my needs but it didn’t go much beyond that,” Delight implies “It went beyond what I needed but in a way that was really pleasurable and helpful to me,” so those changes are important and understanding how to design for that and to measure that is key.
Right, so some people do a Net Promoter. My previous company Netconcepts got acquired by Covario and they were really big on Net Promoter. It was the key indicator to know if they are on the right track or not, and bonuses and so forth who were tied to net promoter scores, and you think that it’s a whole bunch of bunk. Net Promoter is not the right sort of system to be tracking your customer delight and long-term success with, correct?
Yeah, I mean Net Promoter has lots of problems. One is, scientifically, it’s not very good. It’s a sort of crazy metric to begin with—this idea that you take all the scores that are six and below and subtract them from the collective scores that are eight and above, it’s an arbitrary set of rules. The noise that is in the study is your score will change from one survey to the next and the noise that’s in the next survey is hard to separate from the actual signal of something going wrong. It doesn’t take into account people who don’t respond and if someone is so angry with your organization that they refuse to communicate anymore, or someone is so delighted with your organization that they don’t want to take the time to fill out the survey, you don’t get their feedback. And the idea that somehow, the very small returns that Net Promoter surveys tend to get are representative of the bigger population. There’s no evidence that that’s the case so you’re making decisions based on scores that aren’t really representative of what would happen if you talk to all the people who were delighted or frustrated by your product or service.
And then the other thing is that a recommendation to somebody else, is an amorphous thing. I see Net Promoter being used for all sorts of crazy right now. I’ve seen people who were dealing with emergency services and were asked if they would recommend these emergency services to somebody else and that doesn’t make any sense. It’s not a very high level of customer engagement recommendation, it’s like “Yeah, go ahead and fly United. I fly them, it works most of the time, I really don’t like them but I don’t have any other options.” That’s a recommendation so you’re counting that along with the heavy endorsement. It’s not really a very good metric because it doesn’t tell you much. The most important thing is, let’s say your Net Promoter score tanks, do you know what you’re supposed to do differently? There’s nothing in the Net Promoter score by itself that tells you what to do differently. Now, everybody always say, “Well, you wouldn’t do it by itself, you always do it with other things,” and my response to that is, “Well, if you’re always doing that with other things then just do the other things,” Why do you need the Net Promoter? I’m not a big fan of the Net Promoter score, I don’t see it ever adding any value into a process.
Now, you recommended a different survey system, Gallup CE11, I believe?
Yes, so the Gallup CE11 was put together by the Gallup people. They are the same people who put together the election polls and political polls and they’ve done a whole bunch of work in businesses to understand how to keep customers engaged and how to keep employees engaged. They’ve put together something—which is a much more meaningful engagement metric—that actually asks, as a part of it, the question of “Would you recommend this to somebody?” but that’s just one of the 11 questions.
So, the 11 in CE11 stands for the 11 questions?
I love Gallup, by the way. They created the StrengthsFinder test, which I’m a huge fan of. I know my five strengths by heart. I use them every day and I love it! So, listeners, if you want to take a test that helps you understand what your strengths are and how to leverage the strengths in others, buy the book, StrengthsFinder 2.0. It includes about a 20-minute assessment that you take online and it is amazing. It’s just really a great thing. Sorry for the aside, but go ahead with the CE11.
Yeah, so it’s a different team in the same organization who put together the CE11 and they have the intent of trying to understand and answer the questions, “What does it really take to keep customers engaged?” and “How would we measure whether we’re doing a good job on that or not?” and so they went out and they, using a very similar method to the way they developed the StrengthsFinder stuff, analyzed how people think about brands. And one of the things that we never talk about is, people see brands as something that you disseminate or you push out like, “We need to communicate our brand.” What a lot of people don’t realize is that it’s not how it works. How it works is, that brands are built over time. They are the sum of all the experiences that somebody has with your products, your organization, or anything to do with you. You know, brands are affected by a lot of things. The British Petroleum brand took a dive when one of their oil rigs in the gulf exploded and everybody saw them as polluting the gulf and it doesn’t matter how much their PR company promoted how awesome their brand is, that particular experience had an effect and the reality was, they didn’t get a say in any of that. That was just something that happened.
