Reducing Friction: Why Fewer Steps Win More Users

Somewhere between “I want this” and “I have it,” people quit. They quit during sign-up. They quit at checkout. They quit halfway through setting up the thing they were genuinely excited to use ten minutes ago.

The reason is almost never the price or the product. It’s the path. Every click, every field, every “choose one of these” is a small toll, and enough small tolls turn an eager user into a closed tab. The good news: most of those tolls are removable, and once you learn to see them, you can’t unsee them.

Why this matters

Every product asks the user to do some work before they get what they came for. They click, they type, they choose, they wait. The fewer required steps before the payoff, the more people finish. It’s that direct.

And the numbers are sobering. The Baymard Institute, a research group that studies online checkout, found in 2024 that the average checkout runs 5.1 steps long with 11.3 form fields, while most retailers truly need only about 8 fields. That’s roughly a 29% cut just sitting there, waiting for someone to make it.

This isn’t only a checkout problem. It’s a sign-up problem, an onboarding problem, a “create your first thing” problem. Anywhere a user has to do work before they reach value, friction is quietly costing you the people you worked hardest to attract.

What friction and flow actually mean

Two plain definitions, because the rest of this builds on them.

Friction is anything that slows, blocks, or frustrates a user on the way to their goal. Extra steps, extra decisions, extra fields, waits, confusion, mental effort: all friction. Think of it as a cost the user pays before they reach value.

Flow is the ordered sequence of steps to finish a task: the sign-up flow, the checkout flow, the “create my first product” flow. A flow is made of screens, taps, decisions, and form fields. The important part is that you can literally count all of these.

One caveat that matters: not all friction is bad. A confirmation step before deleting a customer’s order is good friction. It protects them. The target is unnecessary friction: the steps that exist for no reason the user cares about.

Map the flow and count its cost

You can’t remove friction you can’t see. So write the flow down.

For each screen, list every action (every tap and every thing the user must type), every decision (any point where they choose between options), and every required field. Then ask one question of each item:

Is this necessary right now?

If it isn’t, run it through four options, in order:

1. Remove it. Delete the step entirely.
2. Defer it. Push it to later, after the user has value.
3. Default it. Pre-set a sensible value so the user just confirms.
4. Auto-fill it. Detect or reuse the answer so the user types nothing.

The count is your diagnostic. A flow with 5 screens, 9 decisions, and 14 fields tells you exactly where to dig. Once you can see the toll, you can start removing it.

Two laws worth knowing

Hick’s Law says the time to make a decision grows as you add more choices. More options on the screen means slower, more anxious users. The lesson: cut or group the choices on the critical path.

Tesler’s Law (the Law of Conservation of Complexity, named after Larry Tesler at Xerox PARC) says every system has an irreducible amount of complexity. You can’t delete it. You can only move it. When you make something simpler for the user, the complexity shifts onto the system and the people who build it.

Picture a great restaurant. The menu is short and the dishes sound simple. The complexity didn’t vanish; it moved into the kitchen, where the chef absorbs it. Good software works the same way. The team should swallow the hard parts so the user’s plate stays simple.

Progressive disclosure: show only what’s needed now

Progressive disclosure is an interaction pattern named by Jakob Nielsen in 1995. The idea is simple: show only the primary, essential options by default, and tuck the advanced or rarely used ones behind a “Show more” link, an accordion, or an “Advanced settings” section the user opens only if they need it.

It lowers mental load, makes products faster to learn, and prevents errors. Fewer visible options means fewer wrong choices. Google’s homepage is the classic case: one search box up front, advanced search hidden away for the few who want it.

Here’s it in practice. On a product form, ask only Name, Price, and Photo up front. Put SKU, inventory tracking, SEO description, and variant options behind an “Add more details” expander. About 90% of shop owners will skip it on day one, and that is exactly the point.

One distinction to keep straight: progressive disclosure is not a wizard. Progressive disclosure splits options into primary versus secondary on the same screen. A wizard splits one task across several sequential screens. More on wizards below.

Every field has a cost

Each form field is another small tax on completion, and the drop-off adds up. The fixes are concrete and reusable:

• Combine “First name” and “Last name” into one “Full name” field.
• Hide optional fields (address line 2, company, coupon code) behind a small link.
• Default the billing address to the shipping address.
• Delay or skip account creation entirely.

Stripe, the payment company, is the gold standard. Its card form is a single combined field: number, expiry, and security code in one row. It detects the card brand from the number itself (no “card type” dropdown), formats the digits as you type, and validates inline. Stripe absorbs all the complexity, validation, brand detection, and security rules, so you type the bare minimum. That’s Tesler’s Law working for you.

