Not all friction is bad. The question is whether it's earning its place.
Every instinct in product design tells you the same thing: remove friction. Make it faster. Make it easier. Reduce the number of clicks.
And for most of onboarding, that instinct is right. The Straight Line from Chapter 5 is all about cutting unnecessary steps.
But here's the counterintuitive truth: some friction is the reason users stay.
The trick isn't eliminating all friction. It's understanding which friction to kill and which friction to keep.
The Hierarchy of B2B User Friction
Not all friction is created equal. In B2B, friction falls into three layers, and each one requires a completely different response.
Functional friction: the product is in the way.
This is the most visible type. Confusing navigation. Slow load times. Forms that ask for information the product already has. Error messages that don't explain what went wrong. Buttons that don't look clickable. Features buried three menus deep.
Functional friction is almost always bad. It's the product failing to do its job. Fix it.
Social friction: the organization is in the way.
This is the B2B-specific layer that consumer products rarely deal with. A user wants to adopt your product, but their manager hasn't approved it. They need a teammate to upload data before they can do anything. The IT team hasn't granted the right permissions. Their department is using a competing tool and switching feels politically risky.
Social friction lives outside your product. You can't fix it with better UX. You fix it with enablement materials, champion toolkits, and making it easy for your users to sell the product internally.
Emotional friction: the user's own psychology is in the way.
This is the deepest and least visible layer. Fear of looking incompetent while learning something new. Anxiety about making a mistake with real data. Imposter syndrome about whether they're "technical enough" for the product. Worry that they'll invest time learning this and it won't work out.
Emotional friction is invisible in your analytics. You won't see it in heatmaps or funnel data. But it's often the real reason users hesitate, stall, or quietly disappear.
Think of these three layers like a pyramid. Functional friction sits at the base: it's the most visible and the easiest to fix. Social friction is in the middle: harder to see, requires a different toolkit. Emotional friction sits at the top: the hardest to detect and the most personal.
Most teams only work on the bottom layer. The teams that win at onboarding work on all three.
When friction is the feature
Now here's where it gets interesting.
Superhuman, the email client, requires every new user to complete a mandatory onboarding call before they can use the product. That's friction by any definition. It's a scheduling step, a 30-minute time commitment, and a human interaction before the user can even send their first email.
But that friction is the reason Superhuman users are so engaged. The call teaches them keyboard shortcuts, sets up their workflow, and gets them to competency faster than they'd get there on their own. By the time the call ends, the user isn't just set up; they're confident. They've had a human demonstrate the product specifically for their use case.
The friction created value that self-serve onboarding couldn't.
This isn't unique to Superhuman. Every product has moments where asking the user to invest effort actually increases the value of the experience. The key is that the effort has to feel earned by the user, not imposed by the product.
The IKEA Effect
There's a well-documented phenomenon in behavioral psychology called the IKEA Effect: people place disproportionately high value on things they helped create, even if the result is objectively mediocre.
This is why users who customize their dashboard are more retained than users who use the default. It's why users who set up their own workflow templates are more engaged than users who import pre-built ones. It's why users who manually configure their first project feel more ownership than users who have it auto-generated.
The effort itself creates attachment.
This doesn't mean you should make onboarding harder for the sake of it. It means you should look for moments where user effort creates genuine investment. Customizing a workspace. Naming a project. Choosing how notifications work. These small acts of creation build psychological ownership.
The line is thin but clear: effort that leads to ownership is good friction. Effort that leads to frustration is bad friction.
The Sunk Cost Effect
Related to the IKEA Effect but distinct: the Sunk Cost Effect means users who have invested time are more motivated to continue, even when the rational thing might be to stop.
This works in your favor during onboarding. Once a user has spent 15 minutes setting up their first project, they're more likely to come back than a user who spent 2 minutes clicking through a product tour. Not because the 15-minute experience was better (it might have been worse), but because they've invested and don't want that investment to be wasted.
This is not a license to waste users' time. It's an observation about how human psychology works. If you design your onboarding so that the effort users invest is meaningful and builds toward their Same-Day Win, the sunk cost effect works naturally in your favor. If you design onboarding that wastes their time on meaningless steps, the sunk cost effect won't save you.
Cognitive load vs. cognitive overload
There's a line between engagement and overwhelm, and it's thinner than most teams realize.
Cognitive load is the mental effort required to process information. Some cognitive load is necessary and even beneficial. Learning a new skill requires mental effort. That effort is what makes the learning stick.
Cognitive overload is when the mental effort exceeds what the user can process. This is where you lose them. Not to frustration with a specific feature, but to a general feeling of "this is too much."
The difference between load and overload often comes down to pacing and context. Asking a user to learn one new concept per screen is cognitive load. Asking them to learn five new concepts on one screen is cognitive overload. Introducing a feature at the moment the user needs it is cognitive load. Introducing every feature during the welcome tour is cognitive overload.
The Straight Line from Chapter 5 manages cognitive load by design. By limiting each step to only what's necessary, you keep users in the productive zone of engaged effort without tipping into overwhelm.
Balancing friction with time-to-value
Here's the tension you need to hold:
Fast time-to-value is critical. Users need their Same-Day Win quickly. But true mastery takes time, and that's okay.
The goal isn't to eliminate all effort. The goal is to ensure that early effort leads to early value, and that later effort leads to deeper value.
First session: minimal friction, fast Same-Day Win.
First week: moderate friction, deeper capabilities unlocked.
First month: productive friction, workflow optimization, team collaboration.
Each phase of the onboarding loop can tolerate more friction because the user has more confidence, more investment, and more evidence that the product works. The friction they encountered on day one would have been a deal-breaker. The same friction on day 30 is just part of getting better at something they already value.
Putting it together
The best onboarding doesn't remove all friction. It removes the wrong friction and keeps the right kind. Knowing the difference is what separates onboarding that users tolerate from onboarding that users appreciate.
Next: Chapter 7, The Journey After the First Win
If you're not sure which friction to keep and which to kill, let's look at it together.
I offer a free 30-minute Discovery call where I look at your onboarding with you and tell you honestly what's broken and whether we're the right fit to fix it.