The way the Gallup people measure brands is that they break it up into five different categories: Loyalty, Confidence, Integrity, Passion, and Pride. They look at each of those categories with 11 questions. They look at how well someone feels about, whether if there’s a problem, you’ll be taken cared of by the organization or if the world would be a different place if the organization didn’t exist. The people’s reactions to those questions give a real insight into how they feel about the brand. What’s interesting about the CE11 is that when we’ve used it, we’ve used it in conjunction with people actually interacting with products and services and what we get from that is this beautiful reading as to what it is about the product or service that causes people’s engagement to change. We can actually see the behavior where the engagement changed substantially and say. “Oh, that’s what’s causing this brand to have an issue!” or “Oh, that’s really delightful!” Because we can see it, suddenly we can react to it. We can do more of it if it’s something positive or fix it if it’s something broken. And that’s what’s really key. It’s not just this dashboard metric that we’re constantly watching and then scratching our heads and saying, “Well, what if we put new stuff out?” You know, it’s something that actually lets us see where the problems happen.
Let’s talk about metrics that are actionable and metrics that are kind of a “red-herring.” You spend a lot of time in your talk discussing this issue. We’re so enamored as business people and marketers, in particular, in analytics and yet, the kinds of metrics that are being tracked or the measures that are being tracked, I should say, by software are kind of a waste of time. They are distracting us from the things that really matter and that’s measuring things and trying to figure out what to do differently but there’s no action that we can take based on the data. So, first, if you could describe the difference among measurements, metrics, and analytics for our listeners—that would be great—and then let’s dive into this topic of metrics that are meaningless versus metrics that actually matter.
A measure is something we can count. It’s the speed at which somebody clicks or it’s the amount of time they spend on a page or it’s the number of instances that a message a pops out—it’s anything that we can count. A metric is a measure that we track over time so common metrics through history have been things like visits or page views—so these are things that we’re counting and then we look for changes and when they change, that triggers us to do something different possibly. Analytic is a metric that computers can count so it’s something that computers track over time. The difference is important because there are lots of things like, how delighted the customer is or why they’re coming that we could count but the computer can’t count because it doesn’t know that. The key thing is understanding what the differences of those things are. And so, common metrics are things that the analytics packages tell you like bounce rate, or time on page, or number of pages visited in a session. Then there are things like conversion rate and then there are satisfaction metrics like Net Promoter Score. These are all different metrics that we can track over time.
However, the thing about metrics is that they don’t tell you why the number are the way they are and they often misrepresent and particularly, if you’re taking large amounts of data and averaging them—so, you’ve got some people who are coming to a website because they are ready to make the purchase right now; some people who are coming to the website because they heard about this thing but there’s no way they’re going to purchase today; and some people who are coming to the website because they clicked on some ad by mistake and they have no intention of actually being on the website and as soon as they get there, they are out of there—and we’re counting each of those the same and we’re averaging them together and that’s not helpful. What we really want to do is separate that stuff out but we don’t really have tools to do that because we can’t tell the people who were there by mistake from the people who are there deliberately. We can’t tell the people who are going to buy today from the people who were not quite ready to buy today.We can’t tell the people who are going to buy today from the people who were not quite ready to buy today. Click To Tweet
So it becomes a meaningless metric that we can’t do anything with and we can’t take an action on. If the bounce rate is high, it could be that we’re solving their problem with one click and one-page view or could be that they wandered into an area that they didn’t mean to and they’re on the website by mistake. Whatever happens that there’s no indication of what the intent was, only the fact that they bounced or the fact that they converted. Trying to base your success or gauge your success, I should say, based on something like bounce rate or time spent on site is kind of ridiculous.
You know people fall in love with these metrics. They love bounce rate, they love conversion rate but they use them to support their arguments or what I call, “agenda amplifiers.” For example, they will only bring up a given statistic if it helps them convince the agenda they are trying to convince and they’ll leave it out conveniently. They’ll never show something it says, “Well, this actually says the opposite of what I was just telling you but that’s because we don’t really understand what this metric is counting so you can ignore it.” You know, that doesn’t show up on the presentation deck. It’s really there to support the agenda and the reality is, if you torture any data enough, you can get it to confess to anything you want, including conversion rates. Conversion rates are definitive that way. They’re the ratio—you can manipulate the numerator, you can manipulate the denominator so I can double my conversion rate by getting twice as many people to purchase.
