Expedia UX Redesign
+ Crawling & NLP

This case study was completed as an independent project. I was challenged to make UX design with help of Crawling and & NLP

Expedia is an online travel agency owned by Expedia Group, an American online travel shopping company based in Seattle. The website and mobile app can be used to book airline tickets, hotel reservations, car rentals, cruise ships, and vacation packages.

Project summery

Challenges

I was checking the Expedia site and saw that there is a lack of information that could help the Users in the purchase process. That is why I looked at the site's reviews and saw that most of the people who gave 1 star were unhappy just because they were not aware of the rules and that was for lack of information or presenting it properly.

here are examples of lack of information in buying process.
as you can see in seat choice, all the item all the same but the prices are different and it's the time that Users get confuse and maybe Users drop the purchasing progress.

solutions

Users get the information they need better and don't get confused when shopping

crawling + NLP
I developed a crawler for the Trustpilot site (where you can find mostly negative feedback from Expedia) as a side project to investigate user needs and problems. The crawler got through mostly, 255 pieces of feedback and analyzed them briefly for some information that I could use to address the users' problems.

result

255
Number of review
108
refund subject
97
customer service
60
cancelled

During my research, I also found that people get confused about the refund process. Maybe they don't know the rules, or maybe they're different from other companies.

Design process

I asked four Users to go through a purchase scenario up to the checkout stage. To understand their moods and thoughts, the test was monitored
Additionally, I asked for "refund" rules, since most reviews discuss refund problems earlier.

research results

- To buy a ticket, they had to pay attention many times to choose a ticket
- The forms were too long
- There is no information in some parts, such as seat and ticket selection, and it is not possible to compare them, which confuses users
- they can not find refund rules easily and there was hard to read.
I found out 2 problems, one of the user, don't set his seat while he was choose the expensive seat just for choosing seat. how that happen ?
as you can see the call to action button in left page is Check out. after the user click to the button, site will remind to choose seat but still the call to action button is go to check out so the site unpurposely lead user to go to check out without choosing seats.

refund problem

the result of research shows that most of the problems users are from refund. when I asked users to find refund page and read it. they have some difficulty, so I take a look at the page.
as you can see the refund rules are like this, ALL CAPITAL, AND ITS HARD TO READ FOR MOST PEOPLE, ESPECIALLY WHEN ITS LONG PARAGRAPH. maybe most of the users find there answer here but they are refuse to read it.

Design

In this section, we have put the list of potential features in the diagram to make the choice easier for us
Site features were placed where pinpoints existed on the site as before
The same process was repeated by four previous users, see the results:
- Three out of four users were more satisfied with the new design
- the rules were much more readable to them
- they can find refund rules easily
- they can choose tickets easily and without confusing

Conclusion

the design itself is an iterative process, and we always try to go toward better versions. in this case study I want to show another aspect of user research, The Big data, and how we, as UX designers can benefit from it. Thank you for staying with me throughout. :)
Powered by Webflow