By Peter Ramsey
19 Aug 21
Selling an NFT on OpenSea
NFTs (non-fungible tokens) are unique digital assets—like an image, or a song—that are stored on a blockchain.
And like traditional art, some NFTs are worthless, while others are hyper-valuable. In February 2021, an NFT from the artist Beeple sold for more than $69 million.
Every day, art that you probably wouldn't even hang on your wall is selling in a JPEG format for thousands of dollars. It's a total NFT-feeding-frenzy.
So people have flocked to OpenSea: the biggest platform for buying and selling NFTs. But is the user experience of actually creating and selling one any good?
To test that, I've created my own collection, and the short answer is: it could be much better.
Summary:
🚗
1. Why you need to use 'Actionable routes'
🗂
2. The trap of sub-tasks
✅
3. Don't forget to auto-save
3 key takeaways
1. Actionable routes
You may have heard of the concept under another name, but an actionable route is essentially a preferable next step at any given time.
For example, after finishing an episode of something on Netflix, one actionable route would be to watch the next episode, another may be to find a different series to watch.
As a designer, you only really care about the actionable routes that are common and commonly preferable. i.e., the user may want to rewatch the same episode at 2x speed, but it's unlikely enough that you wouldn't prioritise it.
It's evident that OpenSea haven't spent much time thinking about these. As an example, look at this:
At this point, the user has clicked into the 'collections' field, whilst adding a new NFT. I imagine that these would be the likely actionable routes:
🌱
Add a new collection
🧠
Learn more about collections
🤷♂️
Do nothing
And so the UX for that empty state really writes itself: you help the users get to their actionable routes as efficiently as possible.
2. Sub-tasks
Studies have shown that multi-tasking increases stress and decreases productivity. Despite the evidence, people still think that they can do it effectively.
One possible reason for this may be the issue of defining 'multi-tasking'—occasionally called 🗯 Context Shifting.
To clarify, because you only have one cursor on a computer it's difficult to literally do two tasks at once. But, the term 'task-switching' and 'multi-tasking' are interchangeable here.
Or rather: the process of switching between tasks carries a penalty. It's mental multi-tasking.
As an example:
⛳️
1. Creating an NFT
The user has a clear task: to create an NFT.
☝️
2. Creates a collection
The user has to switch tasks, to creating a collection.
⛳️
3. Back to creating an NFT
They need to re-orient their focus back to this.
My point isn't that you should always refrain from any type of task-switching—it's often unavoidable.
But rather, that bringing the user back into the original task is more challenging, and what OpenSea have done here is actually make it harder for the user.
Here's the nuance: it's easier to engage with a new sub-task, than it is to refocus on your original task.
OpenSea should be bringing the user immediately back to their original task, with a context buffer like this:
This would confirm what has just happened, and provide transitionary context—i.e., that sub-task is now complete, and you can get back to creating your NFT.
OpenSea would probably find that this approach makes the process of creating an NFT and a collection feel more intuitive.
3. Auto-saving
A major factor in managing the sub-task transition is to ensure that the user hasn't lost any work. You never want them to have to re-enter information.
This can often be solved by auto-saving.
There's another common scenario where auto-saving is immensely valuable: encountering flow-breaking errors.
For example, during the collection-creation process I kept running into this bug where clicking the main CTA would give the impression that it's doing something, and then just spring back into it's default state.
Click.
Wait.
Click again?
You can't avoid every error. It's not even a realistic expectation to build helpful error messages for every potential error. Sometimes things just break.
But, auto-saving can be used like a UX-airbag.
And it would have prevented a lot of frustration here, because I had to refresh the page on OpenSea, and enter everything again.
