Why Frictionless Isn’t Always Better, and What Drives Customer Loyalty

Sam is Senior Manager of Service Design at LinkedIn and host of CX Patterns, which means he spends a lot of time looking at how experiences actually happen: across customers, employees, tools, policies, systems, expectations, and all the messy bits in between.

The conversation starts with a familiar tension. Businesses often need to make short-term decisions that may damage long-term loyalty. Sam uses airline cancellations as a clear example: sometimes they are unavoidable, but they still shape whether someone chooses that brand again.

From there, the discussion moves into good and bad friction. Bad friction is usually accidental, slowing people down because of poor design. Clunky processes, unnecessary steps, information customers shouldn’t have to hunt for.

Good friction is more intentional, giving people agency, helping them pause before a meaningful decision, or making a moment more memorable.

Things you’ll take away from this episode:

• The difference between effortless and forgettable
• The peak end rule (how people remember the highest point of an experience and how it ended) and why the moment that comes before the peak matters more than you’d think
• How employee friction can become customer friction
• Why being more human in your marketing is so important
• How other companies and sectors are using intentional friction to improve their customer experience

Listen now on your favourite podcast platform, or watch the video on YouTube.

Spotify
Apple Podcasts
YouTube

Intro – 0:31
Welcome to the CX Equation, a podcast by Tap CXM. We share actionable insights and real-world case studies to equip you with the tools you need to drive loyalty, engagement, and sustainable growth.I’m your host, Chantelle Casey and I’m Mark Clydesdale

Chantelle – 0:48
On today’s special episode of the CX Equation, we are delighted to welcome Sam Stern, Senior Manager of Service Design at LinkedIn. Sam has spent his career focused on improving both customer and employee experiences, leading service design work that examines end-to-end journeys across one of the world’s largest platforms. Alongside that, he hosts his own podcast, CX Patterns, where he explores key concepts such as trust, anticipation, and expectation-setting in the world of customer experience. In today’s conversation, we’re going to challenge a long-held assumption in CX – that all friction is bad. Sam has been exploring the idea that some friction, when designed well, can actually improve experiences. Without further ado, welcome to the podcast, Sam.

Sam – 1:29
Thank you for having me. Great to be with you.

Chantelle – 1:32
Firstly, let’s take a moment to understand your background and give our listeners some context. You’ve spent much of your career examining end-to-end journeys and experience design. What first got you interested in service design and customer experience as a discipline?

Sam – 1:45
To be honest, I had never heard of customer experience. I was working at Forrester Research at the time, which had a substantial customer experience research and consulting practice. I was taken with the win-win-win potential of it — better for customers, which is obvious, but also, when done well, it has compelling business benefits. And it’s better for employees, too. Spending your day figuring out how to delight customers, solve their problems, and anticipate their needs is a genuinely rewarding challenge. That really appealed to me because there always seemed to be an assumption that treating customers well and achieving strong business outcomes were in tension. Customer experience argued — with compelling data — that they weren’t. It was a both-and situation. The other thing I’d say is I’m drawn to topics that appear simple but aren’t. Customer experience is absolutely one of those. We have these simple maxims: treat people as you’d like to be treated, or the customer is always right. Both of those are actually wrong as a basis for customer experience — and we could explore that if you’d like. The idea of just being nice to people seems straightforward, but there are so many layers and nuances to what it means to treat people well across different situations, different needs, different customers. I find that genuinely fascinating. A lot of people treat it as a solved problem, and it really isn’t.

Mark – 3:23
Do you find challenges in the balance between business goals and customer goals? I work with clients all the time who are wrestling with the tension between hitting a short-term target and doing what’s best for the customer in the long run.

Sam – 3:46
That is the age-old challenge. We know something probably isn’t great for long-term customer loyalty, but there’s short-term pressure to meet a target. I don’t want to be naïve — there are many instances where the decision will come down on the side of the short-term business goal. In those situations, I think it’s really important to at least remind people of the trade-off they’re making. I’ve been researching airline flight cancellations quite a bit recently. Some are unavoidable — staffing shortages, air traffic control issues, severe weather. But the data are very clear: whenever a passenger experiences a flight cancellation, they are significantly less likely to fly with that airline in the future. So yes, some cancellations are unavoidable, but companies should still keep in mind that each one represents lost passengers — and lost revenue — further down the line. Those ten people on a cancelled flight didn’t receive the promise of being transported safely from one city to another. They are unlikely to book with that airline again.

Mark – 5:16
And they’ll tell their friends.

