DNEG ORDINO
ADVANCING REFERENCE MEDIA ORGANIZATION IN FILM MAKING
Currently, in modern digital film production, after a long tiring work day on a movie set, the filming crew gathers massive gigabytes of data. This data becomes the secret sauce for creative and technical artists in the VFX team, helping them weave magic into the movies they create.
COPYRIGHTS @DNEG
SEPTEMBER 2023
In the bustling world of DNEG, sorting through massive data every day after shoot turns into a real time-drainer.
With wranglers relying on laptop file managers and fading memories after a long day, it becomes a bit of a puzzle. Additionally, each wrangler has their unique sorting method, leading to an assortment of folder structures. The DNEG Imaging team ends up untangling the mess before transferring the data for advanced data processing
in Jigsaw, a DNEG restricted tool that requires a DNEG standard folder structure for systematic management; introducing unnecessary redundancies in the workflow.
Capture asset and shot data on set
Capture slate information
Sort data using
file manager, then transfer to DNEG
Redundant data sorting; advanced data processing in Jigsaw by DNEG Imaging team
Data transferred to DNEG directory
Data viewed and utilized by creative teams for shot build
KICK-OFF MEETING
EARLY INSIGHTS
The initial discovery questions discussed during the kick-off meeting with the stakeholders (Imaging and Shoot team HODs) and the development team allowed me to understand and build empathy for the data wranglers in order to make better decisions, build a better product experience, and create problem-centric solutions.
Easy to carry portable machines (laptops) on set with cameras
Jigsaw, a complex, DNEG restricted data sorting tool, requires access to the internet
Non-DNEG employees in shoot teams do not have access to Jigsaw
Sets have poor connectivity, creating a need for offline tools
Time consuming data sorting using multiple file manager windows
Snippets from the current data sorting process (from top left) : Folder structure, data sorting using file manager, on-set crew member, Jigsaw interface
MY ROLE
END-TO-END UX
I led the end-to-end UX process from research to testing for DNEG ORDINO's desktop application since June 2023. I steered my efforts to evolve the development team and the product through research, facilitation, design; identifying and addressing the data wrangler's pain points related to session launch, media import, and sorting experience.
Uncover customer insights, ideation
I uncovered insights and translated concepts into features that address user's motivations, pain points and behaviours.
Define experience strategy
I created frameworks and prototypes to share the vision and design principles, to gain alignment and drive decisions.
Design execution and validation
I executed clickable prototypes for laptop using Figma and
validated with users to meet the acceptance criteria for MVP.
Oversight and coordination
I partnered with a team of developers, project manager and stakeholders to stay aligned with the business goals.
Leadership and raising UX maturity
I presented work to stakeholders to gain buy-ins for the MVP.
I introduced UX practices to the developers to raise UX maturity
THE CHALLENGE
EXPEDITE AND STREAMLINE
MEDIA SORTING
Our goal for the project was to improve the media sorting experience for the data wranglers in the shoot team. The original premise for the project was
clear: enhance the user interface through visual design. However, there were more significant challenges beyond visual aspects that the development team overlooked. The UX team's objective was to improve the user experience by diving into the core problem.
THE APPROACH
GOOD, FAST, CHEAP
Mirroring the industry's rapid pace where workflows evolve with every new show
and every show requires consistent data management, the code was reprocessed
from Jigsaw. This constant adaptation is seen as an asset, but the real challenge lied
in guiding engineering teams to help them experience the entirety of UX, to raise UX maturity in the organization within a short period of time before the launch of MVP.
The assumption was simple : The shoot team is a mix of DNEG employees and show based contractors. Leveraging existing infrastructure by extending Jigsaw's conceptual model that employees were already familiar with, also reduced the effort on training DNEG employees to use Ordino; making the execution faster and inexpensive.
This early architectural decision had a major impact on the quality of user experience we could create; limiting the solutions to conform to the architectural map.
THE PROCESS
DESIGN BY ACCRUAL
The development, product and project teams were involved from the initial phase of the project. With a limited UX maturity, UX became involved in the project subsequent to the completion of the first solution prototype by the development team.
The design thinking methodology underwent slight modification in accordance with the stage in which UX was engaged. Subsequently, I organized workshops to identify potential problem areas in Ordino beside supporting the team with visual design.
DEVELOP
EVALUATE
EMPATHIZE
DEFINE
IDEATE
PROTOTYPE
A snippet from an initial discovery workshop with stakeholders to understand the data structure in Jigsaw, to be adapted in Ordino.
