Video: 5 Questions to Ask in UX Personalization
Learn more about OTT and UX personalization at Streaming Media's next event.
Watch the complete video of this presentation from Streaming Media West, OTT203. Myths of OTT Engagement, in the Streaming Media Conference Video Portal.
Read the complete transcript of this clip:
Ronato Bonomini: What are the big questions that we try to understand in this case? Does changing the UI changes how people actually consume data? Is the same content performing better on a different device, on a different UI? That's one of the questions. Can we, for example, use machine learning--which everyone now is trying to do--machine learning for everything? Can we actually use it for something useful, such as understanding which users might like a personalized experience and which users might not like it? Or, for example, understanding the order of carousels, is that relevant for the effectiveness of content?
So, we listed out these five questions. Let's look at the first one. As a goal for this study, we said, “Let’s understand what everyone is doing, what everyone thinks is actually what's happening, out of real data that we collect on an existing live system.”
The first one is an easy question, that everyone probably has an answer for: Is user behavior different on different devices? I’d like you to take a second to think about what you would answer on this one. We'll go through the questions and then we'll see details and the data behind each question.
The second one is, does time of day matter? And before you think of the answer, we're talking about OTT apps. Content is always available at any point in time. Time does not matter. Content is always there; it's not like it's 9pm Thursday, and you know you're going to watch Shonda Rhimes' content on ABC because it's Thursday night, 9 pm. That's what my wife makes me watch every Thursday night.
But this is not time-bound. It's an OTT app. All content is always available, so does time or day really matter? Think about it. Our position and content consumption really did.
As you can see here, for example, in a traditional grid-like UI--which is a simplified view of how UI is actually made--it works.
Positioning horizontal vertical, has a factor. So, what is that factor? We took a look at analyzing the data, what comes out of the data, in terms of consumption.
The next question is recommendation. We always want to make sure that there's some recommended content. There's some machine learning acting behind the scenes. And whatever happens, every time we work with our partners, we have the team that enters that when we enter the room, they look at us, “Ugh, here comes who's going to steal our job?”
That's not really what we want to do. What we really want to do is to make sure that we use the right techniques, and we use the editorial ingenuity to actually make the content shine, and make the content available to everyone.
One of the questions was, how does a user perceive personalization? Should we truly push so much to use the best algorithm in the world? Or what's recent is just going to do the cut. Or what's popular is just going to be enough.
We tested a lot of those, and came out with some interesting information there.
Last question: The hero banner. How many of you are using a hero banner? And how many of you are thinking, “Oh, we have to have a hero banner. Unless we have a hero banner, we're not going to make it.”
So we tested the assumption that the hero banner is the best way to showcase content. That's a strong assumption. So we put to test, to make sure that the fact that a hero banner takes an enormous amount of space in UI, is actually justified. Because it does.
Related Articles
Ownzones CTO Aaron Sloman discusses how Ownzones provides value to its enterprise video clients through big data analytics in this clip from his panel at OTT Leadership Summit at Streaming Media West 2018.
11 Mar 2019
Nice People at Work's Laura Delcor and Google Cloud's Luke Dyer discuss the impact of personalization and interactivity on the OTT industry in this clip from their Live Streaming Summit panel at Streaming Media West 2018.
13 Feb 2019
Microsoft Principal Product Manager Rafah Hosn discusses the benefits and limitations of a content personalization strategy based on supervised machine learning in this clip from Streaming Media East 2018.
18 Oct 2018
Sea Change International VP Marketing Kurt W. Michel explores the difference between "push" and "pull" video and the value of personalization in push/pull live streaming in this clip from his presentation at Streaming Media East.
02 Aug 2018
Sea Change International VP Marketing Kurt W. Michel discusses analytics- and AI-driven personalization in this clip from his presentation at Streaming Media East.
30 Jul 2018
NeuLion's Phil Green discusses how personalization can help content providers engage and keep OTT viewers and subscribers in this clip from Live Streaming Summit.
19 Oct 2017