HR. Help
- Tullie Design
- Apr 1
- 16 min read
Updated: Apr 24
Designed and developed a global HR content portal that delivers personalized, essential information directly to users' devices while providing a deep-dive portal for further exploration.
Problem
HR was being bombarded with inquiries from a variety of communication avenues which weren’t in sync with one another. This made it complicated for the user to find help and for HR to keep track of where the contact was and the next steps in helping.
Their tools for HR self-service had poor adoption and sentiment from users with only 31% of employees finding it easy to get assistance from HR.
Of those 31% of employees, ~100% of them were employees that had been with the company for over 8 years, which could mean that those employees have figured out a clear path for themselves - however this doesn’t serve the other 69% of employees in the company old and new.
With 73% of employees asking a colleague before “bothering” HR, the blueprint of HR help was less than desirable.
" It's very hard to find clear information. That's why I prefer to contact a person" - Employee, France
For the US alone, there are 47,000 employee interactions with HR over Microsoft Teams chat for a total of ~482,000 messages monthly.
Solution
We created a global platform that was outfitted to meet the needs of global employees in an inviting and intuitive way, so that employees can get the assistance they require.
Instead of focusing on shifting left we aimed to serve up a portal that targeted the users individually by delivering information and data before they had to go searching - this data would be most relevant to their region, rank, service line and other individual data.
Research
Experience Mapping
We created a global platform that was outfitted to meet the needs of global employees in an inviting and intuitive way, so that employees can get the assistance they require.
Instead of focusing on shifting left we aimed to serve up a portal that targeted the users individually by delivering information and data before they had to go searching - this data would be most relevant to their region, rank, service line and other individual data..
We mapped various paths from different regions and found that employees from the same regions had a different mind map of getting HR information. This included a variety of different sites that were unrelated to each other. With no clear “go to” for employees, we started to see why they were asking colleagues or going straight to an HR chat as a solution.
From this mapping exercise we found that in most cases users would be going in circles only to end up abandoning the process of going down rabbit holes that didn’t give them what they needed.This made it very clear why the employees continued to engage with HR through direct channels rather than to self-serve.
"HR expressed sympathy for employees because they know things are hard to find, most saying some version of 'It's hard for me to find things and I'm in HR, so it's got to be impossible for them!" - HR consultant
Qualitative Research: Understanding the Content
Content is too difficult to consume: Policy content that was of high value to employees was presented and written like legal statutes and clauses instead of layman's terms which would be understandable and digestible.
(i.e: One site gave this answer to the user searching for ‘Additional policies and Guidance’ - "Professionals should comply with the Universal Policy on Secondments (link) . It does not apply to the provision of professional personnel to a Client by a Member Firm which should be accomplished pursuant to Loan Staff Contract for loaned staff services, where permitted. Refer to clauses 4.5 and 4.6 of this Policy for further details.” - How does this make sense to the user?)
Users are going down rabbit holes:
Users are clicking on links on links on links that either weren't serving up the information they needed, offering them suggested content that wasn't related or taking them in loops with dead-ends.In some cases employees being taken to a region that the employee was not hired in (and therefore the policies and information would be not applicable)I.e.: An employee searching for “travel and expense policy” went searching on a US site ended up on a different site for a travel policies in the Netherlands.
Getting to the answer:
Content was long and convoluted. Any reference to content was not immediately accessible or was missing. In the best case scenario it took the user 8 clicks to change their buddy/counselor (There scenario was where the user wasn’t taken down a rabbit hole that wasn’t broken or invalid)
Duplication of information:
The Various sites the same topics with similar title but with different information.
(i.e: 'Bereavement leave [policy]' has similar titles but the information under the header pertaining to content was different. In most cases the information inside each of the duplicate headers also contained different information.)
Confusing and missing Information:
In some cases users were taken to a page that says “On this page you’ll find information related to updating your personal information on [another site]”. Once the users transitioned to the other site (which opened up inside a new tab) they were able to see a tab that said “My HR Data” that visually suggested that it was clickable, but didn’t exist.
