There are so many new elements that separate web3 from web2, such as gas fees, tokens, wallets, and smart contracts, that we need to think way beyond just the UI. There are multiple levels.
We need to improve all of those levels to create a good user experience.
How to visualise the different layers of Web3 UX
The User Journey is a Long and Winding Road
Consider a standard app. Nothing Web3, but just a popular app on your smartphone that you use regularly. The actual UI of this app is just the last element in a long string of experiences beginning in the real world, crashing through numerous physical spaces, into digital interactions, through a whole mess of different hardware and software until you eventually find yourself hovering your thumb over a button someone designed.
You probably had a thought, like “play music” or “check directions” or “buy clothes,” which was prompted by something in your physical surroundings.
The “UX” of that experience begins long before you take the phone out of your pocket.
This is under-appreciated when looking at Web3, so let’s take an example.
International payments
There is the “UX” of sending any kind of money to a resident of another country, and there is the “UX” of the particular app you use to do it.
You could argue that sending stablecoins on a blockchain over international borders is already a much better experience than sending FIAT. It’s pretty much instant, and the fees are minimal. If you try to send FIAT it could take a couple of days and you’ll be charged fees at every step. On the other hand, some money transfer apps are pretty good these days. It’s a more familiar and accessible process than the crypto equivalent. Most people have a smartphone and some local currency. Most people don’t have an exchange account, a wallet, and some crypto.
Parts of the user experience are better than the current standard. But a lot of the other steps are worse than the current standard.
By my reckoning, crypto loses out because it’s not as accessible and doesn’t integrate so well with the rest of “the system”.
This is also why there is such opportunity here.
If we really want to understand web3 UX, we need to look at the big picture, and see which levels need attention.
Holistic User Experience Models
The Nielsen-Norman model has three levels:
- Interaction level
- Journey level
- Relationship level
The three together cover the entire user experience from real world to digital. This framework is often used in CX Design (customer experience) and Service Design. In most cases, a person has made many decisions before they even get to your app or website.
If we take a successful company and zoom out even further, we see there is usually a core problem that the company is solving. This is often an annoying issue in real life, and the product has innovated some way of improving this deep frustration.
Annoying_shit_irl --> thing/app_that_might_help --> the_actual_UI
Pioneers have made the biggest impact on people’s lives by dealing with that first bit. It’s just much harder to do, and usually requires major shifts in technology, policy or society.
The classic example here is the apocryphal Henry Ford quote:
What people actually wanted was “to get places faster”. This could be solved by a car. But that solution could only really take off after increasing awareness, plus creating an ecosystem of better roads, overhauling the entire manufacturing industry, and passing appropriate laws.
Seeing the link here…?
The final design of the car was obviously important too, but for this new mode of transport to become as successful as it eventually did, everyone needed to be shown the benefits, and a whole lot of infrastructure, processes, regulations and adjacent technologies had to be improved on too.
If we look at something more modern like Netflix, we can again see multiple levels to the experience. The core user story is something like “people want high-quality entertainment”. The solution to that problem was initially DVDs, but now the solution is a digital streaming platform. Which requires expert commissioning and curating. Users also want to see HD video, which requires good compression on the Netflix side and good internet connection on the user side. And they want to find the stuff they like, which requires a good UI, excellent information architecture, and clever recommendation algorithms.
In both cases the “User Experience” encompasses much more than you might think.
The Four Levels of Web3 User Experience
In the case of web3, I suggest there are four different layers. Each pose specific UX challenges.
A simple framework for visualising the different layers to work on
I see these layers stacking on top of each other. The bottom level is the most important, but also the hardest to influence, because it is most related to technical limitations of whatever blockchain is being used. The top layer is the visual layer, which can be modified fairly easily — though of course what you put there depends entirely on the layers below it.
Below are some of the specific challenges that exist within each layer, starting from the bottom up.
Technology Layer
- Speed
- Fees
- Stability
- Anonymity
Access Layer
- Wallets
- Connecting
- Seed phrases and security
- FIAT onramps
- Interoperability
- Web extensions
- Mobile experience
Functional Layer
- Making transactions
- Liquidity
- Yield farming
- Smart contract interactions and permissions
- User identity
- Governance
- Token types
- Web3 sign-on
- Displaying NFTs
- Ownership of in-game assets
- Sending things cross-layer or cross-chain
Visual Layer
- Visual design
- Information hierarchy
- Navigation
- Content
- Help and instructions
Break it Down
1. Technology Layer
💡 When you look at the actual technology layer, I would argue the UX is already better on DeFi. Sending large payments internationally usually requires a waiting time, fees, and in some cases, security checks with intermediary banks. It’s also hard to track. On the blockchain it sends quickly, cheaply (tho not always 😟), and can be verified by anyone.
