CIAL Dun&Bradstreet
Streamlined report request system: Enhancing efficiency in Report Ordering
CIAL Dun&Bradstreet | User Experience Designer | 2022 | B2B
Led the redesign of the report request system, streamlining processes to enhance efficiency in report ordering. Implemented improvements in language clarity and standardized procedures, ultimately contributing to a more effective and user-friendly system. This case study will cover the process of research, user testing and redesign of the feature.
Context
CIAL DUN & BRADSTREET
CIAL Dun & Bradstreet helps companies from Latin America to make better and informed business decisions. For long, its main product was different types of reports. In the past years, the company has been offering an interconnected system (C360) to not only deliver reports but for clients to manage, research, validate and analyze their credit and supply requests.
REPORTS
Reports are arranged based on requests from companies seeking information about other businesses. Each report varies in content, serving a specific purpose. These reports examine the financial details of the investigated company, its partners, investors, commercial standing, credit data, and more.
In certain scenarios, CIAL's team contacts the target company to request the necessary information. However, there is a possibility of denial. In such cases, the report is still delivered but with incomplete information. In other cases, if CIAL already has the necessary data, the report is produced and delivered within minutes.
This practice of getting reports is commonplace in the business world, serving as a vital tool to assess the level of risk associated with engaging with a particular company. Businesses utilize these reports to make informed decisions, determining whether a company presents a high or low risk in potential business dealings.
GOAL
Enhance the clarity of the report ordering process for our customers, focusing not only on language but also on standardizing how to request different types of reports. Additionally, strive to increase the number of reports that can be promptly delivered (AFF - Answer From File).
MY ROLE
In this project, I was the person responsible for leading discovery, ideation, validation and facilitating data analysis workshops. I worked closely with the PM responsible for this project.
Challenge
The Customer Support team received complaints from C360 users regarding discrepancies between the delivered report orders and their expectations regarding the last update dates. The confusion was coming from the dialogue boxes presented during the 'Ordering a report' process, which failed to adequately explain the procedures for each type of report scenario. This issue resulted in an increase in requests for reinvestigations, putting a strain on CIAL's Operations team.
FIRST THINGS FIRST: IS THERE MORE TO THAT?
The first thing I did was look for data we could possibly already have at the company. I talked with the Customer Support team, Operations team, Business team and the Sales team. My goal was to get a larger view of what users were saying and in which context they were saying. Also, I wanted to understand how they were dealing with the system.
The conversations were insightful once they brought several other problems and complaints users had related to the specific flow. Our main assumption was that communication played a big role in confusing users because they weren't understanding the meaning of the messages. The new information broaden up the issues we had about the problem:
-
The report production flow wasn't clear enough to make users trust the process;
-
The non-optional preliminary report was confusing to users, specially because we would send two emails with different results but withe the same subject title and same message.
-
The fields "Reference" and "Comment" are completely different things for the internal teams but they were seen as the same for users, causing confusion and important information not getting to the operation team.
POSSIBLE SCENARIOS MAPPED
#1
All information updated
Information updated including financials in the last 12 months.
#2
Recent investigation but missing information
Recently updated but some information is missing. Usually the financials aren't available.
#3
Old information
Last update was 13 months ago or more.
#4
No information
The company isn't in the database and the Ops team will then investigate the company and create its D-U-N-S Number to have it registered.
How it was before
When I tested the system, the first thing that caught my attention was the overall feeling of not knowing what to expect. My first surprise was in the case when we have all information updated, I clicked on the main button and the order was sent and I was expecting the system to behave the same way as in the other cases. Buttons look the same, have the same name so they should behave the same way and they didn’t. Here I found two problems:
1
Possibility of control and freedom NOT available. We weren't giving a chance to the person to correct any possible mistake. If the person ordered a report by mistake, one would need to contact customer support to cancel it.
2
Inconsistencies in the process and in the copy. We had one button that looked the same, had the same copy but would behave differently, depending on the chosen report without any clue. It was confusing and had the potential to increase the cognitive load and frustration within users. Also, the order behavior was nothing like the industry conventions that you at least review your orden before you place it.
BEFORE - PLACING AN ORDER
1
Actions had the same button with the same copy but would behave differently depending on the scenario the company is in.
2
The title gives no clue of which report the person is ordering in which company. The person might remember, but we can't rely on that.
3
Recent investigation but missing information scenario. The text is long and doesn't draw attention to what will happen and the possible outcomes.