Content labeling that creates confusion:
Accessing policy content results in employees guessing where to look for answers because of similarities in contact titles.
I.e.: An employee searches for Parental Leave and the search results deliver “Family Leave Policy”, “Family and Medical Leave Policy” and “Personal Leave Policy”The policy labels are redundant and unactionable adding to the frustration of the experience.Clicking “View” does not present content as the employee expects and they cannot access the material necessary to self-help.
On a search results page for the US, Leave policies then included ‘Personal Leave Policy - French’, which has no relation to the US. If they included another countries’ site then all the countries should be included, but there should be none at all.
Labels do not provide navigational or usage guidance:
Product names and HR specific terms make it difficult to navigate and understand how to best get assistance.I.e: Words like “HR Assist” and “HR team” would be used to give the user hope that they were being placed in a location where they would have their needs met, but they would be taken down another rabbit hole.
Main HR site confusion:
The main site that was designed for users to find the entire portfolio of information across the globe, references their own site multiple times on the site as though the user wasn’t already on the site itself.
6 of 14 sites that users would go to for HR help
“Employees overwhelmingly said that lack of single destination was at the heart of self-service challenges.” - Research
Multiple clicks to get an answer:
Employees searching for information would be clicking links upon links to get to an answer if they weren’t taken down a rabbit hole
I.e.: It took a user a minimum of 8-clicks to find out how to change their company counselor.
Non-relevant and unrelated information:
Users were getting information that took them to a multitude of sites instead of seeing all of their personal information in one place.
A user should be able to see the personal information (name, address, ID, social security/national insurance, visa information etc) including region, rate, rank, compensation, accounts, pay stubs, tax statement etc in one place, but they were being linked to other sites that required more login information or other actions to access the information they were looking for.
I.e.: Users would click on insurance and be taken to “All state insurance” to log in to their associated account. Meanwhile they may not know what their sign-in credentials are.
Qualitative interviews
We identified the top topics that employees were going to HR for in order to understand the hierarchy of information that we should be serving up when a user get to the HR Portal.This meant that we didn’t have to go into card sorting to define their mental model but relied on data from previous employee requests.
However we had to take a hard look at the content that was being served up at the same time.
It wasn’t just the multitude of places to get information but also the lack of information that is sufficient to meet the employees needs.
Quantitative findings
Based on research we found that given a good experience that 82% of employees would spend up to 30 minutes trying to self-serve before reaching out to an HR agent.
“I prefer to look for things myself but there needs to be more consolidation, so you aren't going from [site name], [site name], [site name], etc. - A one stop place with all the main helpful links.” - Senior, Audit. US
~80% Many employees stated a chatbot/AI would be ideal as a starting place if it works properly
Almost all employees expected that if the chatbot was not able to answer, then it would connect them to a live chat agent
9/10 employees didn’t have a good experience with the current chatbot solution that had been released in the last year which has caused a heavy amount of distrust in any other future products to be able to meet their needs.
100% of employees said that knowing a live support agent would be available made employees more open to first trying a chatbot/AI for support before loading their burden onto a live agent.
Conclusions of the research
What we found from the research is that users want one place that they can get the information they need.
In addition to that we found that users typically prefer one method of information gathering over other methods with search being the most dominant. Other methods of information gathering included: • Browsing - Where users peruse topically without knowing exactly what they are looking for.
• Gathering - When users look for multiple pieces of information which may or may not be related topically but carry a relationship and importance to the user.
• Being notified - Users appreciate being made aware of information through reactive digital messaging.