If conversion rate for you is, the number of people who purchase divided by the number of visitors, I can double my conversion by getting twice as many people to purchase—that would be good if that’s our goal—but I can also double my conversion rate by actually having the same number of purchasers but having the number of visitors. If I have half as many visitors, my conversion rate is still doubled even though I haven’t sold any more product so is the goal to sell the product or is the goal to have a better conversion rate? That’s where people get into trouble because they often will act as if it’s to have the better conversion rate when in fact it’s to sell more product. Conversion rate is only a proxy for selling product if the denominator stays the same but the denominator never stays the same. I was talking with a client the other day where they were using conversion rate as a metric for lead generation and they were able to generate twice as many leads for the sales force but what they weren’t taking into account was that the leads were far worse quality. They were leads that the salespeople didn’t want because they weren’t resulting in actual sales so now, the salespeople had twice as many people to call but only half of them were producing any outcomes.
Heaven forbid that the team gets incentivized by the conversion rate metric! So, what would be a metric that would be a better fit for a marketer? A more actionable and meaningful metric?
If you’re going to do a ratio, do a ratio that matters. Reduce the denominator while you increase the numerator.
In this instance, I would do quality leads. I would have the salespeople rate the leads and say, “This lead produced a great sale!” and then see how many quality leads were producing and try and look for patterns in what makes a good lead and a bad lead. What is the ratio of good leads to bad leads? If you’re going to do a ratio, do a ratio that actually matters, where you’re trying to reduce the denominator while you’re increasing the numerator.
But this isn’t something that you can just get out of Google Analytics?
No, you have to have the salespeople apply quality to it, right? And quality is a key element.
But you’re going to get resistance from marketers who are used to just pulling data from Google Analytics and not having to—
Well, Google Analytics won’t tell you anything. You can keep pulling data from it—you can pull data from anything. You can count the number of leads that end up on the home page but what does that tell you? What’s the value of just randomly pulling data if the data doesn’t actually tell you whether you’re successful or not. So, the question is, is it more important to pull data or is it more important to have success?
So, if somebody is building a dashboard for their executives based on Google Analytics or you know, some other analytics package, Adobe or whatever, that whole approach is fundamentally flawed because they are just pulling data instead of trying to create something that’s actionable and meaningful. What goes into a dashboard that an executive should see then?
What is it that the executive needs to see? I mean, this is a question that is going to depend on the executive, doesn’t it? This is the whole point of experience, right?
Well, let’s say it’s the team. Let’s say that they have a weekly meeting, all the executive team, and the five minutes are spent each meeting reviewing the success of the online marketing. They’re pulling data from Google Analytics and assembling that into a dashboard and they’re putting green and red and yellow in there to highlight different things like, “Okay, this is on track,” and “Oh, the bounce rate is heading south or it’s getting worse!” What would go into a dashboard in that kind of scenario that the executives would want a quick update on?
Why are the executives investing in marketing? That’s my first question. I mean, if you don’t know what the investment in marketing is supposed to produce, then what metrics are you tracking to see if you have produced it?
So, what would be an example of a good goal or a bad goal?
Lead generation is only good if the leads are good.
Well, lead generation might be a good goal but lead generation is only good if the leads are good so you know, if the goal is to generate more good leads, I’d be counting good leads.
Makes perfect sense! So, what if they’re tracking total online sales?
Okay, sales works. It’s a dollar amount and it goes up or it goes down. You can say, “Well, we made less money this week than last week,” and you can ask the question, “Okay, did we do something that make less money with this week than last week or is it just the natural ebb and flow of data? So, maybe you’ll look at rolling averages and saying, “Okay, the last four weeks have made more money than the previous four weeks,” or year over year or something like that. There are all sorts of ways to do this. It’s going to depend a lot on the business and the goals and the specifics of the context of what people were working. There are no generic metrics that are meaningful.
If it’s a company that is B2B and it’s a considered purchase and it takes a long time and perhaps, the majority of the sales come in offline—that’s a totally different scenario versus not a considered purchase, B2C, kind of impulse buys and maybe it’s an oriental trading company where you can buy all sorts of fun little things for your kids’ birthday party and it’s kind of a no-brainer that you go online and ten minutes later, you’ve checked out. Different business, different sorts of metrics that the dollars matter but the dollars represent different things happening behind the scenes too.
