MGID Ads (Part 2: Design Implementation & Results)

Our central objective in redesigning the Home page was to simplify user understanding without compromising on existing functionality and successful integration of new features.

Updated “Campaign performance” page interface

To enhance page clarity, the first step was removing the large Advertiser/Publisher switch and relocating it to the user profile.

Updated user profile dropdown with Advertiser/Publisher switch

Following that, I focused on improving the Campaign performance page's functionality with the following enhancements:

• Users now have the option to collapse the graph, based on research indicating their preference for focusing on the data in the table.

• Collaborating with a product owner, I enhanced the graph's functionality, making it dynamic and enabling users to select preferred metrics for comparison.

• In pursuit of a smoother user-flow,  I restructured the actions for each campaign in the table, concealing secondary actions under hover and relocating "Statistics," "Optimization," and "Ads" to the campaign page.

UX research found most users, especially affiliates, create multiple variations for each ad during a single session by combining different text and image options for optimal conversions.

The previous interface slowed this process, requiring users to redo the ad creation flow for each image-text pair. In order to streamline the ad creation process we've introduced variative ad creation. Now, a range of text-image pairings are auto-generated, substantially enhancing efficiency and fostering smoother workflow.

Updated Ads creation interface with ads variations

The research findings have also revealed additional challenges that users come across during the process of creating ads:

• Ambiguities in certain features (Category Selection, Teaser Rich, Read Metadata)
• Unclear image focal point configurations and size adjustments.
• Absence of a distinct preview for various image formats.

Significant usability challenges within the previous Ads creation interface.

In order to address these issues, I directed my efforts towards enhancing the flow of ad creation by implementing the following improvements:

1. To streamline ad creation, we relocated CPC selection to the “Limits & Schedule” in campaign level. This allows users to customize it for individual ads in the ad interface.

2. We moved the focal point settings to a dedicated screen and introduced a preview of all available formats for the chosen teaser on the ad edit page. This empowers users to instantly view how the teaser appears on the content page and easily fine-tune the image center for various formats.

3. We implemented time-saving functionalities such as Title Recommendation and AI-generated images, and eliminated redundant fields effectively streamlining the Ads creation process.

My objective was to streamline the campaign creation process for users. Research revealed notable user pain points, including complex form fields, confusing errors and the inability to save drafts in case of unexpected issues.

To address these, I initially focused on major usability concerns:

• Simplified the campaign creation into three distinct steps, grouped settings by type, and incorporated a stepper with step names for better progress tracking.
• Introduced synchronous progress saving and inline input validation to prevent data loss due to connectivity issues or accidental browser tab closures.
• Enhanced user comprehension by relocating errors from the header to relevant fields and refining error messages with the assistance of a UX writer.

Based on user interview feedback, particularly concerning targeting complexities, I undertook the following measures:

• Overhauled the layout for easy scanning and simplified targeting settings, making the process more intuitive and straightforward.

• Enhanced user guidance by offering detailed explanations and replacing ambiguous icons with clear text labels for the include/exclude options.

Furthermore, I fine-tuned the layout and removed any unnecessary elements, achieving this through the subsequent actions:

• Eliminated redundant fields, such as the category field, by automating data collection.
• Segregated traffic insights to a separate page, recognizing that users rarely needed this information during campaign creation.

I had quite a bunch of meetings with the product manager and backend developers, mainly because the way we improved the interface depended a lot on how the backend was set up.

Optimizing an advertisement campaign involves an ongoing process of closely tracking statistics, analyzing real-time data, and comparing it with the goals, then turning off non-performing sources and adjusting the CPC to more effective ones in order to bring the actual values ​​closer to the target value. As you can see, it is quite time-consuming and takes a lot of manual work.

To make life easier, MGID introduced rule-based optimization for advertisers. Automation that saves time and resources and also removes human mistakes and sometimes extra waiting time from people involved in the management process. Interestingly, though, only 3% of advertisers are actually using this type of optimization.

Subsequent interviews highlighted that the extensive tracking settings in the campaign creation process, necessary for rule creation, were acting as a barrier to user engagement with automation. As I had already reconstructed the tracking setup in the campaign creation flow, my aim was to enhance the user experience without extensive backend modifications. This was achieved through the following strategies:

• Improving the discoverability and refining the layout of automatic rules settings.
• Reconfiguring the column order based on insights gathered from interviews.
• Relocating goal settings to filters for a more streamlined experience.

In mid-April, we successfully launched our internal testing involving subject experts. Not only customers feedback was mostly positive, but nearly all participants smoothly navigated the user journey autonomously, without necessitating support.

For the internal testing phase, we meticulously selected 7 participants for both usability testing and usability interviews, allotting 1.5 to 2 hours per session. Our approach encompassed a blend of testing and interviews, serving two purposes:

• Task completion evaluation: We assigned tasks for independent execution, closely observing participants' actions and seeking their insights.
• Post-task experience assessment: We conducted personal interviews to gain deeper insights into participants' experiences following task completion.

Users think that the platform is more user-friendly, which increases the satisfaction of dashboard use. We improved many things, like launching ads after approval, clearer moderation feedback, the errors have become more humane, and they appear in the run time. According to users' feedback, we have improved our platform in following:

• Increased the efficiency of the platform and campaign management speed
• Reduced complexity, errors, problems, and pain points

At the same time, we have found additional insights and launched a series of product improvements based on research feedback. However, the project's scope extends beyond these accomplishments. The next stages include:

• Addressing identified issues.
• Conducting a second round of testing with external participants, our clients.
• Undertaking quantitative research.
• Launching a beta version of the updated platform.

During my involvement in the MGID Ads redesign, I've amassed pivotal insights that now underpin my design philosophy. Effective communication, spanning across teams and departments, stood as a central pillar of our achievements.