The UX challenges at this level are:
- increasing speed to rival or exceed the VISA network (tps)
- improving the stability and security of blockchains
- reducing fees to less than Western Union or Wise
- making transactions verifiable (or alternatively, secret)
These are some of the biggest challenges to be solved. Yet they have nothing to do with an interface. In this light, layer 2 blockchains are direct attempts at improving overall UX.
This sort of thing becomes even more important when you consider non-defi stuff like onchain identity. If one day we are going to be using our self-custody wallets to log in to services and keep our data out of the hands of web2 companies, then we really, really, need fast and secure networks.
2. Access Layer
💡 Imagine the year is 1995 and you’re hearing about this wonderful new thing called the internet. First of all, you’re going to need a PC with a modem and you’ll need to connect it to the phone line. Once you “dial up” to “surf the web”, you won’t be able to answer calls. To actually do stuff, you’ll then have to set up an “email address”, which might be “pop3” or “imap”, use a “browser” and, if you want to be a pro, learn about “telnet” and “ftp”. Somehow you’ll manage it anyway and things will never be the same again.
The UX challenges at this level are:
- making a wallet that is as easy to use as a banking app and/or browser
- creating a web3 single sign-on
- frictionless FIAT onramps
- unifying chains or creating a “layer x”, so manual bridging is not required
- human-readable addresses (ENS is a step in this direction)
- educating people about self-custody and privacy
- putting crypto into the hands of many
- stepping beyond the current internet browser paradigm
3. Functional Layer
💡 What we talk about when we talk about web3 UX. One thing to consider is that some of this stuff is necessarily strange because it’s also brand new. New functions will always seem hard to begin with. A few years ago, nobody had ever heard of NFTs or AMMs. Now we’ve got them, we need to figure out how best to use them.
The UX challenges at this level are:
- making transactions intuitive
- abstracting away complicated strategies (easy yield farming and auto-compounding)
- simplifying actions (e.g. using zappers, wrapping tokens automatically)
- creating flexibility and forgiveness in the system (sending tokens to the wrong network)
- making it easy and beneficial to participate in governance
- making NFTs meaningful and useful
- creating more “killer apps” outside of finance
- moving ownership of in-game assets to the user, not the manufacturer
- increasing the ways NFTs can be shared and displayed
- improving cross-chain and multilayer compatibility (most users probably don’t want ten different forms of USDC)
An important note at this level is that a lot of these problems have been partially solved by some of the custodial crypto apps. The downside is that in most cases, you are trading some decentralisation for better UX. Crypto.com allows you to send tokens for free to other crypto.com users. Most exchanges will let you choose which network to withdraw on (mainnet or Polygon etc), Binance will farm your tokens for you and give you some interest, but you won’t have control over their strategy. Celsius did the same thing and it didn’t work out so well.
4. Visual Layer
💡 What you see and try to understand. Are we using consistent design patterns, based on user research? Is the terminology too technical? Does everything look ok on mobile?
The UX challenges at this level are:
- following accessibility best practice
- reducing the use of technical terms
- making the important stuff stick out
- hiding away the irrelevant stuff
- creating easy to follow instructions
- driving inclusivity
- signalling “friendly” and “welcoming”, not “scary” and “exclusive”
- being A E S T H E T I C
- retaining personality, while building for the many
This represents my first serious attempt to create a larger framework for Web3 UX, and I intend to build on it over time.
I hope that it encourages the discussions our industry badly needs to have about UX. So if you think this is useful, do please give a clap for visibility.
It may not be perfect, but I prefer to ship early and ship often. :) I can always test and improve.
Just remember,
“All models are wrong, but some are useful”.
DALL·E 2022–09–18 19.08.00 — an astronaut sitting in front of a pyramid in space, vaporwave
About Me
I am a product designer and UX consultant specialising in web3. I enjoy scaling and leading design teams, and helping web3 companies at all stages of their development.
If you’re after some more insights, you might want to check my competitive analysis on DEXes, my case study for a project I did with Element Finance, or my series on DeFi Design Tips.
A weekly resourceful newsletter that helps designers stay in the know, be productive, and think more critically about their work. Take a look.