top of page

UX Case Study:


Browse multiple ad formats on a single platform

Reliable Statistics aims to D.I.Y the media planning process. In its first phase towards an MVP, we researched who the target user might be and designed around this user.

Background on media planning

Traditionally media planning is outsourced to media planning agencies. Media planners help determine the best media channels most suited to the target audience of their client’s business. They create a plan based on their client’s budget, objectives, creatives and with tools and data such as Nielsen’s Media Index or Comscore. Within their own agencies, they would have data of their past campaigns which may further aid them in targeting the right channels.

Large scale businesses may also hire marketers with media planning experience to in-house their media strategies. Small to medium businesses tend to turn to social media and online ads where advertising tend to be much cheaper than traditional formats of advertising.

Finding target users

Based on our conversations with the client, we determined that small business owners, media planners and marketers would be the main users of the platform. We then conducted 9 interviews across these profiles. Our key insights can be summarised with 3 personas we created whom we identified as possible users of the platform.


Between the primary and secondary personas

The marketer, Mandy was identified to be the primary persona as she would have the minimum budget to explore more types of channel. AdList would be useful for her as she would not have to research each media channel individually, allowing her to D.I.Y a media plan. She may also have more data about her target audience which would make utilising Adlist in its initial phases easier.


A small business owner like Teck, may try out AdList to find interesting alternatives of channels to advertise. He may find it difficult to input accurate information about his target audience as he hasn’t researched his own clients. He may find that he is only able to afford the usual channels that he is already using and not find the platform useful. He expects the platform to address his pain points of managing channels before he would find it useful.


Yennifer may find AdList slightly useful for its aggregated information on various channels for her, as currently getting rate cards is still a rather manual process. Though with AdList in its initial phase, it would not have the enough metrics and information (such as available ad slots), she needs to calculate and create a detailed media plan. It would also lack the features that she needs to directly buy ads or analyse results from ad buys.

My role

This project was done in a team of four. I was one of the lead interviewers, and contributed towards synthesising, analysing and strategising. The discussion guide for interviews and usability tests were formulated by another member. The sketch prototyping stage was a group effort. I helped create key pages to the mid-fi wireframes. I also collated the designs and created the interactive prototype.

Clients’ brief

Reliable statistics (RS) wishes to design a user interface for their web platform.

In this first phase of the platform:

  • Users can research media channels, create a list and contact media owners.

  • They are unable to buy ads, track and monitor their ads.

In our stakeholder interviews, RS lacks a primary user to focus the product on. Thus we propose user research through interviews to base designs on. The client also had some quandries about his product that we hope user research may shed light on as well.


UI and interactivity considerations

We studied aggregator platforms such as Go bear and Sky scanner to look at how products are searched and compared. We also wanted to reference how users might input parameters as well. Sites such as Expedia and Amazon helped us understand usual practices of saving a product to a list other than in a shopping cart.


Our proposal for the platform is to follow known conventions in aggregator sites whereby users first inputs targeting audience’s profile, a list of recommendations then loads. Users may then edit the profile or use filters to browse recommended products.


The task flow of browsing and saving products to a list on the platform is as follows:


Most useful information for comparing media products

As per one of our client's quandaries, “What information would be helpful to the user to evaluate the effectiveness of his choice of media products?”


Base on our interviews and secondary research on other advertising platforms such as Google and Facebook ads, users would want information on an advertising channel, customised to their target audience.  Based on their target audience, they would like to know what is the channel's projected reach. We also thought an indication of how close the channel’s audience match the the user’s target audience profile might be useful. We have named this “match score”.

Sketch wireframes

We sketched our ideas and consulted one of our interviewees, a media planner.  Based on her feedback about her expectations of the data she would need, the product in its current phase, would make her a secondary user.


We also used these rough sketches to run through with our client (a developer himself) to check if our ideas of projected reach and match score was feasible.


High fidelity prototype

We did not think however the rough sketches had a high enough fidelity to test for content. We then proceeded to is rather high fidelity prototype as content had to be rather detailed for users to make sense of different kind of products.  The interactive prototype may be access here:


Usability tests

2 rounds of tests with 3 users each were done. 1 iteration was done after the first round. The test’s scenario was that of an account manager of a small creative agency with little experince in media planning and decided to explore AdList.  The test consisted of questions to get users’ feedback about the platform as well as tasks based on the scenario for the user to complete.

Summary of test results over 2 iterations

Overall, by the second iteration, users felt that the platform was quite straightforward to use. All users correctly assumed what match score, potential reach and minimum spent meant.

Some content were unclear e.g.:

  • Ad lists were initially called Adset and it was confusing to users. 

  • Users were not sure the difference between interests and lifestyles were at step 1 of inputting their target audience.

  • Price of products were initially indicated by dollar signs but it was not clear enough so minimum spent of a product was used instead.  

Issues from user interviews remain:

  • Getting information of consumption habits of target audience

  • Not being able to buy ads from the platform

More help for the users with formats of products:

  • In the case of newspapers which has many possible formats, it might be useful for users to be able to calculate prices based on the format they choose which later gets reflected in the ad list page. 

Next phases

From the feedback we got from interviews and the usability tests, we prioritized certain features to be added for future phases.

Phase 2

Product Details Page

Price calculator (rate cards, digital buys etc)

  • Ad format filter

  • Images reflecting format as user selects formats/ specs

  • Option to add to list user’s selected format


Product Results Page


  • Package deals for advertising


Listing Page

Scheduling for media time slot

  • manual input of required advertising periods

Reflect available dates from media owners

Phase 3 and beyond

Home Page

Aid users in finding out more about their target audience profile possibly through

  • Using platform to survey customers

  • Other existing data


List Page

  • Ability to purchase the media plan in the platform

  • Auto scheduling for media time slot

  • Budget Allocation

  • Analytics (Advertising goal input at Step 1/ home page)


Media planning was an area that was relatively new to me despite working at a communications department in government. Many follow-up questions were needed during interviews to clarify my understanding of the processes in the industry. My interviews were scheduled to be 30 minutes long but often overran to an hour or so. We are grateful to all our interviewees spending their time with us to help us understand their work better.

bottom of page