For us, this is where Web 3.5 comes in. What we’re going to build are personalized websites. We’re going to explicitly incorporate personalization into the website itself.
The best way to think about this is customization. We’re tailoring the site to the user, kind of like a suit. We start with a basic suit that’s the same for everyone, but then we alter it to fit the individual. The foundation doesn’t change, but how it fits and how it’s experienced does.
That’s the idea behind Web 3.5 personalization.
For a simple keyword tool example, I have a very stereotyped idea of different kinds of users, and then the programming tries to determine what kind of person is sitting in front of the device.
The first time someone shows up, we treat them as a naive user. Sometimes we talk about new users versus returning users—or even worse, you’ll see something like Amazon saying, “Hey, you’ve been a member since 1998.” And I’m like, okay, that’s great.
Again, we’ll pick on my Aunt, who started using Amazon maybe two or three years ago. She gets the exact same interface that I do. Why? I’ve used a ton of features, and someone who just started yesterday gets the same experience.
Now, a lot of lazy sites solve this by saying, “Well, just tell me who you are, and then I’ll figure out what you care about.” But we don’t actually have to do that. We can figure this out. We can do a lot of the work on the backend to determine what kind of user you are.
To make it simpler, we can stereotype just a handful of users.
As a naive user, we’re going to incorporate a lot of patterns like guided interaction. I want you to read the directions. If I just throw everything at you at once, you’re going to be overwhelmed. So instead, it says: press Start to begin. Then I give you a walkthrough.
Select default data so the wizard can fill in some sample data. Now press the Process Text button and see your results to the right. So I process it, and here are your concepts, sorted by frequency of appearance. Refresh the page to try your own data.
Now, when I refresh the page, I’m no longer a naive user—I’m a novice user. And by the way, these are two different kinds of new users: one who has absolutely no idea what your product even does, and one who is just unfamiliar but has now seen it once.
At this point, I’ve done the walkthrough. I know what the tool can do. I can delete the example text, put in my own text, and process it.
Another thing that happens when I refresh is that we still provide help—but it’s contextual help, tailored to the type of user I am now. As a novice user, when I put my mouse in the text box, I get additional guidance automatically. I don’t have to ask for it.
But after I’ve done this enough times, I become a typical user. At that point, the automatic help goes away. Instead, I’m allowed to ask for help—but I’m not constantly being shown it.
I also start getting additional, secondary information. Maybe some input hints. Maybe extra context. And eventually, I can even become an advanced user. Now I get access to more features. I might see settings where I can do more advanced things.
What we’re doing here is using these user types to guide the experience. By figuring out who the user is, we can adapt and customize the entire product around them. And that, right there, is the core idea behind Web 3.5 personalization.