What we learned by observing our users
UX research is now a core part of any product team. It’s no longer an afterthought or done only after building the product to find ways to improve it. UX research is now done at every stage of product development.
We talked to our users during every stage of product development. Despite constantly talking to them about issues they face and things they would like to change. There are so many things that they cannot tell you and the only way you can find that out is by observing them use the product in their natural state.
You can do usability testing by asking them to perform a task on your product while you observe them. But I don’t think that’s their natural state.
They are being asked to perform a task and they will try their best to do it. Their behavior is heavily influenced by your mere presence. The user behavior will be radically different when they use the product when they feel the need or urge to use it.
It’s imperative to understand in what mood or state the user is when they are on your product.
In our case they are probably sick, pregnant, in pain, or have an ill family member than needs to be taken care of. Depending on their condition, the user behavior and time spent on the website differs drastically.
For instance, soon to be moms or first time parents will browse 15–20 gynecologists or pediatricians profiles before they go to one. As compared to a person looking for a GP for a flu, he/she will settle down with one by browsing only 3–4 profiles. It’s imperative to know exactly who your users are and what pain points are they trying to solve by using your product.
We use Inspectlet to understand how people use the website in their natural state. It automatically records videos of users using the website where they are clicking, scrolling and spending time reading content. This helps a lot in designing the UI, changing the placements of CTAs and figuring out problems with low conversion. Moreover, it helps understand the issues customers are facing. It’s a really cool tool. Read more about tools we use at Meddy
Clicking on Elements that are Not Clickable
One thing we noticed that people were clicking on the like icons on Listing Page which were not buttons.
We saw quite a few users clicking on those Green Like looking buttons and getting frustrated when nothing happened. We could have easily fixed it by making them buttons that would’ve allowed people to easily like a doctor. But we didn’t want that to happen because it would lead to lot of users just quickly hitting likes without writing a review on the doctors. Liking a doctor is lot different then upvoting a product on Product Hunt.
Also, the like/dislike kind of gave a Facebook feeling which we didn’t like. The other major thing was that the dislikes on a doctor profile attracted lot more attention than the likes.
So we did couple of things to fix that:
First, we changed the like/dislike to Hearts. The way we thought about is what does it mean when someone says “I like this doctor”?
If you like a doctor that means you’d go to him/her again and recommend it to a friend. If you don’t like a doctor, you would probably not go to him/her again and certainly not recommend to anyone. Like/Dislike or Hearts are simply a way to express your sentiment about the doctor.
Second, we made the entire card clickable so if people click on the heart to like the doctor or anywhere in the card — they are taken to the profile of the doctor. As compared to the previous design where you had to click on doctor’s picture or view profile button to be taken to the doctor’s profile. This way users won’t be that frustrated when they clicked and nothing happened.
Third, instead of asking users to click on heart to express their sentiment. We would ask them in the review whether they would recommend this doctor to anyone or not. Answering Yes would lead to increase in hearts. No wouldn’t make any difference to the heart count.
Lastly, we would not show the review sentiment next to the reviews. Instead, just show the review to make users read all the reviews to decide if they want to go to a particular doctor or not.
Similar Doctors
From the user sessions recorded on Inspectlet. We noticed that lot of users would constantly go back and forth between a doctor profile and listings page. Here is how the user path would look like.
Home page > Dentists Listings > Dentist Profile 1 > Dentists Listings > Dentist Profile 2 > Dentists Listings > Dentist Profile 3 > Dentists Listings > Dentist Profile 1> exit.
The user behavior kind of mimicked like a shopping on amazon or any other website. The user would choose a product from the listing, read all about it and kind of like it. But would go back to browse around and find if something is better than the one he/she already liked. In the end come back to the one they liked in the beginning.
We figured there is lot of friction in going from one doctor to other doctor page. Users always have to hit back and go to the listings page to go to the other profile page. So we added a very simple section at the bottom of the doctor profile page of Similar Doctors with links to other doctor profiles for them to easily browse around.
This led to a massive 20% increase in pageviews on doctor profile pages. It also nicely blend in well with the flow. User would come to a doctor profile page, look at the picture, specialties, credentials and start reading the reviews while scrolling down. When the user reaches the end of the page he/she clicks on some other doctor profile page. Users no longer have to scroll up and click back to go the listings page.
Lessons Learned
- Users can’t always tell you what they want or if they are facing an issue. Because sometimes they don’t even consider that an issue that you think is!
- Understand why people are using your product and what urge or problem are they trying to solve by being on your product.
- See how people use your product in their natural state.
If you enjoyed this article, please hit recommend. That would be incredible.
You can also read this. It’s a post on why we redesigned Meddy
Tap the💚 below then follow me on Twitter if you liked this.
