Designing a Cohesive, Not Consistent Experience

Richard Dalton, Raina Mehta, Mansi Kamdar, Sadia Harper at UXDX USA. Video: https://www.youtube.com/watch?v=OqOUYoycU0I

Readable transcript: edited from the recording's captions for readability (fillers and false starts removed, punctuation and section headings added). Wording is the speaker's own. Timestamps are positions in the video. Names marked [?] could not be verified against the audio.

Cohesive and consistent: two ends of a spectrum, or two different things?

[00:00:00] Richard: Good afternoon everybody, good morning, good evening wherever you are in the world. I'm super thrilled and humbled to be able to facilitate what I'm sure is going to be an amazing conversation. As Catherine just said, we've got three amazing design leaders on the call here, and we're going to jump right in because we only have about 28 minutes if my timekeeping is correct. So I'll ask each of our panelists, as they answer questions, to bring a little bit of their experience and introduction into those, rather than spend time at the beginning on bios.

[00:00:38] Richard: So let's jump straight in, and please welcome questions from the audience as we go through. But let me start off by saying, when you think about the two words cohesive and consistent, I'd love to get some perspective on whether or not we feel that these are two ends of the same spectrum, or are these two completely different things altogether? Maybe we could start with Mansi. Maybe you give us your thoughts on that and then we can go to the other panel members.

[00:01:13] Mansi: Sure, Richard. I would say before we look at whether it's two ends on the same spectrum or they're two different things, it's good to anchor ourselves as to what do they really mean. So when you think about consistency, you're thinking about, is it uniform, does it look alike, does it feel exactly the same as the other thing? And that's the goal for consistency. When you think about cohesiveness you're thinking about, am I offering a better understanding, or am I providing that clarity that makes sense and you understand what it means?

[00:01:48] Mansi: And what I would say, in my experience I found, is that they depend on what's the goal that you're trying to achieve. And that is very, very important, because on the look of it it may feel identical, but like I said there are two different goals that you're trying to drive through. And as you take this concept and you apply it to whether it's design or product development, I think you're looking at consistency when you're trying to replicate certain actions or behaviors so that they're easy to learn, and when they're cohesive you're trying to make sense of them.

[00:02:20] Mansi: So from that perspective I would say they are different goals that you're trying to achieve, kind of ends of different, you know, on that spectrum, and depending on the problem that you're trying to solve is what you're going to approach. And often I find that consistency is kind of the byproduct of trying to be cohesive. But that's what I would say, I've seen that happen in my experience.

[00:02:41] Richard: Raina, I can see you nodding along there. What are your thoughts here?

[00:02:45] Raina: Yeah, I think in particular Mansi's last point about it being the byproduct is, I think, a hundred percent. I mean, when I think of cohesion I think of it as a higher order, it's a strategy. The idea is to align an experience in a unified way, so taking all of the different tactics and elements of an experience and bringing cohesion to them so that you are helping the user stay on path and meet the goals. That is where cohesion is really important.

[00:03:22] Raina: In a lot of cases the tactic that you would employ to help a user get to their goals would be consistency. But I think of one almost as a subset, or Mansi's word byproduct is the right one. Consistency isn't necessarily the goal. Cohesion is the goal. Consistency is one way to help a user meet the goals that you set out for them.

[00:03:52] Richard: That's great. Sadia, I saw you nodding along to some of the points about cohesion in there. What are your thoughts on this?

[00:03:59] Sadia: Yeah, I'm in total agreement with both of these wonderful women. I think they're different things altogether. I feel like for me consistency speaks to execution where cohesion speaks to approach. And I think, Raina, you said it when you're talking about strategy, and I'm a UX strategist and researcher and that's what I try to do. I think when it comes to it, you actually want to deliver a consistent experience where the people that interact with the products don't experience unnecessary frictions, or have a sense of confidence when they use our products. I think cohesion is established earlier with that strategy, so delivering that understanding and insight where everyone who's building that experience is coming with a very cohesive understanding of who we're designing for, in cohesion, and what we want to build for them.

When can consistency be detrimental?

[00:04:42] Richard: So from these three points of view, I'm getting the sense that cohesion is generally always good. It's not like anybody says I want to build a very incohesive, or I want to build an interface or experience that isn't very cohesive. But what about consistency? Are there times that it can be detrimental to the user experience?

[00:05:08] Raina: I will jump in and say I think so. I think there are different types of consistency. There's visual consistency in the design of the branding, and there's also consistency in features. And that's where I think one of the common experiences that I've had with different teams and products that I've worked on is the mobile versus desktop experience. Consistency between mobile and desktop is not something that you want to aim for, because you don't want to clutter a mobile experience. You have limited real estate, and so there are features that sort of get dropped from a desktop experience going to mobile.