4
A comment means any information that could be helpful to the Ops team. It can be a contact number or email of someone that works in the investigated company.
A reference is for the user's personal use only. What is added there, the Ops team does not have access to. However, this caused confusion, once users didn't know and would often add important information for the Ops team there.
5
Old information scenario. In here, the last update on this company was at least 13 months ago. The person needs to choose if CIAL can disclosure the company's name requesting the information or not. It doesn't explain why or what will happen though.
6
All information updated scenario. When the person clicks on Order report button and all the information is already updated, the order is placed right away and the confirmation is a toaster and the person can also make sure it was ordered in the page Reports.
7
No information scenario. Two different buttons but having the same weight. After the person chooses to order a report, there's a form to fill in with any information the person has on the company that should be investigated.
Proposed solution
To achieve a more consistent flow, I focused on clarifying the process through improved communication. I worked with the PM of the project to find a different tone that we were using before.
By adjusting the flows to have the same behavior regardless of the scenario, users could have more autonomy and control. As an initial step, I renamed the 'Order' buttons to terms that more accurately conveyed the ongoing nature of the process, I also added a text tailored for each of the 4 scenarios mapped to explain the context and the possibilities the person have available. After the order is placed, I designed an order summary to act as a confirmation screen.
AFTER - ORDERING A REPORT
1
Select report instead of Order to give a sense that there's more to the process regardless of which scenario the company falls in.
2
Cost was added so the person knows how many units each report costs. This was a test to see if the users would understand this information.
3
Indication of the company selected by the person.
6
I modified the label of the Reference field to "Your reference" to enhance clarity. In addition, I also added a help text to explain the purpose of it.
9
A summary of the order with all the information previously selected and informed.
4
Addition of information that was selected in the previous page. The type of report, the language, the cost of the selected report, and the information on when the report and the financials were updated.
7
I added the possibility to go to the previous step without cancelling everything.
10
Details regarding the cost and remaining balance of the units bought by the company.
5
I added a text for each scenario mapped. The aim is to clearly inform the person about any required actions in a direct manner.
8
A dedicated confirmation screen to ensure clarity and not rely solely on a toast.
11
The text for the second scenario, where the financial information is missing. A new investigation can be placed, but there's no guarantee that the missing information would be available. This possibility wasn't clear enough to users before and is meant to provide transparency to users about the process.
12
In occasions where companies might not require the missing information immediately or can proceed with decision-making regardless, the option to request a report remains available.
13
When the companies want CIAL to try again, I added a section for the user to indicate whether CIAL is permitted to disclose their company name. Additionally, the comment field was added there with a help text to visually differentiate itself from the "Your reference" field.
14
For the third scenario, where information was last updated more than 13 months ago. This update intended to clearly outline the current status, explore available options, and establish realistic expectations.
15
Same behavior as previously explained: Users are given the option to authorize the disclosure of their company name. Similarly, a comment field is provided, designed to offer additional insights
16
In the scenario with no information available, I adjusted the button hierarchy to emphasize the option of ordering a report, drawing the attention to that action.
17
Form with all the fields required for CIAL to initiate an investigation on a company not yet present in CIAL's database. I couldn't change the list of fields, so I adjusted the layout of the Comment and Reference sections to ensure consistency with the formatting used in other scenarios
Results and key takeaways
This section provides a concise overview of the process, a comprehensive and detailed description is available in the next section.
Prior to usability testing, I presented our preliminary findings and concepts to the design team, the PM lead, and a C-level executive. This step was crucial for ensuring alignment on ideas, understanding system and business constraints, and making informed decisions.The feature of showing balance and cost information were not technically feasible at the time, I chose to keep the idea in the testing phase to gather feedback on the concept.
The user testing involved five participants from five distinct companies, including two from Brazil and three from Mexico. My responsibilities included creating the research plan, coordinating with the Sales team, and training the Mexico-based Product Manager to conduct the usability testing. Additionally, I organized and led a data analysis workshop with the team to interpret our findings. Following this analysis, I made adjustments to the interface to reflect the insights gained from the user feedback.
ISSUES FOUND THROUGH THE USABILITY TESTING
1
A button in the listing page that wasn't in our radar was problematic to users because they wouldn't understand its meaning once it was just a random icon.
2
Confusion about the term "Decisions" in some places of the interface. 'Decisions' refers to a functionality that enables companies to assess potential clients according to criteria established by the company. It was a new feature and they didn't know what it meant, thus causing confusion.