Sam – 5:18
Absolutely. And because that’s a real story — not a vague impression, but a specific thing that happened to them — it gets retold and remembered.

Chantelle – 5:37
So in those situations, it becomes about minimising the damage. If the cancellation is inevitable, what happens afterwards matters enormously.

Sam – 5:50
Exactly. Transparency and communication throughout the process are critical. If bad weather is coming and you can anticipate it a day or two in advance, communicate the risk — the probability of a significant delay or cancellation — and put agency back in the customer’s hands. Give them the option to rebook ahead of time, or to extend their trip if they’re already at their destination on a return journey. When you do those things — be transparent, give customers options — you reach the vast majority of people who are reasonable and who understand that bad things sometimes happen beyond anyone’s control. They’ll appreciate that you did what you could.

Mark – 7:00
That leads us on to the topic of friction. You’ve been exploring the concept that not all friction is bad, which runs counter to what many of us have been taught as CX practitioners. What made you start questioning that assumption?

Sam – 7:18
History is important here. The Effortless Experience came out about ten or eleven years ago, and I think at the time it was a necessary correction. There was this prevailing belief that every moment needed to be a surprise-and-delight moment. And the book pushed back on that — when someone calls in to reset their password, make it effortless. I agree with that. However, following that logic all the way to the conclusion that every experience should be frictionless is catastrophically wrong, and too many companies have applied that idea across the board.The analogy I always come back to is storytelling. The equivalent of a fully effortless experience in a fairy tale is the first page and the last page: once upon a time and they all lived happily ever after. There’s nothing memorable in that story. If there is no friction, no grit, no narrative arc or tension with a resolution, there is nothing to remember. The whole point of delivering a great experience is to create a positive emotion that the customer associates with a specific moment — one they will remember — because past experiences guide future decision-making. If they can’t remember their past experiences with you, they can’t remember to choose you in the future.

Mark – 9:25
So it’s like the rom-com — the couple has to fall out in the middle to make the reunion at the end meaningful.

Sam – 9:30
Exactly. Every genre of film has signature moments it must deliver. In a rom-com, if the couple doesn’t have a meet-cute, it isn’t a rom-com. And if they don’t split apart in act two before coming back together in the rain — in London or New York, for some reason — it isn’t a rom-com either. In customer experience, you can design these moments of friction intentionally. That’s the heart of what I call good friction versus bad friction. Good friction is controlled, designed, and intentional — you know exactly what that moment is. The canonical example is IKEA and the IKEA effect. There’s a named phenomenon for it: the confusing assembly instructions and the effort required to build the furniture make you feel as though you contributed to the process. You like the product more as a result. Effort is a form of friction. Rory Sutherland relates in his book Alchemy that when he began working with IKEA, they told him: if you ever suggest that we make our furniture easier to assemble or our instructions easier to follow, we will let you go. They know the friction matters. I wish more companies understood that.

Chantelle – 11:17
I can confirm from personal experience that the sense of accomplishment at the end of building flat-pack furniture is real — and it absolutely makes it memorable. Have you seen examples where removing friction has actually made an experience worse?

Sam – 11:41
There are many cases of good friction that I think most people would recognise as obvious. When a customer of a financial services company is about to sell a holding, there is almost always an “are you sure?” step. GitHub has the same thing when you’re about to delete a repository. Microsoft Word, before auto-save, would prompt you if you tried to close a document without saving. That is good friction in action. If you removed it in the name of making the experience more “effortless”, a small but meaningful proportion of customers would regret what they just did — and it would be too late to undo it. That extra half-second check is genuinely valuable.

Mark – 13:13
I remember the first time I clicked “buy now” on Amazon and it just happened instantly — no verification, nothing. It almost felt too easy.

Sam – 13:33
Right. And Amazon is an interesting case because for things you buy repeatedly, frictionless makes sense. But if you’re buying a book for the first time, and it’s available in three or four formats — physical, Kindle, audiobook — they should probably check that you’re buying the right version. There are so many small moments like that where removing all friction allows people to act almost without thinking, and mistakes accumulate that are so easily avoidable. You’re not gaining any meaningful advantage by being half a second faster at that point.

Mark – 14:47
So how do you make friction feel intentional rather than frustrating?

