My Role

I led the redesign and improvement of the mostly commonly used feature of Zehitomo web application, the User Dashboard.

User-centered insight gathering and Ideation

I initiated the project by identifying user pain points from communicating with stakeholders and translated the user-behavior insights into product improvement ideas.

Design vision and communication

I created prototypes and design mockups to share the design outcome and vision with executives and stakeholders to align product strategy and gain buy-ins. This helped align stakeholder expectations from different divisions and push decision making.

Project planning and scope definition

I drove meetings with the management team to finalize features and project scope while balancing short term improvements and long term product strategy.

Coordination and execution

I worked with developers to plan for implementation details and feature release strategy. At the same time I coordinated with the customer success team for design validation, feedback gathering and user communication.

Background Information

Dashboard is the most used feature by our pro users (pros) where they can find new job postings (requests), check the jobs that they have applied to (applies) ,  chat with clients and manage their jobs. 

Before we go into details, I want to clarify one thing. There are 2 different ways that a pro user can apply to a job: apply manually or be automatically matched with a job that fits their criteria 

The original dashboard had 4 sections: 

New Requests - contains all the new job requests posted by clients that pros can apply to manually

In Progress - contains all the jobs that users have applied to (manually and automatically by the system) but haven’t gotten hired yet.

Hired - all the jobs that pros have been hired for.

Archived - expired jobs and jobs that have been archived.

In “In Progress”, “Hired”, and “Archived” sections, all the jobs exist in the form of a chat with the client in which the pro can discuss job details with the client and eventually get hired.

Understanding the Problem

In the past we have discovered many inefficiencies in the way pros use the dashboard and how the customer success team supports the pros while they use our platform to find jobs. I have summarized all the discoveries and feedback into the following pain points.

• For the pros, most of them have a lot of new matches coming in everyday in the “In Progress” section of the dashboard. It’s hard to identify a high potential client from clients who never respond. They have to scroll through all the active jobs, look at the messages and try to identify the clients with higher potential and then follow up.

• The status of the job is never accurate. Even if the pro actually got the job, the client never clicks hire on our platform, therefore the jobs will never be marked as “Hired” for the pro, which results in the jobs never moving to the “Hired” section of the dashboard. As a result, our customer success team needs to mark the jobs as hired for pros manually on request. 

• Sometimes, pros don’t know what to do when their dashboard has no content 

• It is hard for newly joined pros to understand how things work. Many people don’t understand what “Dashboard” means (especially in Japanese) and the sales team will have to explain the whole system to them in details

After identifying these pain points with the current dashboard, I talked to the customer success team and our pro representatives to find out more about what the pros care about the most. I also talked to the executives to learn more on their vision for the product to make sure that the solutions are aligned with our product strategy. Below are the key insights that defined this redesign project.

Clearly define the goal

After identifying the pain points and gathering key insights from users and stakeholders, I was able to define the goal of this redesign

• Make dashboard section organization easier to understand, more meaningful and helpful for identifying high potential clients

• Make dashboard more customizable to help pros find their job items in an easier way

• Provide guidance in the UI so that pros never reach a dead end and left not knowing what to do

• Make sure the redesign is aligned with the strategy of moving away from manual apply and towards auto matching

The Redesign

Reorganization

First, to make dashboard organization easier to understand and provide meaningful information for the users, it’s important to rethink when each of the dashboard sections is being used and what value are the sections bringing to the users. Like stated earlier, users don’t find the Hired section useful because the status of the jobs there is not accurate, and they use the “In Progress” section the most, but the valuable information is often hidden under the less important information. To address this problem, I decided to get rid of the “Hired” column and separate “In Progress” content into “New Matches” and “Under discussion”(やりとり中). “New Matches” now contains only the jobs that pros just applied to but haven’t heard back from where “Under discussion” contains jobs that received client response. This way, by getting rid of the not so valuable “Hired” section of the dashboard, we can add a section for what pros care about the most --- client engagement.

