May I offer you a suggestion?

The good and the bad of reply suggestions.

Melissa Vela
UX Collective

--

A whiteboard with orange, yellow, pink, and blue sticky notes on it and a paper with a red arrow that reads suggestions.
Image by bluebudgie from Pixabay

Who doesn’t like to save time? I would assume most people do, and with the amount of time we all spend online, it’s pretty convenient to have time savers built into apps and websites. There are many accelerators out there like swiping, double-tapping, and using keyboard shortcuts. But the time saver I want to talk about today are suggestions. We’ve all seen them from Google search to LinkedIn.

A screenshot of a Google search for “what are” with the suggestions undernearth
Photo credit: Google

When done properly they help to prevent users from making a mistake or slipping up. Suggestions are a great example of Jakob Nielsen’s fifth usability heuristic for user interface design which is error prevention.

Let’s take a look at some examples from LinkedIn. For the most part, LinkedIn provides users with good message suggestions. They are coherent and fit the context of the conversation. LinkedIn uses very basic and general phrase responses.

Screenshots of a LinkedIn chat messages with good reply suggestions
Photo credit: LinkedIn

On occasion LinkedIn will not give the user any response suggestions at all.

Screenshots of a LinkedIn chat messages without any suggestions
Photo credit: LinkedIn

LinkedIn has not yet perfected it’s Smart Replies since they utilize machine learning, but they will definitely get better over time. The reply suggestions can sometimes seem a little odd, although they do tend to remain on topic. For example if you tell someone thank you, and they respond thank you back, the reply suggestion might say something like, “Likewise,” or “Same to you.” In that context, it would be strange to say that since you were the first one to say, “Thank you.” If the other person would have been the first one to say, “Thank you,” those response suggestions would have made sense; however, they are redundant and sound odd in the example I gave. Let’s take a look at some of these redundantly odd suggestions that LinkedIn provides.

Screenshots of a LinkedIn chat messages with slightly odd replies
Photo credit: LinkedIn

Now, as you can see, LinkedIn is not perfect by any means, but their suggestions do stay on topic for the chat. Over time, the suggestions will probably get even better.

Another social media site, Facebook, has decided to jump on board with adding comment suggestions to its mobile app. I first noticed these suggestions being offered last Friday. At times that suggestions follow the topic, are spot on, and make perfect sense.

Screenshots of Facebook posts showing good comment suggestions
Photo credit: Facebook

Facebook, unfortunately, has a long way to go in order to make its suggestions coherent. It is as if Facebook is acting like that one kid in class who just says random things out loud to hear him/herself talk. Some of the suggestions that Facebook gives are so far out there, that it’s not even clear why they are suggestions at all. These types of suggestions in no way benefit the user or help the user to avoid making errors. They in fact do the opposite if a user were to accidentally hit on one of these wild-card comment suggestions.

Screenshots of Facebook posts showing odd comment suggestions
Photo credit: Facebook

But then things get even weirder. I’m not sure if it is Facebook basically messing with people or if the suggestions are really that far off. But several people on my friend list posted a similar post about wanting to see what Facebook would suggest for comments on said post. The suggested replies are very, very odd at times to say the least. You should go ahead and give it a try if you haven’t already. Some of the responses are pretty far out there.

Screenshots of Facebook posts showing very odd comment suggestions
Photo credit: Facebook

To sum it up, when done right, comment and reply suggestions can be useful and save the user a lot of time. When done poorly, they are a distraction or a cause for laughter at best, and at worst, they make the users more prone to making errors.

The UX Collective donates US$1 for each article published in our platform. This story contributed to Bay Area Black Designers: a professional development community for Black people who are digital designers and researchers in the San Francisco Bay Area. By joining together in community, members share inspiration, connection, peer mentorship, professional development, resources, feedback, support, and resilience. Silence against systemic racism is not an option. Build the design community you believe in.

--

--

UX Researcher striving to make a positive impact on the world. I’m passionate about travel and other cultures. I love learning languages.