Absolutely. Look at it this way, you and I both go to work. We can talk about whether it’s important that we reduce our morning commute. Maybe you don’t need to reduce your morning commute because you work at your house so measuring commute time for you wouldn’t make sense. Measuring commute time for me might make sense but that depends, right? Why do I need to reduce my morning commute? Do I need to reduce my morning commute to zero? Or do I need to just get it down so that the ratio of my commute to the time I’m actually working is better? Or do I want to spend less time in traffic? It really depends on what we’re trying to do and there is no one solid metric that says, “Well, this is what everybody should do!” We have these things like bounce rate and time-on-page and all these things because the software producers have to give you something when they give the software to count. They just put something there, they don’t know what it does. None of them use it either. They just assume that you’ll figure out something useful for it.
And yet conversion rate isn’t something that is just available out of the box. You have to set up your conversion goals and you have to do a bunch of extra work to get that and companies do that. They put a lot of effort into tracking different conversion events thinking that it is something meaningful.
Right, they do but the conversion rate doesn’t actually tell you about conversion. It doesn’t tell you about whether in fact you’ve gotten good customers or made us a positive difference. What it tells you is that your ratio has changed and you don’t know if the customers you’ve gotten are customers that will have better lifetime value. You don’t if those customers will be so delighted with your product that they will tell everybody or if they’re going to hate it because they thought they were buying something else because it was a bait-and-switch.
What about shopping cart abandonment? What do you think about that?
I think it’s a myth. I think that, when you actually watch people shop, you see very little instances of shopping cart abandonment. You almost never see people in e-commerce sites abandon their shopping carts. It’s some sort of problem in the system. People are not realizing they’re adding things to the cart or they’re not realizing that by putting things in the cart, they’re signaling an intent to purchase. There is something wrong in the interaction model of the system because when you sit there and you watch people side-by-side actually shop, you never ever see people put things in the cart and then say, “You know what? I’m not going to purchase any of this stuff!” It just doesn’t happen. Something else is going wrong in the experience. In many cases, people can’t tell if putting something in the cart actually ends up with the purchase or not. They can’t see whether that happens so they’re at a loss for that.
You showed a case study in your presentation that there are some user login issues that were causing out a great deal of shopping cart abandonment but the correct metric to track here to discover where the issue was wasn’t’ shopping cart abandonment, I think you called it unrealized shopping cart revenue from user login issues’ or something like that.
Yeah, so this was an instance where the client, which was a big bucks retailer, their website was doing billions of dollars a year—about $1.2 billion. They were seeing this massive shopping cart abandonment and they just assumed that this was just a normal behavior and what we learned when we were watching users was that, people would login to the website or they would go to the website, they would shop for their products, they would put them in their cart with every intention to buy them, and they would start to buy them and at that moment, they would run into this problem of the shopping cart or the system telling them that they need to login but they couldn’t remember their username or password so they would try their username and password, they’ll get it wrong, they’ll get an error, they’ll try it again, they’ll get it wrong. Finally, they’ll try to reset their password which then required that they put in the email address for the right place to send it to but they don’t know if it’s the right place to send it to because they don’t remember which email address they signed up for the service at five years before. Finally, they would, maybe, get an email in their inbox and they will click on the email and it would let them reset their password and then they could continue on with shopping.
The first year’s revenues increased by $300 million just because of that change.
What we were seeing was, there was a lot of customers who were getting stuck in that password-mismatch reset loop and they weren’t getting out of it. We saw this because we’re just sitting there next to them, watching them shop. What was fascinating about this was, we then asked how often does this happen in the real world and what we found was that, it wasn’t happening in the real world infrequently and that it was actually happening quite a bit. In fact, when we added up all the shopping carts of all the people who had put in a username and password, gotten an error, and then never successfully logged in—it was close to $300 million dollars every year. And so, the site went on to add a button that let people log in as a guest, without having to say their username and password and as soon as they did that, they saw and they realized $300 million dollars of revenue immediately from that. The first year’s revenues increased by $300 million just because of that change.
That’s amazing! That’s huge! And you wouldn’t have discovered that without watching users shop and get stuck at that particular spot because folks were tracking the wrong metrics, just looking at shopping cart abandonment, trying to solve a problem without seeing what’s happening behind the scenes. So, let’s talk about watching users shop or use a website. I mean, how do you refer to it? User—
User experience work.
If somebody is not happy and they’re getting frustrated, you have a really cool way of getting them to give you that feedback. You have a coin with some tape that they slide. Can you describe this for our listeners?
