Improving The Efficiency Of Kiewit’s Internal Search Portal
My Role
UX/UI Intern
Design Tools
Figma, Mural, Fullstory
Research Methods
Usability Testing, Secondary Research
Project Length
13 weeks
Overview
Kiewit is one of the largest construction and engineering companies in North America.
Context
Employees can be categorized into 3 types:
Skilled Craft
Eg. Laborers, mechanics, welders
Professionals
Eg. Field engineers, Civil Designers
Technology Group
Eg. Project Managers, Software Engineers
Problem
Kiewit employee’s are wasting time and resources trying to navigate Launchpad, their internal portal.
Launchpad provides streamlined access to company approved software applications. Despite the portal having all the resources they need, employee’s are struggling to search for applications, access training material, and seek the appropriate support within the portal. This leads to project delays and puts a strain on their support team.
Solution
Introducing advanced search features to Launchpad.
Files can be tagged with key terms, allowing users to discover related files. Additionally, users have access to various filter and sort options, helping them find what they need faster.
Impact
Simplified user flow and identified design flaws for future iterations.
By incorporating new search capabilities, Kiewit will cut down the average time spent looking for specific applications or documentation. Through usability testing, I uncovered new issues regarding how users look for support. These new pain points will be addressed in future sprints.
How Might We Improve Search Capabilities Within Launchpad?
Project Deepdive
Service Blueprint
I selected the highest priority features after meetings with the product owner and stakeholders.
Before my internship began, the research team conducted extensive research to identify the major pain points within Launchpad. I picked up where they left off, using these findings to document the products current state, pain points, possible new features and feature updates.
The highest priority features for this project were Advanced Search, Customer Success Support, User Feedback, Mini Launchpad, and Material Tagging.
Wireframing
I facilitated two wireframing workshops with several stakeholders.
I collaborated with stakeholders to better understand Launchpad’s current unmet needs and the expectations surrounding possible new features. This provided a space for all team members to contribute their insights into how these features should function.
Four features were desired by most stakeholders: search suggestions, filter / sort options, material categories, and a support button.
Hi-Fidelity Prototype
After several iterations of low-fidelity designs, I finalized the new features and began working on the Figma prototype I would use during testing.
Advanced Search
Before
Material documentation is only accessible through information icon.
Search bar only allows searching through applications.
Results can only be viewed alphabetically.
All material is accessible through the search page.
Users can search any type of document.
a) Filtering by material type.
b) Additional filter and sort features.
After
Ineight Support
Users struggle to find Ineight support services as they are only accessible through the information icon.
Before
Added a direct link to Ineight services on the search page.
After
Tagging Material
Product owners added material documentation within specific application folders. This made it impossible to search through the different files.
Before
Added the capability to tag material with key terms. Allows users to find files using key words, rather than memorizing the folder they’re in.
After
Usability Testing
Participants
I conducted 7 usability tests with Kiewit employees.
0 - Skilled Craft
4 - Professionals
3 - Technology Group
Goals
Observe whether participants can navigate to their desired material.
1.
2.
Learn where users are currently seeking support.
Prototype Adjustment
During my first usability test, the participant ran into major issues with the Advanced Search screens. I was concerned this was lead to unactionable research findings. Therefore, I created an alternative version that addressed my concerns. The remaining testing sessions were conducted with both prototypes.
Advanced Search
Icons take up a lot of space, but don’t help users find what they’re looking for.
The application names are very small and hard to read.
File names are being cut-off.
Version 1 - Icon View
Version 2 - Table View*
Allows users to read the full file and application names. Can be easily skimmed.
* Version 2 is a mid-fidelity prototype as it was created in-between testing sessions.
Research Findings
After synthesizing my findings, I uncovered several pain points. I presented these to stakeholders and provided recommendations for next steps.
Icon View vs. Table View
5 of the 6 participants preferred table view.
Only 3 of the 7 participants found the file they needed.
Findings # 1
Eliminate icon view completely OR allow users to toggle between icon view and table view.
Make application names more prominent so users can distinguish between files with the same name.
Create a feature that prevents files with duplicate names.
Next Steps #1
Icon View: “Definitely more cluttered and hard to consume”
Table View: “It looks easy to search through… you can sort of see everything right there in plain sight”
Overlooked Filter & Sort
5 of the 7 participants did NOT notice filter and sort features in icon view.
4 of the 6 participants tried to use filter and sort features in table view.
Next Steps #2
Conduct further research on icon view to determine:
Do participants believe the filter and sort buttons are unrelated to the material?
Where do other search engines place their filter and sort buttons?
Finding # 2
Icon View: “There is no filter or sort functionality here”
Table View: Participants successfully noticed the filter and sort features
Confusing Feature
Finding # 3
4 of the 7 participants noticed the “Ineight Training Help Feature”, but only 1 clicked on it.
Conduct further research to determine:
What do user expect to happen when they click this button?
When would it be appropriate to select this feature?
Next Steps #3
When looking for support, participants noticed the “Ineight Training Help” button, but did not click it.
Bonus: Confusing Support Options
When looking for support, 5 of the 7 participants struggled to determine who to contact.
Finding # 4
Provide more guidance regarding how users should seek support.
When do they submit a ticket vs. reach out to management vs. seek external support?
Who should they contact first, Product Owner or Product Manager?
Next Steps #4
“I don’t know that it’s clear to me whether I should e-mail these people or whether I should submit a KSS ticket”
Project Outcomes
Impact
After presenting my findings, I passed along my work to the ux team. During the upcoming sprints they will…
Iterate on my table view design.
Conduct further research on the filter and sort features.
Ship an updated version of Launchpad in the next few months.
Reflection
Lessons
Update management throughout the Design and Research process to generate buy-in.
At the beginning of this project I took stakeholders and product owners input on design features, but I didn’t check-in again until the end of my internship. In the future, I would like to update them throughout the process so they’re more cognizant of what the UX process looks like. This would have been especially useful as I was the only designer / researcher on this project.
Learning when to push-back on a design decision is crucial.
When I uncovered major issues with icon view during the first test, I wasn’t sure how to proceed. Mocking up alternatives didn’t feel like an option as the product owner had been very adamant on icon view. The UX team encouraged me to create alternative designs and provided their support by backing up my design decisions and results.
Provide engineers with all design states.
During the beginning of this project I wasn’t designing any error states. This led to the front-end engineer making assumptions on what they would look like. Later on we had to go back to these screens and re-design them, matching them up to the existing design system. In the future, I would consider these earlier on to prevent repetitive work.
Kiewit Luminarium building site, a science center along Omaha’s waterfront.
Next Steps
Iterate on the first version of table view.
Despite my research showing that it was preferred over icon view, stakeholders seemed hesitant to accept it because it’s less visually appealing. I would love to iterate on this design to make it efficient and pleasant to look at.
Conduct usability tests with the appropriate users.
Although Launchpad is used by all employees, there are certain features that only certain employees have access to. In the future, I would like to test each feature with users that more closely match the target users.
Work on individual next steps, listed under research findings.
Other Projects
Increasing the number of RealID appointments that are successfully booked via mobile devices.
Coming Soon!
Oregon DMV
Increasing the number of immigrants who can legally avoid deportation, providing them with employment and education opportunities.
ImmigrationHelp.org