[00:05:51] Raina: And so I think that consistency for the sake of conformity is not always helpful. It can create noise in the experience and in the design, and you have to be really thoughtful about where you're really aiming for consistency and making sure that it's not going to be detrimental to the experience.

[00:06:13] Mansi: And I would say, Richard, I think, you know, what Raina said: if you're just achieving consistency for the sake of consistency, I think that's where it's really detrimental. I would bring a different perspective from a consistency standpoint in some cases, because it could be a byproduct of your experience, you may achieve consistency as a result.

[00:06:37] Mansi: So I would look at, when we work at Walmart and you're thinking about walmart.com, the customer's engagement when they're shopping for no matter what product, you're going to go through the payment and the checkout experience, and you're always expecting that to be a consistent feel or an experience that you would land with. But at the same time that's not the goal. So you don't start off with saying, well, I'm going to make everything consistent so that every customer that comes in and shops with us is going to see and look and feel everything exactly the same.

[00:07:06] Mansi: But I think what you really start off with is to say, what is it that I'm trying to bring to my user? So I'll use an example. If you're trying to buy a phone on the website, or you're trying to buy a gallon of milk, they are two different personas, the mindset of that user at that point in time, and you want to design for that. And then you would achieve consistency where it makes sense, because there are very, very standard behaviors that you want to drive out of that experience. So it can be detrimental in the case where you just start off to say I want everything to look consistent and then try to design your product from that angle. It can really hurt you. But then you achieve it where you see it's necessary.

[00:07:48] Sadia: Yeah, I totally agree. I feel like when you said goals, that's so huge, because hopefully early on you've identified the right goals for who you're designing for, like their needs, where they're coming from. If you've done that successfully then hopefully your consistency won't hurt them, where you choose to implement it. And if you don't, then you do have a chance of it maybe not being so good, if you're not being consistent for the right reasons. Totally agree.

Consistency across brand and every touchpoint

[00:08:14] Richard: I think that, Sadia, when we spoke yesterday, we talked a little bit about your experiences in the brand world. And when I think about consistency, it's interesting to think about, so what do we want to be consistent? We've spoken about different levels of consistency, or types of consistency here. There's that consistency from a user experience and a usability perspective, but then there's also consistency from a branding and a narrative and a content perspective. Can you speak a little bit about your experiences there?

[00:08:48] Sadia: Yeah, that's something I've seen where I've worked on teams where we've had consistency from a product design standpoint, and that is great, like all of our features are really singing and speaking the same language. But then what we find marketing is doing in terms of those assets or ads being developed, or even social campaigns, completely different language. So I think it's, again, how do you do consistency not just within product design or UX but kind of scaling up to that larger aspect where it's every touchpoint? And so how do you make sure that what we deliver from a brand perspective really permeates through every element and actually is experienced or expressed, again, through certain features, through animations, and then through the content that you choose to put out there in the world outside of your product as well?

Consistent to what standard?

[00:09:32] Richard: That's great. I think about, we've talked about different types of consistencies you can see here, and when I was preparing for this panel I was thinking about what consistency implies, which is that you have two things and you want to compare them to each other. You have to be consistent to something. And a lot of times in my own past I've thought about being consistent as in, we're looking at one thing that we have made and we're comparing it to other things that we have made, and are they consistent?

[00:10:02] Richard: But what if we were thinking about being consistent to a different standard? What if, instead of being consistent to the things that we've made, we were more consistent to somebody's body of knowledge, or their expertise, or their understanding, or their task at hand? So maybe bringing more of that consistent treatment against a different standard in mind. How have you all used design practices and tools and techniques to introduce more of the user view into consistency, rather than just this comparison of things that we have made perspective?

[00:10:54] Mansi: I can definitely start with this one. I could say, in terms of the tools that we use, I think in some cases when you're trying to understand our customers, our users, or even associates, we build applications or products that are used by in-store associates. So there are different roles that they're playing and how they're interacting with the same application.

[00:11:17] Mansi: And I think it's very, very important, I think we can't emphasize enough sometimes, to say understand, understand and understand your user. If you understand what's the role that they're playing, where are they coming from, how are they going to interact with your product, that will allow you to then go in and define that experience for them. So I think definitely starting with that understanding, having the research around it, doing some strong discovery and understanding that process of how they interact with our products, becomes a very important tool that we work with our design teams on, and we work in that collaboration.

