Natalia Golova's Portfolio

Document Helpdesk Redesign for an Identity Verification Platform

ROLE

Product Designer, UX Researcher

TIMELINE

May'23 - Jul'23

As part of a broader initiative to modernize the AuthDoc identity verification platform, the Document Helpdesk redesign focused on improving the internal tool used by Helpdesk employees. Document Helpdesk enables employees to manually assess documents that fail automated verification or require additional human verification for heightened security. In the legacy system, the Document Helpdesk interface was essentially a variation of the main user interface, leading to a cluttered and cumbersome experience. To address these issues, I aimed to enhance both usability and efficiency, aligning the tool with the organization’s goal of achieving a human verification speed of under 2 minutes for most cases.

Note: This project is based on a real-world engagement under a non-disclosure agreement (NDA). All sensitive data and proprietary information has been omitted or anonymized in accordance with the NDA terms.

Goal

To redesign the Document Helpdesk interface to streamline manual document verification, increase processing speed, and reduce cognitive load for Helpdesk employees. This would enable faster, more accurate assessments, helping the organization meet its goal of under 2 minutes for human verification in the majority of cases.

Outcome

achieved average time on task 1 minute 55 seconds in beta testing

User Research

Constraints

The Document Helpdesk redesign was subject to several key constraints that shaped the approach and design decisions:

Limited Timeframe

The redesign had to be completed within a tight, two-month window to align with the larger re-design timeline of the AuthDoc identity verification platform. This required a focused, agile approach to ensure delivery within the timeframe without sacrificing quality.

Small User Base

The Document Helpdesk is used by a specific group of fewer than 20 Helpdesk employees daily. This specialized audience allowed for a more customized design approach but also required a deep understanding of their unique workflows and pain points. The redesign needed to prioritize functionality that directly supports the high-stakes, accuracy-driven tasks these users handle.

No Retraining Period

Given the time limitations, there was no opportunity for comprehensive retraining. The interface therefore needed to be extremely intuitive and easy to navigate from day one. Every element was designed to align closely with user expectations and existing workflows to allow for a seamless transition without disrupting productivity.

Research Preparation

To conduct foundational research, I employed qualitative interviews and the Contextual Inquiry method. This approach allowed Helpdesk employees to demonstrate their document verification processes directly within the platform, enabling me to observe and measure the time spent on each verification task in real-time.

Following these observations, I organized a product outcome prioritization workshop with the key stakeholders. During this session, we aligned on a primary outcome and a relevant performance metric: keeping the average verification completion time under 2 minutes for at least 80% of cases. This outcome became the central focus of the redesign, driving decisions aimed at improving efficiency and streamlining the verification workflow.

Interview Process

One unique aspect of the Document Helpdesk redesign was that many of the platform’s users also held key stakeholder roles. As a result, I had the rare opportunity to engage a large portion of the user base directly during the research phase. This enabled me to conduct nearly comprehensive interviews with Document Helpdesk employees, providing an in-depth view of their day-to-day challenges, needs, and specific pain points with the existing system. At this stage I conducted 16 interviews in total.

However, the extensive input from nearly all users also introduced challenges. While I gained valuable insights, the feedback quickly became highly specific, as many users had their own individual preferences and workflows that weren’t necessarily generalizable across the team. Balancing this variety of perspectives required careful management to avoid focusing too narrowly on individual preferences that might not benefit the broader user group.

Setting the Scene

Document Queue and Document Details

Through the research phase, I identified two critical screens that significantly impacted the verification workflow and its overall speed: the Document Queue and the Document Details page.

Document Queue

The Document Queue serves as a comprehensive overview of all incoming documents and their current verification statuses. This screen allows Helpdesk employees to quickly prioritize tasks, assess document volumes, and monitor which items require immediate attention.

Document Details

The Document Details page provides an in-depth view of individual documents, displaying scanned images, Machine Readable Zone (MRZ) data, and the results of any automated checks. This screen enables employees to closely examine document details and identify any discrepancies or issues that may have prompted manual review.

Focusing on the usability and efficiency of these two screens became a primary objective of the redesign, as optimizing these touchpoints would have the greatest impact on achieving the goal of reducing task completion time. We set the goal of refining the structure and flow of these two pages.

Legacy Userflow

Problems

Actions are heavily concentrated within the Document Details screen, resulting in a high cognitive load for users during the verification process.

Documents are not differentiated by status; all documents are grouped together regardless of their verification stage, making prioritization difficult.

