Research

I used three sources for research purposes, anonymous usage data from the “Marketing Automation” application, which is the existing marketing automation product for iCIMS customers, user interviews with current MA users as well as current CRM users alongside competitor research focusing on Mailchimp & Hubsopt. Its worth noting that this research was carried out for a holistic review of the entire Recruitment Marketing / Sourcing user experience. 

Competitor research

The competitor research was carried out to identify features and patterns and not focused on branding. 

Marketing automation findings (Quantitative)

My objective in examining this data was to get an understanding of current user behaviors. What I discovered was;

• Out of the 20+ automation nodes available, only 2 were ever used Send Email (Action) & Check for email open (Condition).
• Almost all automations were setup to send a single email with very few utilising the automated nature of the product
• Engagement scoring was different for every customer, indeed for every user, meaning no consistency of scoring across MA customers and even within a single clients setup.

USER INTERVIEWS (QUALITATIVE)

I, along with the product team conducted interviews with 10 marketing automation customers, focusing on their experiences with the application, particularly their usage of the workflow builder and engagement scoring. The interviews revealed key usability issues aligned with several of Jakob Nielsen’s 10 usability heuristics:

Visibility of System Status: One customer mentioned, “I find the workflow builder intimidating, just worried I’ll break something or send the wrong email to 10k people.” This highlights the lack of clear system feedback during critical moments, leaving users unsure about the impact of their actions.

Recognition Rather than Recall: “I just use it to send newsletters, I know it does more and it was explained, but the person who understood it best has left.” This speaks to the difficulty users face when they need to rely on memory instead of intuitive cues in the system. The platform fails to make the most essential features easily visible or accessible.

Error Prevention: “Engagement scores seem useful, but they don’t seem to ever go down, how can someone be constantly engaged?” This suggests that the design isn’t providing enough checks or clarity around how engagement is calculated, leading to confusion and potential errors in interpretation.

Flexibility and Efficiency of Use: “It constantly crashes when the journey gets too complex.” This comment demonstrates a lack of flexibility, as the tool can’t handle more advanced workflows without crashing, limiting users’ ability to scale their processes.

Help and Documentation: “I only ever duplicate, never start new. I don’t really know how to use it TBH.” This reinforces the need for better in-context help and guidance, as users don’t have the necessary documentation or help within the app to confidently explore features beyond their immediate knowledge.

NOTABLE TRENDS WITH THE INTERVIEWEES WERE:

Lack of Basic Skills: Many users lacked recruitment marketing knowledge, highlighting the Match Between System and the Real World heuristic—there was a disconnect between user terminology and the app’s technical language.
Engagement Scoring Confusion: Users didn’t understand how to properly use engagement scoring, reflecting an issue with Visibility of System Status—the app doesn’t adequately inform users about the scoring mechanism.
Limited Automation Usage: Users mainly sent job-specific promotional emails and newsletters, showing that the automation capabilities were out of their grasp. This indicates a Consistency and Standards issue, where users expect similar behavior across tools but don’t experience it here.
Knowledge Transfer Gaps: When the trained individual left, app usage declined due to a lack of Help and Documentation in the application.
CSM Challenges: The CSMs lacked marketing expertise, and therefore couldn’t assist clients in setting up significant automations, pointing to a gap in User Control and Freedom where users need more guidance and flexibility to solve their own issues.

Desires

• Templates 
• In app knowledge and strategies around recruitment marketing.
• Simplified UI for editing workflows.
• Less admin and setup around engagement tracking.

SYNTHESIS

After analyzing the data, I synthesised the findings into manageable reference points, beginning with user personas and completing the contextualising with empathy maps. Contextualising the target audience’s experiences with the current workflow builder will help meet their needs and goals and allows me to mould a palpable representational persona. Formed strictly from my research and interview quotes, I synthesised a user persona applying reference tools like Claritas Prizm, and Myers Briggs.

Empathy Map

After analyzing the data, I synthesized the findings into manageable reference points, beginning with an empathy map. Created from interviews, secondary research and observational analysis, the empathy map helped contextualize future users’ encounters with the Workflow Builder & Engagement Scoring features in CXM.

User persona & story

Contextualising the target audience’s brushes with sourcing and how a potential technology could meet their needs and goals allows me to mould a palpable representational persona. Formed strictly from my research and interview quotes, I synthesized a user persona applying the Myers-Briggs Framework.