Sam – 14:52
I think you want customers to feel that the friction is putting them back in control — giving them agency. One of the more striking examples of this is self-exclusion in casinos. In the US, problem gamblers can voluntarily ban themselves so that no establishment is permitted to allow them to play. To me, that’s the ultimate form of customer agency — acknowledging a limitation and creating a structure to address it. A less extreme version is Instagram prompting users who have been scrolling for a while to consider taking a break. It’s a perfect form of good friction: it interrupts the doom-scrolling loop and gives the person a conscious choice. They may choose to keep scrolling, and that’s fine, but the interruption restores their agency. So, to summarise: first, use friction to put agency back in the user’s hands. Second, slow people down when the action truly warrants a conscious decision — selling an investment, deleting a file, closing a document without saving. Third — and this is where it gets more provocative — you can introduce friction at key moments to make experiences more memorable. An example that’s stayed with me for nearly fifteen years: I was at an Enterprise Rent-A-Car airport location. We came off the shuttle bus, joined the queue, and there was one person working behind the counter. Everyone grew frustrated. Then, about ten or fifteen seconds later, a door opened and six members of staff came out. The queue vanished instantly. I saw the same thing happen at another location a few months later, so I asked about it. They admitted, without quite admitting it, that yes — it was deliberate.
I call this the “Enterprise pause.” They added ten to fifteen seconds to the experience. But in doing so, they created a mini service recovery moment. Your conscious brain switches on: why is there a queue? Why isn’t anyone helping? And then boom — the resolution arrives and you’re paying attention. That is remembered far more vividly than if all six staff had been there when the bus arrived and there was no queue at all. The resolution of the line is what’s memorable. The seamless alternative is pleasant but forgettable.

Mark – 18:30
And keep it manageable, I suppose.

Sam – 18:32
Absolutely. They had the right number of staff. They knew when the shuttle was arriving. They simply waited fifteen extra seconds. Completely manageable, intentional, designed. And I’d add: test it. See whether complaints increase. See whether loyalty behaviours improve at the locations where you try it first. These approaches won’t work in all situations, and I’d rather someone test it and come back to tell me it didn’t work than take my word for it.

Chantelle – 19:17
An example that comes to mind from my own life: if I’m walking past a small, busy restaurant, I’m more likely to go in and wait than to walk into somewhere large and empty, because I’m willing to trade a bit of waiting time for the expectation of a better experience.

Mark – 19:38
There are lots of restaurants in London where you can’t book, and I think that’s exactly why. The wait makes it feel more special.

Sam – 19:45
And that’s effort as a signal of quality. If you had to queue, you’ll rationalise to yourself that it must have been worth it. We tell ourselves stories. We’re not fully objective, not even close. An empty restaurant doesn’t give you a story — it gives you a worry: did I make the right choice? Why is no one else here?

Chantelle – 20:18
Just briefly — for listeners who aren’t familiar with service design as a discipline, could you outline how you think about it?

Sam – 20:31
I was in customer experience for twenty years before starting the service design team at LinkedIn two years ago. At that point, I had to ask myself what distinguished the two. Both involve substantial customer and user research. Both involve mapping end-to-end journeys and considering the broader context of what someone brings to an interaction — their expectations, their prior knowledge of you, what you’ve promised them. Where service design goes further, I think, is in looking at the full ecosystem that delivers the experience. We research LinkedIn employees extensively — the tools, processes, systems, and policies in place, and the degree to which those either enable or obstruct the delivery of a great customer experience. You might trace a policy all the way back to the legal team and ask: why does this exist? Often there’s a legitimate reason — protecting against bad actors, for instance — but that same policy may be preventing a customer-facing employee from answering a customer’s question. Service design takes responsibility for that full sweep, crossing departmental boundaries and accounting for the different roles the same person might play. Most listeners are probably LinkedIn members, but not LinkedIn customers. The same individual embodies different identities and mindsets depending on whether they’re looking for a job or hiring for one, and service design has to hold all of that.

Mark – 22:43
Thinking specifically about LinkedIn — where in your service design work would introducing friction be most beneficial?

Sam – 22:52
LinkedIn Learning is one example. Between videos or sections of a course, we include quizzes and reflection questions, prompting learners to pause. We know from the broader e-learning research that if you don’t stop to reflect and process what you’ve learned, it won’t stick. That friction makes the learning stronger. Another example is our Campaign Manager marketing product. If someone sets a budget we know is too small to reach a meaningful audience for their stated goals, we’ll flag that. Equally, if they’ve cast too wide a net and are likely to overspend reaching the wrong people, we’ll intervene. Those moments of “are you sure?” are good friction in practice. There’s more we could do, but those are two live examples.