After this reorganization, we are left with New Requests, New Matches, Under discussion, and Archived. This made me think again, can we simplify it even more? This brought me to think more on the pain point where many pros have a hard time understanding what dashboard is. Dashboard is kind of a technical term and it’s not very descriptive of what it is used for, because inside the dashboard, there are client requests that pros can apply to, there are also jobs that they applied to that turned into messages. Why not categorize them by the status of the job, unapplied, which means it is simply a job request and applied, which means it is not a chat with clients. If we categorize them this way, we can actually separate New Requests out from the rest of Dashboard and we can rename Dashboard to something more straightforward and representative of what it stands for --- Messages.
By taking New Request out from Dashboard to the main menu and changing Dashboard to Messages, we made the purpose of each section clearer. This was also done with the intention that if one day we decided to completely get rid of manual apply and focus on auto apply only, then it is possible to get rid of the entire New Request section without touching Messages.

Customization

To make each of the dashboard sections more customizable so that users can find what they are looking for more easily, I added a filter button and a search bar to each of the sections. 

Filter button provides different filter options on the status and properties of the job such as starred vs no starred, Client available vs unavailable for phone call, replied vs no replied. This helps pros filter and identity the high potential clients. I also added a Hired vs No Hired filter in the “Under discussion” section to provide the option to see hired jobs since we got rid of the hired section. 

Search bar provides filtering on job details, such as location, job type, or even client name. Pros can filter the jobs with specific details and find what they are looking for right away.

More guidance, more action, more success

With the new section organization, we help the pros focus and prioritize important information by pointing them to the “Under discussion” clients, however at the same time, it could be discouraging when pros don’t have a lot in the “Under discussion” section especially when they first start using the app. To support our pros every step along the way and provide them with useful insight towards success, Customer Success team provided their insights on the most useful actions to take when pros have an empty inbox, and we added them as the placeholder content when there is an empty inbox to encourage pros to take related to get more jobs.

Communicate the purpose of the redesign to our end users

How new features and design changes are introduced to the users is as important as the actual redesign. In order to communicate the purpose of the new design to both existing users and new users, I added an introductory tutorial page for the New Requests and Messages sections to explain what purpose each of the sections serves.

Design Validation & Rollout Strategy

When it comes to validating the redesign and collecting feedback, we first ran a few usability tests with an interactive prototype with some of our pro users. This was to get sanity checks on whether the general direction of the solution we provided was a good direction and whether it helped solve some of their frustrations. We asked users to navigate through the interactive prototype and asked them to find their jobs and messages by using the filter and search bar. At the end, we asked some followup questions to get their feedback. The test users thought the new organization was very intuitive. They said having the filters was very helpful for them to find what they are looking for. They didn’t understand some of the filter options very clearly and we adjusted the wording according to their feedback. 

Because this is such a big redesign of the existing site, we needed to have a very well thought-out release strategy. Since we have a large number of pros who are using the dashboard feature on a daily basis, we do not want to disrupt their behavior abruptly and without getting their consent. After discussing with the engineering team and customer success team, we decided that the best way to present the new design to our users would be a soft release with a way to “opt in” to try the new design. We will present users with options to try the new design and stay with the old design. If they decide that they don’t like the new design after trying it, they could always switch back to the old design. We also provided a feedback survey on the site for users who tried the new design to send us feedback on what they think of the new design. This way, we were able to roll out the new design and collect feedback from the users without forcing it onto the users. From an engineering perspective, we kept the old version of the dashboard while building a completely new one and made the two interchangeable with one switch. Because of the way it was planned and implemented, we were able to continue to deploy new code to production without showing any of the new design to users while making sure that none of the new code was breaking production.

Reflection

This entire redesign project took almost 3 months from understanding the problem, ideating, to design, development and complete rollout. We experienced some delays. Many times we had to take a step back because we decided on one solution and encountered problems while pursuing it whether it’s design issues, technical difficulties or strategy changes. Although these issues did prolong our project timeline, I do believe that the time we spent going backwards were worthwhile and saved us from future struggles. At the same time, I do think this project could have been rollout in smaller increments if we were not so ambitious in solving everything with one big redesign. That might be something that we could improve on in the future when tackling a big redesign like this. Currently we are collecting feedbacks from uses and will plan on implementing smaller adjustments and improvements in the near future.