Goals & Features

Product Feature Priorities

• Prioritize an intuitive, user-friendly interface that allows recruiters and sourcers to easily build, customize, and manage automated workflows. This should include access to pre-made templates.

• Ensure smooth integration with existing CRM and ATS platforms to minimize disruptions to users’ current workflows and enable seamless data sharing. This integration will be crucial for recruiters who rely on these systems for candidate management.

• Develop robust reporting features that provide real-time insights into campaign performance, allowing users to track email opens, clicks, and overall candidate engagement. Easy-to-access and exportable data is essential for making informed decisions.

• Focus on automating repetitive tasks like follow-up emails and lead nurturing, while ensuring that the system allows for personalized candidate interactions. Maintaining a balance between efficiency and a human touch is key to improving candidate engagement.

• Implement an intuitive engagement scoring system that automatically tracks and scores candidate interactions, such as email opens, link clicks, and form submissions. The scoring should be easy to understand and manage, with options for users to adjust criteria based on specific recruitment needs. The system must allow recruiters to identify high-potential candidates quickly, without requiring in-depth knowledge of marketing automation. Simplifying engagement score management at a global or workflow level will help increase user adoption and reduce fears around misconfiguring campaigns.

User persona & story

UI Designs 

prototype & testing

After an initial wireframes were created, we quickly moved to a high-fidelity prototype. Testing was carried out with a selected group.

‍

usability tests using the hi-fidelity prototype of CXM were conducted.

There were seven tasks created to test the usability and critical objectives of the feature including,

• Creating a new workflow
• Editing an existing draft workflow
• Using a workflow template
• Addressing errors within a workflow
• Activating a workflow
• Reviewing workflow analytics

The test goals included the following:

• Identify any critical errors or incomplete task flows within the app
• Evaluate if the user comes across challenges with navigation or design
• Understand successful task flows and identify helpful UI features
• Assess to see that relevant information is in the right place at the right time
• Can the user setup and activate a workflow
• Can the user edit an existing draft workflow

Design iterations

Original design, the goal was to improve this design without sacrificing any functionality and taking into consideration all research.

first iteration

I had seen grid based systems like the above within role playing games. I thought the grid aspect might make the experience more easily digestible and that by decoupling the Yes/No aspect of the decision nodes that it would make it easier to design a simple positive candidate flow. However after having to explain the concept in-depth to my peers and to users of customer xero, I realised it was overly complex, not mobile friendly and very confusing. 

second iteration

I had seen grid based systems like the above within role playing games. I thought the grid aspect might make the experience more easily digestible and that by decoupling the Yes/No aspect of the decision nodes that it would make it easier to design a simple positive candidate flow. However after having to explain the concept in-depth to my peers and to users of customer xero, I realised it was overly complex, not mobile friendly and very confusing. 

Third iteration

Expanded out the view for the horizontal scroll, this is first version we started to integrate the visual idea of funnels into the builder, as well as keeping context specific decisions grouped under associated actions. I think this helped alleviate some of the issues from the previous iteration by allowing for much more space for more complex flows, but still didn’t meet our criteria for readability and explain-ability for users.

Fourth iteration

Cleaned up funnel view, we’ve tried to correct for a lot of wasted space on the previous iteration by removing the unnecessary pipeline entry marker, as well as experimenting with colour markers instead of arrows. This helped to add in being mobile friendly as we could use a carousel style navigation to swipe through on mobile. From initial reviews however, the layout of nodes and introduction of function style nodes (combined decisions) was going to be too confusing for new users.

Final iteration

Bringing back the arrows while retaining colour tagging has helped bring some readability back to the flow while still accounting for mobile navigation. We also fleshed out the goal funnel to become part of the main campaign flow, we think this will help beginner users start to understand what they want to get out of a campaign, as well as create great opportunities for analytical data for a dashboard.

Design comparison

Both designs represent the same campaign properties and outcomes.

Results of testing

Of the six tasks, all task flows were completed sufficiently and with expediency by all participants. All participants were impressed with the visual design, design patterns and the second-hand nature of the navigation.

Each stated that this application would meet their individual needs and was successful in reaching their end goal.

There were many valid and interesting suggestions:

• Add language variations on templates.
• Be able to send SMS messages from the workflow builder
• Be able to duplicate workflows along with the content.
• Be able to fully test workflows.