[00:11:51] Mansi: I was listening to a talk earlier and they mentioned that very often discovery or research isn't done at all, or in cases they're done in the beginning, in the middle and the end. And the whole continuous discovery and learning process as a tool becomes very, very important as you work with engineers, you work with product, you work with design teams. And thinking about those practices that you make as a part of your standard set of tools that you want to use becomes extremely important in defining those cohesive experiences.

[00:12:24] Mansi: And then for certain actions, like if you were going back on a page, or let's say you were going to delete something, there would be a certain consistent understanding of what that action is. And I think that would be one type of consistency, Richard, you were referring to, where you would look at that action and say, yes, can I derive some type of consistency that I can ensure is in place?

[00:12:50] Raina: I would say that in addition to that, two examples come to mind. One is that we have used, in our initial conceptual testing of ideas, as we're trying to figure out how we want an experience to shape up, we've used competitor models in our testing, just to create prototypes, to understand how existing paradigms, let's say in video, you might have a video carousel and it's pretty standard across YouTube and other video providers how video is engaged with, how discovery happens of additional videos.

[00:13:38] Raina: And so we might use existing paradigms as successful models that we don't have to recreate. We don't have to recreate the wheel here. There's a consistency that we can bring in from other models, because they're so prevalent and so highly used and users are so trained in this, that it helps us accelerate our discovery, accelerate our design and experience, our wireframing and user flows. So I think leveraging best practices, and again, consistency is really helpful when you are trying to get users up a learning curve quickly, and you know that the large majority of them have embedded mental models, you can utilize those.

[00:14:18] Raina: The other thing that we've thought about at Kaplan is, as we're moving people out of in-person learning, particularly during COVID, and into digital learning experiences, how do we utilize a really successful model of in-person teaching and learning with a teacher, where it's very interactive? And how do we look at other platforms besides maybe Google Meet and Zoom, to create more of a dynamic that emulates the classroom? So bringing consistency from the real world experience, because that's a successful model, into the digital experience and trying to emulate that is another way that we've looked at consistency, as again a way for us to get started and start testing based on a model that works.

[00:15:08] Sadia: Yeah, I think the way we've looked at it at Bright is looking at consistency from an execution standpoint and then from a process standpoint. So I think in execution it starts as high up as the vision, so the vision we have for the company, to the mission, to our design principles, to our design system, to really put those guardrails and tools in place so our UX teams can just move forward as quickly as possible, again, working within the same space to create that consistency.

[00:15:36] Sadia: I think process is a huge part of it too. Process is, in every project, really identifying do we need to do exploratory research, generative research too, so we have the same amount of understanding. We've even started implementing UXRDs[?], which I had used previously in Intel, but I feel like they're really helpful now. We're just understanding, how are we approaching this, the same consistency around how we identify goals, how we communicate goals or articulate them, what does success look like. So even our evaluative measures, making sure we're doing that in the same way, how are we evaluating not just our work but also what does success look like. So really using the same parameters for every feature around consistency. And I think if you do that then you're kind of approaching each thing the same way, and of course they're going to be different projects, but if our approach and our process has some consistency there, I think that's going to turn out to have a more consistent product as well.

Audience question: what role do design systems play?

[00:16:33] Richard: That's great. We've got a couple of questions from the audience here that we're going to try and address. So the first one is asking about what role design systems can have as a mechanism or tool to ensure cohesion or consistency that's meaningful. Who wants to speak to that first?

[00:16:56] Raina: I can definitely take that one. So we definitely use the design system at Kaplan. It's been one that has been evolving. So I think that design systems are really important for so many reasons: efficiency, not just in the discovery and design phase but also delivery. The engineers are using these design systems, and so the efficiency operationally that it creates is important. But obviously it also creates that known sort of understanding and heuristic for the user.

[00:17:33] Raina: And I continue to think that that design consistency has a very important place, but as I said in the beginning, ours is constantly evolving. So I would only say that a design system has to continually be refined, and it should be learning from the testing and from the usability research that you're getting back. It shouldn't be something that stays stagnant. And that's where you don't want to stay consistent, again, for the sake of consistency. It's really important that you're measuring against KPIs and goals, and when it doesn't work, that you're looking to evolve it.

[00:18:17] Richard: Yeah. Mansi, anything to add against that?

[00:18:21] Mansi: No, I would say I go with what Raina has said. I think we do adopt design systems here at Walmart as well. They are continuously always evolving, I think, and I agree with that. So I would just echo, I think she said it really well.

