Technology
I've met Rebecca (Linkedin) at the smind conference in Ljubljana Slovenia and I really enjoyed her presentation. She and the company she works for, look at e-commerce websites and learn from what works and what does not work (if nothing else, go and check out the FREE blog). The presentation has a couple of great ideas on how to minimize checkout abandonment. Here is the link for the mobile cheat sheet. The transcript of the podcast: Rebecca: 23% of users in one of our studies cited that a too long or complicated checkout process was a reason for abandoning the site. [music] Peter Mesarec: This is Time4Marketing, the marketing podcast that will tell you everything you've missed when you didn't attend the marketing conference. Hello, and welcome to the Time4Marketing Podcast, the podcast that brings you the best speakers from marketing conferences all around the world. My name is Peter and I'll be your host today for the episode number 29 as we're slowly ending the 2019 year, the second year of this podcast. Before I introduce you to an excellent guest that we have tonight, please go and subscribe to the podcast if you like it, and of course, rate the podcast in your favorite podcast listening app. We now have a website, it's called time4marketing.com. The number four is a number. That sounds very logical. On the web page, you can also subscribe to the newsletter so we'd send you interesting information about the podcast and marketing conference. Now, we go to our today's guests. We have with us-, I'm very glad to have Rebecca Hugo. Hello. Rebecca Hugo: Hello. Thank you for having me, Peter. Peter: Very glad that you are here today. I saw you speaking at the Slovenian Conference e-Commerce Day sMind conference. You are the UX auditor at Baymard Institute. What is it, what do you do and what do you do there? Rebecca: Baymard Institute are an independent usability research company. We specialize in helping other sites improve their e-commerce givings to their users. We do all of independent research. From there, we distill a lot of our findings. We found, I think it's 11,000 I think is our current number of individual issues that all users have come across when they're testing various sites across all industries. From there, we distill those down into-- We're over 750 guidelines at the moment. The number's still growing because we have a couple of research studies on at the moment. Those guidelines look to the design patterns that either positively or negatively are reacted to by the user. That is anything from a product detail page layout, to how filtering options are presented to the user, to how the checkout is or is not optimized, depending on the site. From there, my role as a UX auditor, our clients will come to us and say, "Could you look at our site?" Basically orders us, "Let us know what we are or are not doing well." I suppose it's almost like taking your car in for a service. It's, "Your oil is a little bit low, your windscreen wipers need tightening up, but the leather in your seats are fantastic." We'll do a similar thing with the order. It can be anything from just looking at a single section, doing the entire site, or even doing prototypes. It gets really quite exciting in an odd way, looking at how different industries present essentially the exact same information to their users and also the nuances thereof in those instances to really create a great experience for that user. Peter: I do a lot of SEO audits. When I begin my SEO audit, there's always the one thing that I'm going to go and check if it's done right, that's the canonicals and the language alternates. When you start such an audit, you probably have a workflow that you have to go through, but what is the one thing that you think that companies are forgetting about and shouldn't be forgetting about? Rebecca: That is a ridiculously hard question, Peter. [chuckles] It's so specific. Because depending on the industry for a start, so if you're looking at a gifting website, one of the core aspects of that is going to be wildly different from a heavy text-back website or anything. For example, selling a laptop or even a fridge, mass merchants and so on. One thing we can fairly consistently find that is still an issue with all the sites that we look at is their search. Their search is predominantly quite poor. That's anything from the varying types of search, your exact search, your feature search, you have slang, you have abbreviation, you've got thematic. All of these aspects across the board, the search for majority of e-commerce sites is still surprisingly weak. There are a great deal of users who just prefer to use search. Knowing that if your search is particularly weak, then not having a particularly good category taxonomy to back that up, it can cause just so many issues for that user. It's still consistently interesting to find and look at what the search landscape looks across the majority of sites regardless of industry. Peter: One of the SEOs on my LinkedIn feed was showing a couple of examples of, I believe UK e-commerce retailers who had no results when he searched for Black Friday on Friday. It seems that search is something that people are forgetting about or just using the default settings over whatever their search is. This is something that we often see. Very right. I've mentioned that I've invited you for this podcast because you had a very interesting presentation at the Slovenian sMind Conference. That was called the 6 Findings from Testing the World’s Leading Checkout Flows. Before we go to your presentation, how is-- That's a weird question because I live in Ljubljana. How was Ljubljana and how was the conference for you? Rebecca: Ljubljana was-- It was beautiful. I sadly didn't get as much an opportunity as probably deserving of such a beautiful city to really look around it. One of the representatives from e-commerce sMind was so-- Sorry. e-commerce-- I can't even say it now. Shopper's Mind e-Commerce Day was kind enough to actually take myself and another speaker around a little bit one of the evenings, so we did get to see some parts of it. It's such a beautiful city. I would really like to visit again. The day itself was great. The atmosphere was fantastic. Everyone was so kind. It was wonderful having just people being comfortable enough to come to you for feedback for a start. You never know how these go unless someone actually tells you exactly what they have or have not been able to take away, any improvements and so on. It was a lovely crowd. It was really well put together. It's a real testament to what Ljubljana, Slovenia, and obviously the Croatia side of things as well and what the company has been putting together. I felt very touched to be able to have the opportunity to come and speak. Peter: Excellent. All right. Let's go to your presentation. 