The system contains multiple outdated and deprecated document statuses, adding unnecessary complexity.

Document lookup is limited to document numbers only, requiring cumbersome workarounds to search by document bearer.

Proposed Queue Userflow

Solutions

Users can access documents at various stages of the verification process, improving workflow flexibility.

Verification status is now clearly visible, allowing users to see which documents are actively being processed.

Enhanced filtering options enable users to sort documents by multiple parameters, improving search efficiency.

Progressive filtering by last name and given names streamlines the search process for specific individuals.

Users can view the history of previous document edits, providing context and transparency for verification actions.

Ideation

To implement the improved user flow, I conducted multiple rounds of ideation and validation, using mid-fidelity prototypes to test concepts directly with Helpdesk users. I categorized user pain points through affinity mapping, grouping feedback by themes aligned with key outcome of faster document verification, and related outcomes, such as faster document retrieval and better verification status visibility.

The following sub-outcomes were identified as essential:

I then applied impact-effort analysis to prioritize solutions that would yield the greatest improvements in efficiency and ease of use. This approach helped identify the most impactful solutions to implement first, allowing for a phased introduction of additional functionality.

DOCUMENT QUEUE

Manage the Incoming Queue Effectively

v.1 Simple Queue

Older documents appear at the top, positioned just below high-priority documents.
Documents with a final status are removed from the queue immediately and can only be accessed through cold storage.
Filter by document number.
Progressive filtering: Surname → Given Names → Date of Birth.

v.2 Simple Queue + In Progress

In progress: some user has started editing the document.
Second opinion: some user has requested second opinion.
In progress documents get pushed to the end of the queue.

DOCUMENT QUEUE

Manage User Privacy Expectations

To address user privacy expectations, stakeholders introduced a requirement to separate data storage into hot and cold storage, with Helpdesk employees permitted to access only hot storage (temporary storage of the most recent documents). However, research indicated that access to older documents is essential for addressing user support requests, which predominantly occur within the first 48 hours after submission. After discussions with key stakeholders, we reached a compromise allowing Helpdesk employees to access recent documents while still maintaining privacy safeguards.

I designed a Queue with the following four tabs to accommodate this structure:
- New: Displays all newly submitted documents awaiting verification.
- In Progress: Shows documents currently undergoing the verification process.
- Last Reviewed: Lists documents that were reviewed within the past 48 hours.
- All Documents: Allows users to request access to specific documents through an administrator. The timeframe for document retention in this tab will be tested for optimal support.

DOCUMENT QUEUE

Manage Training

In discussions with key stakeholders, they emphasized the importance of systematic training for Helpdesk employees, citing several key purposes:

Studying Forged Documents: Reviewing detected forgeries can help employees recognize similar traits in future verifications.
Learning from Past Errors: Reflecting on missed details in past verifications provides valuable insights for improvement.

Solution

For training purposes, employees can use the "Last Reviewed" tab to access recent documents from the past 48 hours, or search within "All Documents" using a Status filter. Additionally, an option to anonymize documents and add them to the Reference Data—a separate application within the AuthDoc ecosystem—will be provided. This allows for long-term reference without compromising user privacy.

DOCUMENT QUEUE

Manage Ongoing Work

Problem

Helpdesk employees currently lack visibility into who is working on a document and what specific changes have been made to the original submission, aside from the most recent modification.

Research Insights

Employees occasionally need to collaborate with colleagues or re-check a document together. This requires access to a full history of modifications to analyze previous edits effectively.

Solution

Introduce a History section that displays all saved edits, including timestamps and the author of each edit, enabling a clear audit trail.

Add a Cross-check feature on the Document page, allowing employees to easily collaborate and review documents with colleagues when needed.

DOCUMENT QUEUE

Statuses and Notes

Most existing statuses are deprecated or outdated, creating confusion and inefficiency. To streamline workflows, new Helpdesk statuses were introduced:

New (Helpdesk)
OK
Bad Scan
Doubt
Unknown

Additionally, a Notes feature has been added to allow employees to attach relevant details or observations, improving document navigation and collaboration. Examples of notes might include any combination of comments on document quality, priority or user comments.

DOCUMENT QUEUE

Final Queue Layout

My findings determined by 3 rounds of usability testing with 8 participants have shaped the following final Queue layout.

DOCUMENT DETAILS

Common Issues

Research Insights:

There is a need to minimize repetitive actions, reduce the number of clicks required to access key features, and limit the screen space occupied by rarely used fields to streamline the verification process.