How to build a frictionless customer experience (and how AI is changing things)

Having a frictionless customer experience is essential for the success of a SaaS business. That’s because in this era of tech, customers expect things to work fast and to work 100% of the time. Our CEO, James, actually just wrote about this in our newsletter:

That’s a great goal to aim for. Even if you don’t achieve perfect uptime or perfection, you should work to become as close as possible, because you want your user journey to be free of as many obstacles and hindrances as possible.

Why do we put so much importance on a frictionless experience?

It’s not just about pleasing the customer in the moment.

The less friction there is, the easier it is for the user to achieve their goals, which leads to better retention, less churn, increased overall loyalty.

That impacts your retention metrics directly, but also can improve your NPS score, your CSAT score, your CES score, and other relevant metrics.

And as these metrics improve, so does your lifetime value and your overall ability to upsell and expand to customers.

This is obvious.

But how is AI changing these dynamics? If you don’t bring these functionalities into your product, you risk getting left behind, but it’s also easy to get caught up in the hype.

Let's jump into this concept of frictionless experience!

What exactly is friction in product?

Before we can talk about creating frictionless experiences, we need to talk about what exactly friction is.

When we think about customer friction points, I'm referring to obstacles or issues that stop, slow, or frustrate the user as they attempt to complete a task.

You can find these in all stages of the user journey, from onboarding to support to task execution and more.

One example might be an onboarding process that is not well designed and has too many fields required. Users find themselves having to put in information not relevant to their persona.

Or, you can imagine a customer getting frustrated because they are unable to get in touch with the support agent or the help they need.

Points of friction might seem minor. But they’re usually indicative of a broader problem in your product design, and if you let them fester, they will boil over into massive churn and headache.

But we also have to talk about the reality that friction doesn’t just come from product design, but also from your own internal team, values, and division of roles and responsibilities.

Even if you have a very slick interface for your support chat, if you don’t have enough agents, or if they aren’t well-trained enough, it doesn’t mean anything.

Creating frictionless experiences demands a holistic effort of great UX and a well-organized team to put the user first.

But how do you identify customer friction within your products?

How to find friction

First of all, you need to focus on the user’s needs — not your wants.

You have to obsess over the collection of user data, feedback, and engagement metrics.

Important questions to ask:

• How is your software used? According to your users.
• What is the customer effort over the last six months?
• What are you hearing from user interviews, and within survey and feedback responses?
• What are your analytics and engagement funnels showing you?
• What does your testing seem to indicate? Does that line up with what you’re seeing in real life?

Once you identify these core concepts, you can begin to analyze your funnel and user engagement data to identify friction points.

Creating frictionless user experiences

Creating a frictionless user experience requires a comprehensive and multi-pronged approach that incorporates every stage of the customer journey, from first touch to retention or churn.

But that doesn’t mean that each stage is equal, even if they’re all important. The reality is that different parts of the journey are more important than others when it comes to reducing friction.

For example, without frictionless user onboarding, you can't even get users into your product! Thus, it makes sense to start with optimizing your onboarding process before you worry about how to upsell users.

Now, I'm not saying that you don't need to optimize every area of your product, but the reality for many startups is that engineering and design resources can be quite limited, and thus it might make sense to focus your energy first on onboarding before you devote more time to secondary or tertiary items.

What are some of the best practices to create frictionless user onboarding?

Lead with clarity and intuition. When you have well-designed workflows that are easy to understand and follow, and which correlate to industry and user expectations, your making it easier for the user to take the happy path.

Also, ensure that users have easy access to support either in-app with nudges or tooltips, or with a quick line to an agent or an AI assistant.

Your general design approach really matters as well. When you build, you shouldn't just build pretty things for beauty's sake. You should design with the user in mind, with a focus on reducing the effort it takes to complete tasks within your product. Think of what is logical and clean while also being aesthetically pleasing.

The good news is that there are now a variety of no-code solutions, such as Wix and Squarespace, that allow for the creation of aesthetic and intuitive interfaces without requiring advanced design or coding skills. These tools offer drag-and-drop functionality, enabling those without a technical background to build user-friendly online experiences that align with UX design principles.