METHODOLOGY
CHASING
AGILE-SCRUM
The combination of a fixed launch date for a show and
the aggressive scope created an intense environment
with timeline and co-ordination challenges.
Working backwards from the launch date for the MVP, we followed the Agile-Scrum methodology to complete a set of goals to be completed within a bi-weekly sprint;
with frequent planning and goal setting for the scrum team to stay within budgets.
I self studied the tool while simultaneously auditing the UI and did extensive usability tests with 5 participants to highlight and validate the key issues leading to a unintuitive tool and identifying factors contributing to loss of engagement.
EVALUATING THE INTERFACE
Self educated the information architecture and UI
Mapped task flows
and validated with the development teams
Audited the user interface against NN/g heuristics
Conducted usability test to collect qualitative and quantitative insights
THE DISCOVERY
A COMPLEX,
UNINTUITIVE INTERFACE
The initial build of Ordino posed a significant challenge for data wranglers due to an overwhelming interface, and a dictated way to organize data strictly within a predetermined structure, which proved less intuitive and constrained. While DNEG views this structure as essential to diminishing redundancies in the pipeline, it ultimately led to decreased engagement with Ordino.
RESEARCH INSIGHTS
OVERWHELMED, DISCOURAGED, CONFUSED USERS
I conducted extensive research labs with 6 on set shoot team members and data was analyzed and presented to the stakeholders and the engineering team.
My goal for testing the initial interface was to understand the challenges users were facing when sorting their data using Ordino, which led to minimal adoption and lower satisfaction ratings. Participants were selected across DNEG locations in North America, UK and India to reduce a western bias when developing tools at DNEG.
A poorly executed user interface led to frustration, confusion and wasted time among data wranglers on set using Ordino. This was due to not gathering real data and making assumptions about the users mental model and implementation of the solution without iterating and usability testing.
THE REDESIGN
A SEAMLESS MEDIA SORTING EXPERIENCE
In a complex VFX pipeline with crucial timelines, the revised user interface of Ordino saves time by making reference data sorting quick, and effortless and essential for users, while still following the DNEG standardized folder structure.
Ordino makes the journey simplified, understandable, actionable, and discoverable for the exhausted data wranglers after a long shoot day.
INTRODUCING SESSION MANAGEMENT
Ordino allows users to initiate and customize sessions, adjusting settings before entering the sorting mode. It also facilitates retrieval of saved sessions, ensuring no loss of data from ongoing work.
A streamlined workflow with clear, essential, and digestible tasks minimized clutter on the interface, to ease cognitive load.
ENHANCED MEDIA VIEWING
Prior to organizing data, the crucial step involves previewing media. The media viewer serves as a panel for exploring folder/s data, offering detailed metadata and contextual information. A user can effortlessly locate a slate image by navigating through the viewer.
DISCOVERABLE, HASSLE-FREE
DATA SORTING
The media viewer serves as the primary panel for organizing media into appropriate folders, improving discoverability through a user-friendly predefined folder structure.
This sorting process feels natural, reducing reliance on drag-and-drop,
and enhancing the overall interaction experience.
VISUAL RESTRUCTURE OF INTERFACE
User challenges drove a revision of critical user journeys in the Ordino workflow, influencing the reframing of visual structure. Proportional allocation of viewing and sorting zones was established on the user interface.
THE IMPACT
DATA SORTING
LEAPS 8-FOLD
It is still early days for Ordino, yet the follow-up usability tests exceeded our expectations. The redesign of the journeys and the interface has had a positive impact on the overall pipeline. The efficiency of the data sorting process improved by approximately 16 times than the current process, with a higher satisfaction rating.
THE FRAMEWORK
HOW WE GOT
THERE..
Th biggest challenge faced in the project was guiding the engineering team and moving the needle of UX in the organization while balancing research, experience strategy, design and testing within the crucial timeline of launching a minimum viable product (MVP).
Managing feedback was even more challenging as the team did not collect through appropriate usability testing and hence did not have a structured way of collecting and analyzing feedback to drive a design decision.
The following questions informed my strategy:
-
Who are we building the application for?
-
What contexts need to be considered?
-
What does their current data sorting experience look like?
-
How can the current build of Ordino be improved?
Ordino poorly empathized with users who were not the accurate reflections of users working on set.
The goal was to create design solutions that are grounded into the VFX pipeline, yet future-forward and within the constraints.
