Session 1: Introduction and evaluation of Feedly

Today was the first session of the Human-Computer Interaction (CHI) course. We were introduced to the course and evaluated Feedly’s user interface as an exercise.

As with every course, we started with an introduction. Long story short: the user is always right, and the fault is always with the developer.

In the afternoon, teams of three were formed for the project. As an exercise, we then evaluated Feedly (an online news aggregator) from a user’s point-of-view. Our team found quite a few good positive remarks about the application:

  • Green buttons lead the user to the primary tasks: Get Started, Follow and Add New Source.
    This makes it very clear how to start using Feedly. The focus is on effectiveness to ensure that new users actually make it to and through the sign up process. We find this a clever choice: Feedly really needs users to sign up.
  • Users only have to log in when he wants to follow a source.
    This allows them to experience the look and feel of the application before having to register. We think this is a good choice: requiring the user to register immediately would probably scare away many new users.

Our major issues with the user interface were the following:

  1. The Preferences page is just a long vertical list of options.
    We found this completely unusable. Users have to scroll through the whole thing trying to find the appropriate option.
    Screenshot of Feedly's Preferences page with too many options.
  2. When clicking “Follow”, the user has to click “Cancel” on the bottom of the opened side pane to exit. Clicking on the empty space on the right does nothing.
    The “Add New Source” screen looks like a modal dialog (similar to the article preview window), we assumed we could cancel it by clicking outside the dialog. In some way, it makes sense though: users would probably be more annoyed when they accidentally close the dialog and lose their work on categorizing the new feed. Effectiveness is more important than efficiency in this case.
    Screenshot of Feedly's Follow screen.
  3. Responsive design: the side menu opens and stays open when the browser window is large enough.
    Although this is fairly common in modern web applications, we think it may still confuse some users.
    Screenshots of Feedly's side menu opening while resizing window.

We also found some minor issues, including:

  • Drop ‘n drag isn’t supported in the feed list as it is on the organize page.
    We find this confusing, we expected to be able to do some basic management from the sidebar’s feed list. Apparently, the sidebar only allows for navigation.
  • Some buttons are hard to find.
    Buttons on articles such as “save” (in any feed) and “unsave” (in the Saved for later feed) are in a small, grey font and only show on hover. We think these are too well hidden to be used effectively.

When all groups had finished their analysis, we went through the most important remarks. There were many different issues with the application, which is a bit unsettling given that the website serves some 12 million users. Often, it was unclear if a certain issue was an actual bug or an intended feature. For example, you need to log in when you want to follow a feed, but you don’t need to be logged in just to browse a feed. Some students thought that the user should have to log in before browsing a feed, others thought that they should not even have to log in before following a feed (and instead store the followed feeds in a session).

We concluded that users all have very different issues, with some issues even conflicting between users. This makes the task of designing a good user interface with a good experience for all users very challenging. With this project, we have to try to achieve just that. Follow us to see how we do!

EDIT 1: Own remarks were added
EDIT 2: Added more detailed explanation and opinions to our own remarks, as suggested by our friendly commenters. Thanks for your input!



  1. I find that some of your remarks on Feedly are more like established facts than they are remarks. Maybe you could add some more opinion into it? I.e. why do you mention it, what do you think of it.
    For example, I don’t see what you mean with the first remark (the green buttons): do you find it a problem or rather something positive?

    I noticed the drag and drop problem too. Apparently, it does work when you drag the feed title (as opposed to the rectangle). The drag icon on the entire rectangle is really misleading, one more thing they should fix!

  2. I am not really sure the session thing will be so a good.
    Sessions tend to expire and nothing is so frustrating as lost and unsaved work.

    Just searching for feeds and viewing posts, like it’s now, is more then enough for me as a teaser to what feedly can do.

    They would probably have more success if they also accept other registering sources like facebook or so.

    1. We felt the same way: storing the feeds inside a session makes it very likely to lose the user’s preferences. That should end up in some kind of permanent storage, so you’ll need to be logged in by then. For the record, we/ find the current approach just fine.

      Adding more options may help to attract new users from different sources, but I think it may also scare away users if too many options are added. It’s a thin line, and they seem to do just fine with just Google users.

  3. Personally, this post reads more like a diary entrance of the lesson than a review of Feedly.

    Not that it is not a good post, of course, but it might be better to focus more on Feedly itself.

    Also, try to elaborate more on your own opinions. The short bullet-points of your evaluation only seem to identify the problem. They do not really seem to share your opinions about them very well, which is unfortunate because I was really looking forward to read them.

  4. Would you recommend someone to use Feedly, or rather say, “its okey but you better search for something else”?

  5. A nice short and concrete summary of the afternoon. Maybe some screenshots to illustrate your examples would have been nice?

  6. The bullets with your remarks are really easy to get a quick overview of Feedly. But maybe it would be useful to talk a little bit more detailed about one or two points (those who really cached your eye).

  7. From the title I expect a concrete evaluation of feedly but what I get is just a description of what happened in the first session. What I miss is the evaluation of Feedly itself: a description of what you think are the problems with the remarks you mention and some screenshots to make your point more clear.

    Besides this the post reads well, but I would discourage to use bullet points because it does not encourage users of reading the rest of the article.

  8. We hear you! We’ve updated the blog post with more detailed explanations and screenshots for the major issues.

    Thanks for your input! 🙂

  9. nice post, summarized the exercise in a clear and concise way 🙂

    I really like that you give a conclusion that is focused on the course, and not on Feedly, because making connections to your own project can be very useful. Nevertheless, I’m wondering a bit what your final conclusion on Feedly itself is. You could say that it isn’t that good because of all the issues, but on the other hand, maybe it’s not possible to please everyone, and Feedly is actually doing a pretty good job?

    I also agree with you that the log-in before subscribe idea is a good thing, and I strongly believe that sessions would be useless in this case. 🙂

  10. I read the version that was apparently updated based on earlier comments. Now, it looks quite good, and indicates some typical trade-offs in user interface design. I also like the fact that you start with the Good News. (Though you could leave out what happened in class when it is not so relevant…)

  11. I like the structure, the bullet points make it very clear and easy to read. I also like the pictures provided within the text. I agree that reading this post feels like reading something like a diary entry, but personally, I like it, it’s easy to read.

  12. Personally I agree with your explanation why the sidebar only goes away with cancel. An accidental click is easily made so it makes sense to prevent it from happening.

    I also think the current ‘try-before-you-sign-up’ approach goes far enough. You can see most functionality without an account and only when it makes sense too, the application asks it. More options than only a google-account would have been nice.

Comments are closed.