Chantelle – 24:27
And on the internal side — are there examples where addressing employee friction has had a significant CX impact?

Sam – 24:38
Absolutely. We spend a great deal of time on the employee experience within my service design team. We’ve been running waves of research over the past two years — shadowing employees, sitting with them, watching them work. Observational research is invaluable because I don’t know what a salesperson’s day actually looks like, so I ask what might seem like naïve questions: why are you clicking there? Why are you going to that tool? And you learn things like: this dashboard has more up-to-date data than the CRM, so I always go there instead. Or: I don’t trust the information in Salesforce because activity tracking isn’t done consistently, so our team maintains an offline spreadsheet. These were real examples from our research. Activity logging was a major friction point. After a customer call, a sales rep would need to take time to clean up their notes and enter them into the CRM so that future account managers, customer success teams, or support staff would have the history of that account. That doesn’t benefit the rep in the moment — it’s purely administrative. So we’ve been working on automating it. We’re recording calls through Microsoft Teams and Gong, using AI to capture and summarise notes, and logging activities directly on the employee’s behalf. The output is higher quality, and it saves the employee time. That’s a case where removing friction benefits almost everyone in the ecosystem.

Chantelle – 26:37
And the employee isn’t resenting a task that feels pointless to them in the short term — particularly if they’re in sales and motivated by immediate outcomes.

Sam – 26:49
Exactly. And there’s a funny irony in those conversations: in almost the same breath, employees will tell us that the customer information in Salesforce is poor, and that they hate logging activities. Well — there’s your explanation. Everyone is in that same bind: no one wants to do the logging, so no one has good information.

Chantelle – 27:16
That crosses into the MarTech space as well. Employees often don’t understand the downstream value of what they’re capturing, which makes them resent the additional task even more. So, Sam, if someone wanted to rethink friction in their own organisation, where should they start?

Sam – 27:50
The first step is a mindset shift. Friction gets cast as always bad. Change that. Acknowledge that there is good friction and bad friction. There is absolutely a great deal of bad friction to remove from the experiences of your customers and your employees — that work is important and ongoing. But don’t assume all friction is bad. Once you have that mindset, start asking: is this good friction or bad friction? Bad friction is almost always unintentional — not designed, not chosen, there by accident or happenstance. That’s what you remove. And as you remove it, ask yourself: do we actually want this to be a fully effortless experience? Paying a bill should be effortless — as few steps as possible, unmemorable. But there are other moments where, once the bad friction is gone, you might ask whether some well-placed good friction would help — so that the customer is consciously engaged at the moments that matter, and is likely to remember your best moments rather than glide past them. The Enterprise pause example illustrates this: it cued the customer’s conscious attention just before the peak moment of the experience — a warm, efficient interaction with a member of staff — making that moment far more likely to be remembered.

Chantelle – 29:52
Are there particular moments within a customer journey where friction tends to add the most value?

Sam – 30:03
Yes. This connects to the peak-end rule, identified by Barbara Fredrickson and Daniel Kahneman around thirty years ago. People overwhelmingly remember two things about an experience: the peak moment — whether positive or negative — and how the experience ended. Recent research suggests these two moments account for around 94% of the memory of an experience. That’s remarkable. Two moments dominate everything. So if you have a peak positive moment in your experience, you can amplify it by creating contrast with the moment just before — and that contrast is often achieved through friction. The Enterprise pause takes the experience from a seven (everything fine, staff available) to a four (there’s a queue, why is no one helping?) and back up to an eight (resolution arrives, friendly staff, efficient service). That trajectory from low to high is far more memorable than a consistent seven or eight. Use friction around your peak positive moments to make them land harder.

Mark – 31:23
I really enjoyed what you said about staff. Empowering your team with the right tools and the latitude to do the best for the customer — that feels like such an important part of this.

Sam – 31:58
I’ll add one thing, if I may. I’ve been looking recently at how much more likely we are to remember a human moment than a digital one. It’s part of our evolution — we are the most social species on Earth, and we got that way because being in large groups was a survival advantage when you lack sharp teeth or great strength. That evolutionary wiring won’t change as quickly as digital screens disaggregate human interaction. A human moment is another form of friction. You might say: this could be faster with self-checkout, but we’re going to have a member of staff serve you instead. Trader Joe’s in the US — consistently the highest-rated grocery chain — has no self-checkout, no delivery, no curbside collection. You interact with a member of staff. Their staff are friendly, they’re engaged, and you leave feeling noticed. That’s more memorable because it was delivered by a human. So my final piece of friction advice is this: there are moments where keeping the human element — even if it slows things down — is the right choice, because it will be remembered more positively.