Guest checkout and the forced-account trap

The single most damaging field of all is the one that forces strangers to create an account.

Baymard found that 24% of US shoppers abandoned a cart in the past quarter purely because the site forced them to register first. Another ~18% leave because checkout feels “too long or too complicated.” Yet 62% of sites still fail to make guest checkout easy to find.

The fix is an order-of-operations change, not a feature: offer guest checkout as the main path. Invite the user to save their details after the purchase is confirmed. Once they trust you, “save my card for next time” feels like a gift, not a gate.

One-click checkout: deferring friction, not deleting it

Amazon’s “1-Click” (patented in 1997, expired in 2017, since copied by Shop Pay, Bolt, and others) collapses an entire multi-step checkout into a single tap. How? It stored the card and address once and reuses them forever.

The friction wasn’t deleted. It was deferred and collected a single time. That’s the whole trick, and it’s widely credited with hundreds of millions in extra revenue. (The exact uplift percentages floating around come from secondary blogs, so treat them as illustrative, not gospel.)

Common misconceptions

A few myths worth clearing up:

• “Simpler means dumber.” No. Tesler’s Law says the complexity doesn’t disappear; it moves into the system. Simple for the user is often harder for the team to build, not easier.
• “Fewer steps always wins.” Mostly, but not blindly. A short 3-step wizard beats a 20-field wall of inputs. The goal is less load per screen, not fewer screens at any cost.
• “All friction is the enemy.” Good friction, a publish confirmation that states consequences, a check that blocks a $0 un-orderable product, protects the user. Keep it.
• “Progressive disclosure is just a wizard.” It isn’t. Disclosure splits primary from secondary on one screen. A wizard splits a task across many screens.

Wizards versus a single page

When a flow does need multiple steps, choose the shape deliberately.

Pattern	Best for	Watch out for
Single page	Short tasks; when users want to scan everything at once	Looks overwhelming if the task is long
Wizard (multi-step)	Long, complex, or branching tasks; reduces load per screen	More clicks; can hide total length; drop-off between steps

The rule of thumb: short flows fit on one page; long or branching flows deserve a wizard with a clear progress bar and a back button. A 3-field single screen beats a 6-step wizard, but a short 3-step wizard beats a 20-field wall.

Onboarding: time-to-value and the aha moment

Time-to-value (TTV) is how long it takes from first use until the user gets something real. The aha moment is the instant they finally get it: “oh, this is what it does for me.” The whole job of onboarding is to shrink the gap between the two.

Teams pick an activation metric, a measurable action that predicts long-term use. Facebook’s growth team famously used “7 friends in 10 days.” Dropbox aimed for one file in one folder on one device. (Treat the exact thresholds as well-known industry lore, not precise science. The principle is what’s solid.)

There’s a famous story behind this from Clayton Christensen. A fast-food chain couldn’t sell more milkshakes by tweaking flavor, until researchers watched buyers and found bored morning commuters “hiring” the shake to make a long, dull drive less boring. Design for the job the user came to do, then strip everything that doesn’t serve it.

How to use this

A practical checklist you can run on any flow this week:

1. Map and count. Write down every screen, action, decision, and field. Get a real number.
2. Interrogate each item. Ask “is this necessary right now?” If not, remove, defer, default, or auto-fill, in that order.
3. Disclose progressively. Show only the essentials up front; hide advanced options behind “Add more details.”
4. Cut field count. Merge fields, hide the optional ones, and pre-fill anything you already know.
5. Offer guest checkout. Make it the main path. Ask people to save details after they get value, not before.
6. Collect once, reuse forever. Store the card and address a single time so the next purchase is one tap.
7. Aim onboarding at the aha moment. Pick one activation metric and remove every step that doesn’t lead toward it.
8. Keep the good friction. Confirmations and validation that prevent costly mistakes stay.

Run a quick test of your own product: count the steps from “I want this” to “I have it.” If you can’t say the number out loud, you’ve found your first thing to fix.

Conclusion

Here’s the one thing to carry with you: complexity never disappears, it only moves. Your job is to keep moving it off the user’s plate and onto the system, one defaulted field and one deferred step at a time. The product that asks the least of its users, while quietly doing the most for them, is the one people actually finish.

Reducing friction gets people to value faster. The next question is what keeps them coming back once they’re there, and that’s a different craft entirely: the psychology of habit, triggers, and the small rewards that turn a first visit into a daily one.