3
Despite the help text, the 'Your Reference' field remained unclear to users. Feedback from testing indicated that this feature was of minimal utility, as the participating companies were not accustomed to categorizing their report requests.
4
The concepts of 'units' and 'balance' were not easily understood, likely because the idea of reports costing a certain number of units was unfamiliar to them. Typically, companies purchase a specific type of report in a fixed quantity annually. To successfully integrate this concept, it would necessitate adjustments in the sales strategy and how the services are marketed.
5
There was confusion regarding the scenario where CIAL might not have a specific company in its database. This situation led to a perception of decreased credibility for CIAL among users, if they ever encounter such a scenario in actual usage.
6
Confusion by the meaning and the purpose of a preliminary report. They questioned the rationale for receiving a report containing outdated information before the completion of a new, updated report.
We could address all the issues in the new interface except for the last one because of technical reasons about the rules of the preliminary report. At the time, the agreement was that a new round of research and ideation would happen to address the issue of preliminary report.
A fully detailed approach
RESEARCH GOALS
I wrote a research plan aligning the goals, methodology, hypothesis, metrics tasks and other topics with stakeholders. The goals were:
-
Evaluate if the different scenarios and their outcomes are understandable to users;
-
Understand the value users give to the preliminary report and the value they perceive in requesting an investigation;
-
Find out in which cases users would consider getting a report with old information,
-
Discover people’s thoughts about the balance and cost and how they feel about the overall experience.
For this research, we wouldn't get all the answers we wanted from the usability test, so I added a qualitative interview to get people's thoughts on the topics we wanted to know something.
Five users were selected, three from Brazil and two from Mexico. I prepared the prototype and conducted the research with the users from Brazil and the PM responsible for this project with users from Mexico because of language issues. The tasks were:
-
Task 1: The company Brasil Modelos (DUNS: 914669031) requested credit to your company and you wish to evaluate some of their data. Brasil Modelos is a recurring client and, as part of the current policy, you need a simpler report to start the decision. Using our platform, request the report you think is more appropriate. Success: Be able to request a report from the available list in the first try without hesitation and understand the outcomes of each scenario.
-
Task 2: The same company, Brasil Modelos (DUNS: 914669031) requested a higher amount of credit. Even though they are a recurring client, as part of the current policy, you need to have access to several data about the company to make your decision. Using our platform, request the report you think is more appropriate. Success: Be able to request a report from the available list in the first try without hesitation and understand the outcomes of each scenario
-
Task 3 – A company called Aviagaca LTDA requested a considerable amount in credit to your company. They are a new client, and you don’t have any previous information. Using our system, find and request a report that you think is more appropriate. Success: Be able to request an investigation in the first try without hesitation and understand the outcomes of each scenario.
DATA ANALYSIS
For the analysis, I gathered the people within the company that participated as observers in the research and the design team lead to analyse the data together. All the material had to be translated to English first. The workshop had these steps:
-
Selection of important feedbacks by participants
-
Affinity grouping analysis
-
Dot voting
I split the group in two and each one was responsible for selecting feedbacks from one user at a time until we finished selecting. After each round, we would present, give some time to the other group to see if something was missing and then move on to the next user. We had boards to add the post its divided by steps (each step or variation of the flow was a step).
Then, I grouped all the post its together and split the group into two to analyse each step. The process was the same as in the previous steps.
For the dot votation, each person had 5 dots to vote on which topics they thought were most problematic and should be fixed right away. We presented our choices to each other and explained why. We reached an agreement on the priorities and feasibilities at the time to adjust the process.
Key takeaways
Leading the first research initiative from the product team within the company marked a significant milestone, both for the company and for me personally. It wasn't the first time in that position for me, but it was a complete different and unique challenge. Compared to previous experiences in a B2B company setting, I found harder to contact customers. Not exactly because they refused to do so, but because the communication was almost fully centralized around the Sales team.
So my first challenge was to convince and align with them on the importance of the research, with presentations, showing the research plan, the prototype and doing a beta testing with them. It was an interesting process that led me to identify their concerns, structure my speech and then delve into each one to alleviate their concerns. When I finally conquered their trust, they helped me greatly in the recruitment process which was smooth and had zero dropouts.
The process of collectively analyze the data brought the team together to agree into the best solution considering feedbacks from users and technical constraints. Unfortunately, I was unable to keep up with the implementation of the changes, as my role shifted to a new position within the company-funded startup, which required my full attention.