I won't go into all of the general UX best practices that you might want to adhere to, but some of the most important ones are ensuring consistency in design structure and taxonomy across your site, as well as clear next steps for users within each workflow.

How AI is changing things

One of the main ways to battle frustration and friction is with intuitive automation.

What I mean by this is that it's not all about having fancy and super complicated AI products that can take over the entire product right now.

Rather, what can be helpful is intuitively adding AI power for core repetitive tasks that users do regularly.

For example, if you're an HR logistics platform, you know that your users have to run payroll twice a month.

And so you might examine where in that user flow you are identifying friction from your user surveys and feedback and funnel analysis.

Then, you might think about how you can bring AI in to automate some of those repetitive practices.

Now, that's just one example.

But the mental model that you should adopt here is to not think about how AI can transform your entire product in one broad stroke.

What's more realistic is to think about where it can be best leveraged with the least intrusion. Just adding an AI agent in the bottom right and shoving it down your users' throats is not going to create a frictionless experience. It has to be intentional, mindful, and generally in line with your design principles and the desires of your users.

That's why it's so important to have great user data and an understanding of your personas.

For example, if you're in construction tech, as I once was, you're likely aware of the fact that not all of your users are as tech-savvy as those that you might find in a B2B SaaS product for enterprise cloud.

And naturally, that dictates what you can and cannot do with AI-powered tooling. It's not to be derogative and to say that those folks don't understand what's going on behind the scenes. In fact, they could be the ones who could most benefit from AI.

Rather, what I mean is that you want to keep things simple and within the realm of your product. Don't overreach – don't do AI for AI's sake.

The transition

One of the things we're most excited about here at Command Bar is how in-app messaging can be transformed from highly intrusive and un-targeted towards very user-friendly, user-focused, and un-intrusive. What I mean by that is that the pop-ups of yore, which were big interstitials, or even today, with many being hard to close out or otherwise purposefully misguiding, are on their way out.

They're no longer going to be tolerated by users, and in fact, they're bad for companies as well because users are so conditioned to just quickly click out of them.

The reality is that putting the user first doesn't mean sacrificing your team's needs. In fact, when you create product tours, nudges, and pop-ups, what you're actually doing is improving the state of play for everyone involved, because not only are your users getting more targeted and timely information and prompts, but you're also better able to segment your messaging, leading to higher conversion rates, higher click-through rates, and a more satisfied user base.

What this does in the long run is create a sense of trust and communication between the product and the user.

They won't click out of those pop-ups as quickly as they've been conditioned to because they remember that the last time they saw one, it was actually a relevant invitation to a beta testing pool or a discount on an upsell that they actually wanted. That's the power of a user assistance mindset, which puts them at the forefront and not as an afterthought.

We've really extended this into our Copilot product as well. We've been so excited seeing our customers get improved engagement from their end users. But that's not the only reason that Copilot has thrived as an AI agent and as an augmentation to support strategies. It's also because it's been able to deflect tickets and help companies pinpoint where they need to improve their help documentation through analysis of queries and their fallback and dead ends. That's user intent data that you can't find elsewhere!

The future of frictionless customer experience

The future of frictionless customer experiences means not only continual optimization of user flows and user-first product design, but also a reliance on AI to reduce friction and improve users' ease of use.

I think that the priority on model speed right now, as we see every single large tech company invest billions into improving their large language models, only goes so far.

What actually has to be prioritized is the end user experience, right? It's about how AI actually powers better experiences for users. And in the narrow model of a ChatGPT or a direct conversational AI tool, certainly pure speed is a benefit and a competitive advantage. But within your product, the way that AI is really going to reduce friction and create a better customer experience is by solving the right problem at the right time.

And that requires a lot more than just a fast model. It means incorporating a clear vision for these AI products into the fundamental design of user flows, and backing them up with all of the user research and feedback data that you would use to inform any other area of your product design.

I think it's been easy over the past two years to let product strategy get hijacked by AI.

But the companies that will create truly frictionless experiences and be category-defining are the ones that will find ways to leverage AI in ways that truly solves problems that users care about.