Quikr Cars

Quikr Cars Post Ad

About This Project

Our post-ad pages are place where user would place their used vehicles on the platform. We currently had a separate desktop and mobile experiences. The process was iterative ones and we did usability test and conducted user interviews in each cycle to come to the best possible solution. I have broken the flow in design in Desktop and msite to highlight the design process.

Project Details

Quikr
UX, Visual design, UI
Sept '15 – Ongoing
www.quikr.com

Mobile first approach

Before proceeding further I would like to elaborate that we initially had a separate mobile site(msite) and desktop flow. Desktop was largely better performing than msite but as we had witnessed a gradual but steady increase in out msite traffic hence we needed to focus on the shifting pattern.

Status Quo

Old post ad in Mobile view

Our mobile flow was broken in 3 steps: image upload followed by user details and car details. There was a significant dropoff in each steps and as a result the overall conversion was poor. A lot of information asked was not required and some were outright obvious and due to some legacy reasons they were kept.

Evaluation of current page

Using unnecessary fields in the page
Dropdowns such as Brand, locality, city had too many option without any search
Incorrect usage of dropdown
Mobile method was heavily unoptimized
Error handling was irritating

User Research

As mentioned before we already have a generic flow that caters from Real-estate, jobs, cars and pets. It began with a simple craiglist like approach but soon balloned into a giant single page form.

Our initial user reseach was limited to usability evalutaion by fellow designers of the current flow and user interviews(telephonic) of users who either uploaded or tried listing their cars on platform. This helped us determine the current set of issues users are currently facing. Later on I started listening customer care calls to find issues

Who all users our mobile post ad

In the redesign we were more focussed on Individual sellers as they were facing the main issues and there were other method(via CMS tool) from which they can list their cars. Also car dealers were listing vehicles via android/iOS app or via Desktop. Also whatever improvements we make here would be passed to car/bike dealers.

In India, a lot of folks have mobile as the primary device to connect to browse the web and hence the mobile users were distributed all across; significant car sellers are from smaller cities or towns.

User persona of bike seller at Quikr

Problems

As mentioned earlier due to legacy issues we were using form element in a very inefficient manners. We improved the design in an agile method where we reached the final design by tweaking the design over time and getting constant feedback.

Post our user research we narrowed the user problems on priority level and thereby tried to tackle issues by priority. Certain problem we had to scope out due to tech constraint or platform constraint, such as the edit flow had to be done via legacy method only even though it came across as a critcal issue for us.

Information hierarchy was all over the place. Post the category selection the image selection(non mandatory) box seemed strange as per information. A lot of users reach the step then leave the process as they didn't have images at the time.
Unnecessary fields made the form longer and also difficult to fill. Legacy fields like title 'Ad title' was doing more harm than good. But a lot of users got confused and thought for the perfect title for the car. Same could be said for car description.
We had a lot of redundant fields which were not necessary any more. Also due to legacy issues we were using incorrect UI elements for many fields. Fuel type had only 5 options yet it was a dropdown.
Testing the flow in real setting highlighted the bandwidth issues and further discussion with the customer care team highlighted the issue of bandwidth.
In later stages of refinement we saw that our dropoffs would be highest in the image upload section. Compound that with users on bad network and it would almost be impossible to list your vehicle.
The current text copy was confusing for first time users. Key amongst it was the use of word 'AD' (Post you AD, AD details etc). Users always would correlate 'AD' with a paid listing.

Problem Solving

Number privacy issue

Urban users were apprehensive about sharing there details to random users. Also most complained about spam calls that many people get. We added a feature that allowed users to hide their mobile numbers and could share it via chat or sms if the seller thinks the potential buyer was genuine.

Initially we allowed users to post number publicly

Easing selection

We wanted to ease users in making the selection and we wanted to avoid big dropdowns and input that users needed to fill. The dropdowns were searchable but they users can select their car more easily as we populated common option for him.

Initially we allowed users to post number publicly

Image Upload

Image upload was the section with most hassle, in the conginitive walkthought a lot of users complained that they had no feedback what was happening during the image upload. There were tech issues as the mobile got hung up in the upload process.

We in our final approach decided to use an image compression on client side via JS to reduce the image what was being given to us. In the design part to optimize for the buyer UX we wanted a consistent landscape images.Also suggested a link like what kind of images should the users upload and order of images.

Initially we allowed users to post number publicly

Social sharing

A lot of urban users increased the buyers by sharing within the network and these buyers are generally more trustworthy and understanding. Also suggested listing their ads of group like Flat and Flatmates could even help sellers leveraging their network. Also on a plus point most of the these boards are individual buyers.