6 Findings from Testing the World’s Leading Checkout Flows. Rebecca, here are your five minutes. Rebecca: Okay. Obviously, the checkout is such an integral part of any e-commerce store. If you can't purchase online, it's not really e-commerce. Having a robust checkout that's really going to perform well for a user is so important. What we really found over a lot of our data studies was that 70% of users who put something in their cart would end up abandoning it. That's two-thirds of users we're going through all the trouble of finding a product that they liked, added it to their cart, but they'd still ultimately not purchase it. If you took away all of those users who were simply not ready to purchase, which is completely fair, there's only so much that a site can do about that, but when you look at the reasons that were left, so many of them could be improved with relatively simple checkout optimization. Some of the core things that we were particularly interested in, one of which is checkout length. 23% of users in one of our studies cited that a too long or complicated checkout process was a reason for abandoning the site during a checkout process during their checkout flow. Our most recent 2019 checkout UX benchmark, which is something that Baymard does, we look at 60 top-grossing US and European sites and use those to take a look at what the landscape looks like for e-commerce UX, we found that the average e-commerce site in 2019 has 12.8 form fields within their checkout flow. This may not seem like a lot, but considering you could actually essentially half that number, sites could get that number down to six to eight form fields for a guest checkout. 12.8 is actually quite a lot. There's such a disproportionate amount of time that users will spend with any open-text form field. Increasing time, causing issues, causing errors, that being able to minimize the amount of form fields, essentially the amount of tasks a user has to do can really create an improved performance and improve experience for the user. The other thing that we found really quite fascinating, at least I know that I did, was the perception of site security. Because the perception of site security can be just as, if not even more impactful than the actual site security that is present. 17% of users in the same survey, they abandon the checkout process because they just didn't trust the site with their card information. Users, we found we're believing that part of a page, so if you feel the box and area in [unintelligible 00:10:46] were more secure than other parts. Even though from a technical standpoint this doesn't make sense, the page is either encrypted or not, the fact that we were aware of this fact from our users, we can then leverage this misconception. Creating a visual robustness, leveraging the importance of site seals, and also what sites seals are more beneficial. A fascinating thing, we've found that some large companies were not given the same weight as just a simple padlock because that's something that users recognize. Knowing about these particular instances and how users react to that information can be so powerful in how we're able to create a comforting and a secure experience, as well as a good experience for our users. Finally, just a simple factor of mobile keyboard optimization. There are still issues across so many sites that are simply just not optimizing the mobile keyboard for their users during the checkout, and with the occasional even dire consequence. In fact, granted that dire is quite a drastic term, but that is exactly what it feels like for a user who doesn't receive their package. Anything from it needing to be for a big event or a wedding or a birthday, not receiving something that you're so looking forward to can be really quite detrimental. We're finding that even something as simple as not disabling the autocorrect feature for fields that don't benefit from it, name or address fields, that can result in the dire consequence of not receiving a package. It's something essentially simple. Our data is showing that 79% of mobile sites are not disabling autocorrect for those fields. On top of that, it's just needless friction from not finding or utilizing the optimized keyboards for email addresses, telephone numbers, credit cards and so on, the alphanumeric, numberic, the @ symbol and so on. Even on top of that, doing it consistently. 25% of mobile sites weren't consistently evoking optimized keyboard. For example, credit card came out with numeric, fine. As soon as you entered security code, it'd go back to alpha. Again, it just really comes down to this needless friction. When the companies really pay attention to that and how you can alleviate it, it can just make such a difference to the site. Peter: Thank you. I have a couple of questions coming here. I've listened to your presentation and the idea of thinking about the number of fields and the checkout compared to the idea of the number of steps for the checkout was something that was enlightening for me. I was like, "How did I ever not think about that?" Of course, I went home and started minimizing the number of fields on my website. I came down to, I think what was six fields. The question that I had then while watching the data coming in was, "Did I go too far?" I would like to hear your thoughts about that. I feel that customers are a bit used to having a bit bigger number of fields where the post number and the city are two different fields, and maybe name and surname are two different fields. I somehow felt that people are now misunderstanding my checkout fields. Is this something that you also see in your tests? Rebecca: Things like it will come up occasionally. It's part of the reason for a good placeholder text, good tooltips. Whenever a user does encounter something, not unusual, but something they're not necessarily used to, it can immediately be a little bit jarring. Microcopy is such an important aspect, especially within the checkout. I think a lot of sites don't necessarily pay attention to their microcopy the way that they should. There's a reason why UX copywriters are becoming such a big career choice within our industry. I think it's so important and something that's just not paid attention to the