[00:18:36] Richard: Yeah, I think from my own experience as well, I've seen that it's important to know when to use components of a design system. It's also important to know when not to use components from a design system, because in certain cases there can be this desire to, like, well, we've got this component that almost works for this situation so let's use it, because we want to be consistent and we want to be efficient, we want to be good reuse and have less technical and experiential debt, so let's use it. Doesn't quite fit our circumstances, let's use it anyway. That's where we run into, I think in many cases, this kind of foolish consistency. And so it's important, from my experience, that the design systems, as Raina was saying, be flexible and extensible so that they can be tailor-made for the design problem that we're trying to solve.

Audience question: how do you make cohesion efficient?

[00:19:27] Richard: This may well be somewhat related, so it's a good follow-on, but another question is, and I'll just read it verbatim: if our team goal is to get rapid feedback and make significant change that's a better experience for the customer and the business, what's the best approach to making cohesion efficient? Now, I thought the interesting take there, the word that was used there, is cohesion efficient, not consistency efficient. So when we're agile-ing our way to customer experience, how can we make sure that we are keeping everything within that larger picture? Because from a cohesion perspective I think it's oftentimes about how things fit into a larger whole, right, how things can be cohesive. So, without putting too much overhead into the agile process, how do we do that? Sadia, I can see you smiling. Yeah, it's so easy, because you've solved this problem, or you haven't solved this problem.

[00:20:32] Sadia: No, I think it's so interesting when I hear that phrase, rapid feedback. My question again, as a strategist and a researcher, I was like, well, what's the feedback for, what are we doing it to inform? And so I think not all feedback is used the same way. So if we're trying to understand how someone is using the product, or in what context, then that's helpful. But if we're trying to get a kind of deeper knowledge into, like, is it working, is it not, or actually, what's the other way around that was more deep, is it working, is it not, is something else? I think we have to be really calculated in what feedback we're trying to obtain and how we're going to use it.

[00:21:10] Sadia: And I think that's different for cohesion, because I think when I think of cohesion, to me that means more like understanding and thought. So understand, like, okay, how is this working for the person, is this meeting them where they are? And that's a different type of research and feedback. So I would say not all feedback is the same, and that's how I would approach them that way.

[00:21:29] Sadia: I think it's just so interesting, because we never just ask questions just to ask them, they're always informing something. So are they informing something from the consistency or product standpoint, or is it informing something from a larger strategic place? I think that's what we want to know, because that's where the two are different, cohesion and consistency. Because I feel like, again, cohesion being around understanding, consistency being around execution, I think it's much easier to establish consistency when you have cohesion than it is to establish cohesion when you have consistency. So I do think one comes before the other. So when you're thinking about that feedback, think about what it's actually being applied towards. Is it something early on, the strategy, or something later on, execution?

[00:22:12] Mansi: Yeah, and I would say in the question two things that kind of stood out for me was wanting that rapid feedback, so you really want that feedback loop really quickly, and then you're trying to get that cohesive understanding of how is this going to work in the larger ecosystem. I think a few things, I don't know if there's, I wouldn't say we've solved the problem, I think this is an ongoing thing. As Sadia said, it depends on what goals you're trying to achieve. But I think starting with knowing exactly what you're trying to measure, like what's your key measure for this, what are the things that I'm going to track as a part of my OKR, and what is it that I'm going to consider success when I'm trying to get feedback? Having that understanding up front really helps.

[00:22:53] Mansi: And then I think in the aspect of wanting to get some rapid feedback, understanding the ecosystem of the users that you have and making sure that you have a variety of feedback that you're getting back, because you're probably serving many different roles that these users are playing. They're interacting with your system in many different ways, it could be channels, or it could be, as I think we had said earlier here, you're looking at it from a marketing standpoint or you're looking at it from a product standpoint.

[00:23:18] Mansi: So understanding, I think Sadia was mentioning that, having that understanding and getting feedback across the board, making sure those focus groups are providing you that feedback, and then knowing the questions to ask to meet that initial goal that you set out to do. So I think if you use some of these tools, having that core understanding of what are you trying to achieve out of this, having a good understanding of that user base and getting that segmentation right, that you're hitting them fast enough, and then getting that feedback from them to see where you land, I think that usually tends to help you. But again, it's a process that you learn, using these different tools together to see what works best.

When not being consistent is the better experience

[00:23:58] Richard: That's great. So we're getting close to the end here. Another question came in which has given me pause for thought as I was looking at it, which is, do you have experience that you had a better approach but not consistent? So the way I'm interpreting that, which may not be the way it was asked, is, are there times when each of us has had an experience where actually not being consistent was a better experience?