• Conversing - Where users have a conversation to help guide them to take an action or complete a goal (this could be online or offline though an AI tool, chatbot or messaging)
In summary of the research findings
There is no perfect way that users were gathering information as they all approached their searches in different ways depending on what was available to the user, but it was time to deliver a platform that would capture all the ways in which a user was looking to gather information. It was however clear that the primary goals for users was to verify that they are getting the correct information, which means:
• Trusting that the verification of the topic is correlates to the users need
• Being familiar with the primary topic
• Understanding the specificity of the topic
• Comprehending the timeline pressure on the information gathering process
• Acknowledging the the privacy of information that is used in the gathering process
• Knowing that their information if up to date
Next Steps: Taking what we know and setting the stage
Taking the research and exploring the opportunities - we conducted a workshop where the team spent a week in Georgia USA with the global team that had been working on all the aspects of the project. From stakeholders, to Project Managers, Product owners, Analysts, designers and research - we came together to explore our findings, thoughts and expertise to build a site that would accommodate our global employees to the best of our ability.
From Sticky notes, to card sorting to Information Architecture, group thinking, siloed think tanks and everything in between. Each asset delivered their findings and then we workshopped our way into design. It was important to the design team that we offered the opportunity for other team members to show what their solution to the problem my look like.
I find this always gives the designers an opportunity to see things from outside the “designer mind frame” with all their training and which easy we think we should go vs the mental model of the users who aren’t the designer. It was great insight to see our other team members' versions of how they envisioned the site and then voted on a rough idea of which way we were going to move ahead.
Design
Wire framing the foundations
We created a wireframe from the Global HR site that users were meant to know to go to (despite it not giving the user what they needed at all times)We wanted to see if any portions of the current site would be valuable to bring along into the new design.
Serving up the necessities - Targeting our priorities:
Set a Destination: We initially thought that the design should be a spoke that floated on all the pages where users may go for help from HR, but after exploring the importance of all the ways in which a user may want to interact with the product we realized that although a spoke may be able to help the user inline on the page that they interact with the spoke on, they may need to dig in further for more information and be able to save portions HR to their own profile which wouldn’t be possible with the spoke model alone.So we landed on the destination being a HUB and Spoke design.This aligned with employee mental model and telemetry data.
Create Constant Awareness: Create a spoke design. Remove any other confusing HR names and replace it with a spoke that will not interfere with any of the other site but hover as a resource on all other pages - meet people where they are and improve their search results.
One-Size-Fits-All: Search, browse and chat need to be the same for all users
Shifting Left no more: When a chat-bot isn’t meeting their needs they need to be handed off to a human to help, however research showed that Employees are open to
chatbots/AI if it can provide accurate information and is easy to find
Digestible Content: needs to be easier to consume with additional contextual support around the content. Copy needs to be clear and concise in layman's terms so that it is easily digestible without forcing a user to look into topics further.
Concise Labeling: Align labels with mental model and fuzzy match any like terms with content I.e.: “Maternity Leave” may also be “Parental Leave”, “Paternity Leave” or “Family Leave”.
Global Language Support: The site needs to translate into all the languages used for conducting business at the company.
Disability Options: There should be an option to dial-in or get called instead of navigating a site. (*Specifically driven from US and UK data)
Automatic User ID Identification: The site should understand a user and their region role and service up the most commonly asked questions before the user has to go searching for them.
Self Profile Management: Users should be able to change the information about their personal details and updates without having to go through an HR person directly.Notifications: Giving users insight to changes, updates and processes.
Building on to the foundations
As we built out the Information architecture of the product we landed on a rounded out view of the intial product as a team and began to build around the foundations.
The Destination - The Hub: The destination should be a HUB and Spoke design.This aligned with employee mental model and telemetry data.
Constant Awareness - The Spoke: The new spoke hadn't been implemented on any of the companies sites before and would comfortably sit on other site pages without interrupting the flow or information on a variety of sites. The spoke is expandable and collapsible for easy access and to give the user access to a full range of the products uses.
One-Size-Fits-All: Because of the streamlining of the content and the UI the Search, browse and chat is the same for all hires, roles and regions, as we will find out through the references of our new design.
Shifting Left no more: When a chat-bot isn’t meeting a users needs, they are put in contact with a human.
Digestible Content: Content encapsulated UI layout and copy for easier understanding. By compartmentalizing content into chunks that were related or within context, we divvied up the content into buckets for breathability and digestibility.