Yes, it’s a simple little trick. I just take a piece of masking tape that’s about 2-3 inches long and on one end of it, I draw a happy face and then on the other end of it, I draw a frowny face and then I put five little marks evenly spaced between the happy face and the frown face. Then I give people a quarter and I put it in the middle. I ask them if the coin is in the right place and they just take it and then they slide it to the frown side if they’re frustrated or the happy side if they’re delighted and as they’re using the website, or the product, or the service, I just ask them to move it. It’s a simple little hack and it tells me, in real-time, how frustrated or delighted they are and I can ask them why, which is really important.
And do you do this every few minutes or when you start seeing them getting—
Yeah, every time something important happens. If I hear them sigh or say something that lets me think that maybe it’s changed, I will ask them. And overtime they’ll just start moving it on their own when you ask them frequently enough. Every time they get to a new part of the website, is the coin in the right place? And they just move it. Then, I’ll ask, “Why did you move it there?” and then they’ll go, “Well, I’m not happy with, you know, I think I shouldn’t have to do this.”—okay, good.
And these are not just people off the street. These are people who are in the industry, who buy perhaps from the same company or from competitors. These are—
Yeah, these are customers. These are qualified people, I mean, it depends on the study we’re doing, but they’re often qualified customers of the organization. These are the people who should be buying, if we’re doing an e-commerce study. People who should be signing up for a lead, who should be converting. Everything about them is who we want as a customer and if for some reason, they’re not then we want to know why they’re not. This isn’t just some random person who might or might not be interested. These are highly qualified folks with the product or service.
So, you’d given us an example on your presentation of how designers, if they’re chasing after the wrong metric maybe just to keep their job like monthly active users—you really railed on that metric—that can create a really poor user experience which will, if in user tests, the person will be sliding the coin to the frowny face all day long because it’s such a frustrating thing. One example you gave was, Instagram when that’s hooked up with Twitter, which is just a simple settings tweak, and then now Instagram the tweets that show up as you are pushing out your Instagram posts are basically, hobbled to not include the pictures even though it would be a better experience for a Twitter user to show the pictures. They specifically break the experience for the user in order to increase monthly active users. Could you talk more about that?
A lot of Silicone Valley now are chasing these metrics that are really sort of false metrics.
A lot of Silicone Valley now are chasing these metrics that are really sort of false metrics. You see this a lot with businesses that don’t have a real business model like Instagram where users of Instagram don’t give Instagram any money. Advertisers may give Instagram money but money that Instagram gets from its users are not there. They have to somehow measure the success that the users are so they measure this thing, they create it, and then they called it: monthly active users. On the surface, it seems like a good idea because in the surface, it feels like, “Well, if someone is active on our site, they’re more valuable to us than someone who’s not active.” The problem comes in when you take that to its logical conclusion and you start to design to make it happen. The problem is, the marketers are hobbled and they can’t tell whether someone who sees a picture on Twitter is using the service or not. If they just view the picture, are they using it? Well, if the purpose of the website is to publish pictures and have people look at them, one could argue that looking at a picture is a use instance and that people who look at lots of pictures are using it.
However, they can’t measure how many people see the picture on Twitter so instead, they say, “Well, we’re not going to show you the picture on Twitter. We’re going to make you, the user, to click through the website and we’ll only show you the picture on Instagram because there we can count you and we can show how many people are so excited to be on Instagram if they are clicking on the pictures.” The reality is, they’re actually getting less use because a lot of people aren’t bothering. They see the link, they don’t know what it means, they think, maybe, it’s a trick, they don’t understand that there’s a picture on the other side. You know, there’s all sorts of things that create a problem here so they’re not actually engendering more loyalty to the service but they’re actually inconveniencing their users. They’re doing it because it’s the only way they can measure whether in fact people are using the service in the first place—and that’s because users don’t actually contribute any monetary value so they can’t measure what an e-commerce site would measure, which would be actual number of sales. This is the problem of most advertising-paid websites. They start to rely on these crazy metrics that are independent of the actual use scenario.