[00:24:26] Richard: Something that came to my own mind there is that when I've been listening to y'all it seems that consistency is about predictability, right, from a user perspective. And in many of the experiences that we're all designing, like e-commerce experiences or learning experiences and things, predictability can be very useful. But in some instances predictability is maybe not the goal, or is not the most useful approach. I think about entertainment type or gaming experiences, or even customer service experiences.

[00:24:56] Richard: I was, several years ago, in the checkout line, this wasn't a digital experience, and they couldn't find the price for a pack of soda that I was buying. And instead of asking me to go back to the aisle and find the price, or have somebody go and find the price, or whatever, which was my expected consistent experience, the checkout person just said, we'll just charge you 10 cents for that, how about that? And I was like, okay. They were empowered to create that delightful experience which was not consistent, but it was a great experience. So have you all had maybe experiences like that where consistency wasn't the thing to do?

[00:25:39] Raina: I guess I would say that generally speaking, the more multi-varied your experience is, the more you really want to personalize it around where that user is in the journey. And so different visits are going to be in different parts of the journey based on that particular visit. So it might be that you're presenting them with a different landing page, or a different sort of messaging or call to action, potentially every time they visit. Maybe that sounds a little bit too diverse and too confusing, but the point being that a first-time visitor, versus a student in our case who is a first-time student and coming into their learning material the first time, is going to get some kind of orientation about the materials and they're going to start at the very first stage.

[00:26:42] Raina: But if you're a student who comes in every week and does their learning, then we're going to pick you up from where you last left off, or we're going to put you in that next thing that we think you should do. And maybe today it's a live class, but tomorrow when you come in it's actually that you haven't done this reading and so we're going to drop you into the reading. So the more personalized that journey is, the less consistency they might have from visit to visit. But it doesn't mean it's not cohesive. Again, the cohesion comes in because we're keeping them on task to a goal. We need them to finish their coursework, and so our job is to deploy the right mechanisms at a particular time to meet the user where they are, to get them to that goal. And I think that is how I perceive the difference between the two.

[00:27:33] Sadia: And I would also say, in terms of that moment, there's always a reason behind it. So if I'm going to do something that's very atypical of what we've done before, it's either because in some way what we're doing is not working, and so either it's not serving who we're building our product for, or it's not serving the direction we want our brand to go in. So I feel like that's when you do it. You don't just mix things up for the sake of mixing things up, or being inconsistent for the sake of being inconsistent. There should be a reason.

[00:27:58] Sadia: And so I think that's the question I would ask. I think every instance is very individual, but knowing the reason why, and again that being cohesive with how everyone is thinking or should be thinking of, hey, we've all decided we're not addressing their needs so let's do something different. Or, I've noticed in terms of the marketplace we're being perceived this way, that's not how we want to be perceived, so let's try something different. I think you have to understand the why behind the action.

[00:28:22] Mansi: Yeah, and the last thing I would add is I agree with Sadia, you need to know the why behind the action. And I can definitely jog my memory to say some instances where we've chosen to do it differently is because, like if you think about, like I said, you're in a store, your application or your product is going to be used on a tablet. The form factor is very different than a web experience, and so you deliberately are not going to try to make them consistent. You're deliberately trying to make them cohesive, because you're going to focus on that and you know from the get-go that they could be different from that aspect. So there are instances that I would say where, if you know the why, you may approach to have them differently.

[00:29:03] Mansi: And I would still go back to that same experience. I think you talked about entertainment earlier, Richard, and entertainment is one of the portfolios that we have. And if you think about purchasing a cell phone, it's different than buying a gallon of milk like I said. So you're going to go in, you're going to choose your products, you're going to get approved for something that you can qualify to buy, and they're larger value purchases where it would be very different. So you don't want to drive consistency, and you know from the get-go you're trying to bring that cohesiveness into that digital experience that you're creating.

[00:29:32] Richard: Well, that's great. We're at time and I was going to wrap up and try and summarize, but I think you all did it better than I can. I think what I took away from your closing comments there is that it's all about purposeful consistency, right, that we should be asking ourselves, why are we trying to be consistent here? Is that a meaningful reason, a purposeful reason, or is it just because of habit or behavior? And we should question that if we find ourselves doing that, because it might not be in the best interest of the interface and the experience.

[00:30:01] Richard: Thank you so much for this conversation, I've thoroughly enjoyed it, I hope everybody watching has too. There were a couple of questions we didn't get to, apologies for that, but it's been a pleasure.

[00:30:09] Sadia: Thanks, Richard.

[00:30:12] Mansi: Thank you, Richard. Thank you, guys.