way that it should. There are multiple reasons for anything, [chuckles] sadly. If users are used to seeing 12, suddenly seeing 6, it can be jarring. That's not to say that it's a bad experience because it's jarring, it's just a new experience. There will always be a little bit of habituation time, but it's knowing that what you're able to actually offer the user is that improved experience. Bolstering the microcopy, bolstering the placeholder text, bolstering the tooltips if appropriate. It's also just determining what is actually useful for your particular site. If your site is niche in any way, six to eight is the average for "a typical e-commerce site". It's ensuring that you don't overly assume anything to be typical. If you're a gift predominant site, then changing your address fields or matching your billing to shipping address by default isn't immediately beneficial. If you've got something very important for an industrial site, then yes, you might need company details over just having a standard address. That is the nature of heuristics, heuristics are a rule of thumb. We are finding that, more often than not, these are the best situations to be and these are the best patterns to follow. However, never ignore the niche that you're in if you happen to be in a niche. Peter: The workflow is go and check out the expert findings that you have on your website, then change it on your website, but still measure the impact and see if it can be directly used on your website or not. Rebecca: Yes, in a great deal. Not only look at we're suggesting but why we're suggesting it. With all of our data, we try and back up with what the issue is, why it is an issue, so what we're seeing during all of our research to lead us to the conclusion that X is happening, therefore implement Y should alleviate. Because the other side of it, and it's something that we will do during auditing is mark something as issue resolved. The site may not necessarily be doing what we have specifically recommended, but they are circumventing the issue through another implementation. It's bearing that level in mind as well, whether what we're suggesting of the 774, 767? We'll eventually keep track of that number. There will be a great deal of these that just simply won't be applicable to your industry or to your site. There's always going to be a little bit of a pinch of salt because that's the nature of heuristics, but that's not to say don't pay attention to what we have found. Peter: Especially in e-commerce sites that feel that search engine optimization is important for them, we can usually see people starting to add category descriptions to their category landing pages. Whenever people start adding that to their e-commerce sites, the question comes up, should we have content above the products or should we have the content below the product? Should we have the content hidden and "read more" button or should we not? Do have any data on how that content can influence people coming to the category pages? Rebecca: Our data at the moment and our research findings, we don't have anything specific with category descriptions. What we do find is what that page looks like. For example, the filters on the actual main list are out of line, or if something looks too much like an advertisement or a promotion rather than actual benefit or actual product, it can mislead or distract users the same way that banner blindness will just mean that they pay absolutely no attention to it. With a lot of the information, and UX can often stand not in the way of SEO, but they're not always aligned with their needs quite often, that it's very important to essentially just understand what is the user needing to get from that page. Anything that a site therefore needs to do from a company perspective, is it stopping that user being able to do the fundamental task that they're looking for? If that is needing to filter and needing to understand the amount of products on a page, needing to create a visual comparison against products. Whatever the size or company needs to add from, again, a company perspective, as long as it doesn't stop the user being able to complete their task, then it really is a design preference. Anything too large that pushes down the content of the page-- Because many users, they're are not adverse to scrolling when they've determined a purpose, so going through, say 100 products on a visual push site like apparel, that's fine. Users are quite accustomed to that. If it's taking them a disproportionate amount of time to understand what the page is actually doing or selling them, then that could be off-putting. If that category description is particularly long and showing that above the fold, then that could possibly-- Again, we don't have any hardcore data to say it will, but that could pause the user or hinder the user from being able to understand exactly what it is and what is the page they've landed on. Peter: All right. You spoke briefly about the mobile keyword optimization. I'll just add the link to you. If I remember correctly, you have a page where you gave examples of how mobile fields should be optimized on your Baymard website, is that right? Rebecca: Yes, we do. It's on our little cheat sheet. You can just find all of the code snippets and the attributes to establish and actually implement the best way to optimize these keywords. Some of them like card filled for example, they don't have a direct code phrase as it were to trigger numeric, you have to do it slightly more manually. Peter: All right. Excellent. Rebecca, I think that's it. Where can people find you? Do you have conference plans? What are the social networks where people can find you and your company? Rebecca: You can find us on LinkedIn. You can search for Baymard Institute and you can search for myself, Rebecca Hugo on LinkedIn if that's your prerogative. At the moment, we don't have any specific conference plans for Europe, but if you are interested in getting any of us in the company to speak, please don't hesitate to be in contact. If not, we've also got Twitter. Baymard Institute, we're also on Twitter. We love a bit of a chat. You're more than welcome to contact us there and we look forward to speaking with anyone. Peter: All right. Excellent. Rebecca, thank you very much for being on the podcast and talking about the UX of e-shops. Next time you're in Ljubljana, we definitely can meet and [inaudible 00:22:31]. Rebecca: Fantastic. Thank you so much for having me today, Peter. Peter: Have a great day. Rebecca: You, too. Peter: Bye.