You see designers doing all sorts of things like making people click on links, or putting stories across multiple pages, or putting in slideshows that make you click 10x so that they can get ten more times impressions. Every user hates them and every advertiser hates paying for them but the publishers love them because it makes it feel like something’s happening and it creates a bad experience. Going back to brand, this is not helping your brand. This is not engendering favors from those folks. It is actually frustrating them and your brand is always hurt when you’re frustrating people. It’s not necessarily the right thing to do but when you have no other metrics, you tend to do it and what we see is that, when you have no other metrics, you build these things in and you end up going down this road and the downside is that you create the worst experience and you end up shooting yourself in the foot. This is all because you didn’t build a decent business model to begin with so if you don’t have a good business model to start with, you’re going to run to these problems.
And if you’re not measuring and tracking frustration levels of your users, you’re also going to be blind to this.
So, what would be some metrics to track frustration? Error message numbers and things like that?
Every time someone gets an error message in your system, you’re frustrating them for some reason.
Yes, so every time someone gets an error message in your system, you’re frustrating them for some reason, right? Nobody gets an error message and is excited about it. Maybe they put in the phone number with dashes and spaces and you’re making them take that out. Or, maybe they entered their credit card number wrong. Or, maybe they got their username and password wrong. Whatever it is, you’re making them go through work. Do you know what the most issued error message on your website is? Most companies don’t and yet, those error messages are causing frustration—they’re creating support calls, they’re generating all sorts of cause, they’re losing sales, there’s all sorts of problems associated with them and yet, you don’t know if they’re spiking today— so every error message should be instrumented and that should be a dashboard. The right element should be is, how many error messages did your customers get while trying to buy your products?
And that goes back to this idea that if you spend money on retaining a customer, that’s going to be money better spent than on bringing a new customer because the cost is way lower to retain a customer than to bring in a brand new one.
Exactly! And you know, a lot of that costs are things like what you’re spending on delivering an error message and losing a customer as a result of it.
I know we’re out of time here so this has been an incredible wealth insight into how to drive better user behaviors and get them the result that they’re after then do this in a way that enhances the bottom line of the company at the same time. How would somebody who wants to perhaps, work with you and your company, UIE—how would they learn more? Where would they go?
They can learn more by visiting us at UIE.com. We have articles and the talk that I gave is actually available on the website. There’s a section called Jared Live where companies can go and see if they want to bring me in. All the talks that I give are up there so they can go visit that and see. This one which was, Is Design Metrically Opposed? and other talks that I’ve given on design and user experience, they can contact me on LinkedIn or follow me on Twitter at @jmspool where I tweet frequently about design and design strategy and the unusual habits of the airline industry.
Thank you again, Jared. Thank you, listeners. Be sure to check out the marketing speak website at MarketingSpeak.com for the transcript for this episode and the show notes. Included with the PDF transcript will be a checklist so you can take some actions from the learnings of this episode. Thank you all and we will catch you on the next episode of Marketing Speak. I’m your host, Stephan Spencer.
Your Checklist of Actions to Take
☑ Gather a group of qualified customers together and use the coin trick to rate their satisfaction while they are using your website or service.
☑ Consider marketing a lifetime activity and work on progressing the client experience throughout their whole journey, not just for the sale.
☑ Use the survey from Gallup CE11 for more personal metrics.
☑ Stay on top of error codes on your site; know why they are happening and fix the issues to prevent loss of sales.
☑ If you try the coin test, put together a list of questions beforehand. Make sure you ask “why” when they are moving the coin.
☑ Use a good experience to allow your website or company to market itself-ensure that your clients are happy, and they will refer others to you.
☑ Pay attention to shopping cart abandonment. If it’s happening, dig deeper into what part of the sale they are abandoning their cart.
☑ Have team members rate your leads on a scale of good to bad, to see the type of customer you should be targeting.
☑ Use your data to retain current customers- keeping the customers you already have should be your focus.
☑ Allow your website to work for your client. Have a guest tab for checkout, in case a client does not want to register for an account.
About Jared Spool
Jared Spool is a guru on usability, and is the founder of User Interface Engineering (UIE). He’s the co-founder of Center Centre and author of the book, Web Site Usability: Designer’s Guide. He’s also co-authored Web Anatomy: Interaction Design Frameworks that Work. He is the conference chair and keynote speaker of the UI conference in the UX Immersion Conference. He’s been doing usability since 1978, and is considered the “godfather” in the field. He helps clients understand how to solve their design problems, explains to reporters and industry analysts what the current state of design is all about, and is a top-rated speaker at more than 20 conferences every year.