WORKING BACKWARDS
While the solution built by the development team was supposed to reduce the data sorting time for one slate data from 6 minutes to 1.5 minute, the users failed to complete the task successfully even after exploring for 4 minutes.
I invested the time to talk to the users, document on-set scenarios, feelings, their interaction with the interface, while cross-verifying why such a powerful solution did not work. Documenting this upfront and presenting before diving into the solutions was time consuming, but eventually helped the engineering team and stakeholders understand the gaps and pain points.
Understand challenges users faced on set and with Ordino during a usability study
Analyze qualitative and quantitative data collected from think aloud, note taking
Present the usability debt
to stakeholders, engineers
JOURNEY MAP
MAPPING THE EXPERIENCE
Understanding how users spent their time on set, sorting data in intervals or all together at the end of the day while going back home or in the hotel room, allowed me to think beyond just the data sorting experience.
Plotting the journeys allowed everyone on the team to understand how users should interact with the initial build of Ordino, their motivations, touch points, feelings, pain points, usability issues and a set of recommendations.
INSIGHTS
TO IDEAS
Through insights analysis, collaborative brainstorming sessions, idea explorations and design reviews conducted with the development teams, I explored solutions to solve critical, moderate and minor challenges that were highlighted in the usability lab analysis before the launch of the Minimum Viable Product (MVP).
IDEATION
OVERWHELMING TO INTUITIVE
The system needed to be instinctive for the wranglers to guide them through the entire journey in Ordino. The envisioned journey was a revision of the workflow, influencing the system architecture, that solved two critical challenges:
-
Splitting the journeys created a step-by-step, self guided process.
-
Session management allows wranglers to continue to work in intervals through a saved session, reducing clutter on the data sorting interface.
IDEATION
CLUTTERED
TO CONCISE
Introducing session management created a need for restructuring the UI, allowing the user to work with Ordino in phases. On creating a new session or opening a saved session, the interface primarily focused on the area for menu, tools and actions, input, sorting and output zones, thus influencing the information architecture of Ordino and improving the visual hierarchy.
IDEATION
HIDDEN TO DISCOVERABLE
Through the usability test, we identified how users currently sort their data. We learned that:
-
Previewing the media and slate information was the key to sorting their data.
-
File managers had a huge influence in the data wrangling process.
-
Drag and drop was an indiscoverable and tedious interaction.
Based on these insights, the data sorting experience was revisited to make it easier for the users to improve decision-making by revealing the structure, thus reducing the cognitive load.
TESTING
FOLLOW-UP USABILITY TEST BEFORE IMPLEMENTATION
Following the white boarding session with the development team, follow-up tests were conducted to get an understanding of how the solutions were perceived by the users on set and impacted their overall workflow, considering the time budget, resources and the acceptance criteria for the MVP.
The quantitative and qualitative results highlighted that the improvements in data sorting performance improved by ~16 times with higher satisfaction ratings.
PROJECT OUTCOME
The outcome of this project was a significant reduction in reference data sorting time, and a consistent data sorting structure by DNEG and non-DNEG employees. The overall time spent collaborating with the product team also elevated UX maturity within Agile teams at DNEG. This contributed to a massive boost in UX requests from stakeholders and development teams.
KEY TAKEAWAYS
UX AGILITY IN VFX
DESIGN FAST, FAIL CHEAP!
Emphasize rapid design iteration to minimize costs and maximize efficiency.
Integrate UX early in agile development to steer non-user-centric teams towards effective product design and development practices.
Advancing the science of HCI
Comprehending human X technology interaction is fundamental for crafting technology that enriches rather than impedes the human experience, fostering innovation.
Prioritize knowledge transfer
Establishing frameworks and conducting learning sessions, along with weekly design reviews, fosters the evolution of user-centred design within development teams.
Integrating UX into Agile methodologies
To fully adopt the UX process, begin with easier tasks before tackling the most challenging issues, and engage the team in practicing UX methodologies to familiarize them.
Challenges and opportunities in VFX environment
Key priority is requirements to make the show as efficient as possible. VFX agencies will not fund million dollar projects in cross-contrast environments in film production.
Understanding budgets, and constraints
Prioritize, measure
risks, and conduct rapid usability testing with limited resources; adjust research and design approaches according to time and budget constraints.
NEXT STEPS
Post-MVP launch, usability testing identified the necessity for an
even inclusive interaction model catering to outdoor environments lacking desks for placing laptops. This user requirement will be validated through research and resolved accordingly.
An accessibility audit will be conducted to incorporate design principles for enhanced accessibility and wider usability.