Mark – 33:18
I’ve just come back from holiday and went through Heathrow, where it was all self-service bag drop. There must have been fifteen members of staff helping people use the machines — and you can’t help thinking that if those fifteen people were simply checking bags in the traditional way, the whole thing would have been faster and more pleasant.

Sam – 33:44
And they haven’t actually reduced their headcount at all, have they? The experience is worse and the staffing levels are the same.

Mark – 33:46
Not at all. Though I have recently been telling people about a very good experience I had with a digital banking app — I had an issue with a transaction, asked the AI chatbot, and it gave me genuinely useful, specific advice. So it can work. But with financial services especially, sometimes you just want to speak to a person. You want the reassurance.

Sam – 34:14
That’s true. And AI interactions are increasingly capable of feeling human — we’re at a point where people genuinely ascribe personalities to large language models and feel a sense of relationship with them. That’s been happening, in a more rudimentary way, since the earliest chatbots fifty years ago. Giving your digital experience a personality and a point of view makes it feel more human, and in the many instances where defaulting to a human-to-human interaction isn’t the right answer, that’s a meaningful alternative. Customers say they prefer self-serve — but they remember human interactions more positively. So the goal should be to make your digital interactions feel as human as possible.

Mark – 35:17
Brilliant. We have one final question, which we ask all of our guests: who has been the greatest professional influence on your career, and what did you learn from them?

Sam – 35:33
A manager I had at Forrester — he was only with me for about a year, though I was at Forrester for sixteen years in total, so he passed through my life relatively quickly. His name is Michael Gazala. I want to name him because he was by far the best manager I’ve had, and I’ve had some very good ones. He focused on making me a better version of myself — genuinely strengths-based development, with a real commitment to coaching. When he took me on, I was a mediocre research analyst. By the end, I was one of the stronger ones at Forrester. What I take from Michael is this: focus on people’s strengths, set them up to use and build on those strengths, and commit consistently to coaching and development.

Mark – 36:21
Good leaders. At TapCXM, Chantelle is leading on making sure our leaders are good leaders — not because they were the best technically in their area, but because they know how to grow and develop people.

Chantelle – 36:39
I always say to my team: I’m not the smartest person in the room — that’s exactly why I have all of you. It’s about empowering them and making their jobs easier, not micromanaging.

Sam – 36:51
And there’s a funny thing about leadership done well — when you’re at your best, it can feel like you’re not doing very much. You’re deliberately not making decisions for people. You’re deliberately not telling them what to do, because you want them to develop the confidence and capability to do it themselves. Sometimes it’s hard to tell: am I being lazy, or am I being a good manager?

Chantelle – 37:12
I’ve started calling myself the queen of delegation.

Mark – 37:18
Brilliant. Thank you so much, Sam. It’s been a real pleasure.

Sam – 37:21
My pleasure — thank you both for having me.

Chantelle – 37:24
Really interesting to hear the thinking behind friction, and to reflect on examples from my own experience where it’s shaped my perception of a brand. Great conversation.

Mark – 37:43
An interesting one for me this week. Sam really challenged the way I think about things. I’ve always been focused on seamless, frictionless experiences — so it was genuinely thought-provoking to hear someone argue that sometimes too easy isn’t actually the best outcome, and that a well-placed pause can make the overall experience better.

Chantelle – 38:10
What struck me most was the idea that not every good experience is a memorable experience. Creating a narrative, building contrast between negative and positive moments — the idea that past experiences guide future decisions is really important.

Mark – 38:30
And the balance is not making it cynical or forced. Just those simple things — are you sure? Is this the right moment? He talked about LinkedIn Learning, Instagram’s scrolling prompt. I was thinking about Netflix asking “are you still watching?” after six episodes. Maybe there’s something in that.

Chantelle – 39:01
All great practical takeaways — including the internal stuff about employee experience and how that flows through to the end customer. Brilliant.

Mark – 39:17
We hope you’ve enjoyed this month’s CX Equation, and we’ll see you again next month.

Chantelle – 39:22
See you next month.

Outro – 39:27
The CX Equation is brought to you by TapCXM. To find out more about what we do and how we can help you, visit tapcxm.com. Search for the CX Equation on Apple Podcasts, Spotify, or wherever you listen to your podcasts, and make sure to subscribe so you don’t miss any future episodes. On behalf of the team here at TapCXM, thank you for listening.