Policy copy needs to be parsed out into legal and laymans terms. The Copy UI requires considerable spacing for legibility and breathability with Anchor links to the top of the page and return links for longer pages.
Contextual topics need to be considered for Top References regarding the inquiry so that the user has insight to other questions surrounding the content that may be in relation to the topic at hand. This gives the user a way to navigate through
browse - reducing cognitive load.
UI is different color for users profile vs the global information and policy.
Concise Labeling: Label copy is concise and translates across the globe through fuzzy matching. This reduces clicks to get to an answer. When user types in "Maternity leave" it will redirect the user to 'Parental Leave' in the USA, for France and other countries labelling will defer to the most commonly used terms for parental or maternity leave (where leave policies are offered). Labels UI will be clearly marked through spacing and colors that will be the same across global HR platforms.
All duplicates are removed and each heading will only be served up one per country/region.
Global Language Support: The site needs to translate into all the languages used for conducting business at the company.
Disability Options: There is an option to dial-in or get called instead of navigating a site, the number should be accessible on the footer of all pages.
Automatic User ID Identification: The site understands a user and their region, role and service line and serves them up the most commonly asked questions as 'trending help topics'.
Self Profile Management: Users should be able to change the information about their personal details and updates without having to go through an HR person directly. Notifications: Giving users insight to changes, updates and processes.
Other Features
Ability to apply In-line without the need for HR Intervention: Users should be able to apply for programs on the site without just getting the information. They should be able to access the information, get the application and apply on applications within the platform.
Accessing Chat from Every Place: Chat is accessible from the Spoke and the header of the Hub. When a user clicks on the Chat with HR Help, they are automatically taken to a new tab with HR.Help and the Chat bar open.
Branding
Naming the product
Naming the platform and creating the visual representation of the platform was created towards the end of the project when we were figuring out how we could create a simple and direct name for the product that wasn’t masked by a “cute, clever or modern” name - but instead cut straight to the point of what the platform is and what it’s there for.
"Based on reactions to labelling and buttons, people would prefer something straightforward, easy to remember over something cute, clever or 'modern'." - Research
The workshop We created workshop with 9 of our key product stakeholders to come up with a name before navigating to the visual design of the branding from that point onward.
Step 1: Aligning on keywords that describe the personality and feel of the service before we attempt to name it.
(Write 3 words or phrases that you feel encompass the service offering we have created - then discuss it with the group)
Step 2: Spend a few minutes coming up with potential names for the service.
(Write names you think would best encapsulate the service based off the principals we discussed previously)
Step 3: Each person gets 3 votes to vote on the 3 names you like the most.
Step 4: Consolidate the top 5 to 6 top voted names and discuss each one out loud.
Step 5: Vote on the best name
🏆 The Winner: HR. Help
Branding & Identity
We took the Words from Step 1 to do a search for images that aligned with the core words that we knew would align with the brand such as “Ask”, “Concierge”, “Guide”, “Service”, ‘Help” and “Team”.
We were served up with a plentitude of images but we wanted to keep the image simple and clean - decipherable and articulated the product in one glance.
We landed on a chat-like icon with a human silhouette and heart - indicating care and help, a human experience and the ability to chat (quick intervention with a live agent).
Branding & Naming conclusion
Although the exercises ran for a short period of time, the stakeholders opted for not testing at the time of the launch as we were testing the capabilities of the platform more than focusing on the branding of the product.
The company aimed to continue to explore the branding but were fixed on the name of the product for its launch.
Conclusion
Wrapping up our efforts
Building the platform was a pleasant challenge and I truly appreciated the research portion of digging
into the users' conflicts and frustrations.
With projects like these, learning about the users through testing and analysis the qualitative and quantitative results gave the UX team a very clear outline of the needs that were to be met.
The responsibility to delivering this dynamic product was a joy.
The outdated design system wasn’t a challenge in this product as we we given more autonomy than usual. Overall, once we built out all the screen and finalized the documentation, the project was given to development to be built out and then go-live.
