Modernising data imports

01

Overview

The Advanced Data Automation Wizard (ADAW) was a platform-wide initiative to replace a legacy import experience that had become the final dependency preventing the full decommissioning of an older version of Glow.

While the primary objective was feature parity, the project presented an opportunity to rethink a workflow used by customers to import large volumes of operational data into CargoWise.

Rather than recreating the existing experience, I worked to modernise the workflow, reduce friction, improve discoverability, and make data imports easier to understand and validate before committing changes.

The resulting solution was adopted across the platform, replacing the legacy importer and enabling the retirement of a key dependency on the older Glow architecture.

02

Context

Data imports play a critical role in enterprise software.

Customers often need to import large volumes of information such as contacts, shipments, rates, and operational records. When these imports fail, the consequences can be significant, ranging from wasted time to large-scale data quality issues.

CargoWise already had an import solution within the legacy Glow platform. However, the experience had not evolved alongside the rest of the product and felt increasingly disconnected from the modern web experience.

As part of a broader migration effort, the importer needed to be rebuilt within the new platform.

This created an important question: should we simply recreate the legacy workflow, or use the opportunity to improve it?

03

Understanding the existing experience

The project began with a detailed analysis of the existing importer.

I mapped the end-to-end workflow, reviewed customer pain points, and examined how similar products approached large-scale data imports.

Competitor analysis included platforms such as Airtable, Monday.com, and Notion, each of which approached data mapping and validation in different ways.

This research helped identify both industry conventions and opportunities to improve the existing experience.

• Limited keyboard efficiency — drag-and-drop mapping was inefficient for power users working with large datasets
• Technical language — platform terminology increased the learning curve for less technical users
• Lack of import visibility — users had limited visibility into how imported data would appear once processed
• Fragmented navigation — moving between mapping and configuration required traversing multiple menus and submenus

04

Defining the workflow

Before designing interfaces, I focused on understanding the complexity of the import process itself.

The existing workflow contained multiple decision points, validation paths, and error states.

Mapping these journeys exposed opportunities to simplify navigation, reduce unnecessary transitions, and create a more coherent experience.

This work became the foundation for later design decisions and ensured that improvements addressed the workflow itself rather than simply refreshing the visual design.

05

Exploring solutions

With a clear understanding of the problem space, I explored multiple approaches through collaborative workshops, design critiques, and iterative concepts.

Throughout the project, I worked closely with stakeholders to balance three competing priorities: maintaining feature parity with the legacy importer, supporting existing customer workflows, and improving usability wherever possible.

This required careful evaluation of which established patterns should be preserved and which could be modernised.

06

Designing a better import experience

One of the most significant improvements was the redesign of the column-mapping experience. Rather than relying primarily on drag-and-drop interactions, the new approach provided a more structured and scalable workflow that better supported high-volume imports and keyboard-oriented users.

To improve confidence before import, I introduced a preview experience that allowed users to inspect how their data would appear after processing. This made mapping errors easier to identify and reduced the likelihood of issues being discovered after import completion.

I redesigned the relationship between import configuration and data mapping to reduce unnecessary navigation. Users could move more efficiently between related tasks, making it easier to validate changes and iterate on mappings without repeatedly traversing nested menus.

Where possible, technical terminology was replaced with language that better reflected user intent and task completion. This helped reduce cognitive load while making the workflow more approachable for less technical users.

07

Outcome

The Advanced Data Automation Wizard was successfully delivered and adopted across the platform.

The project achieved its primary goal of replacing the legacy importer while also delivering meaningful usability improvements beyond simple feature parity.

• Replaced the legacy import experience
• Adopted platform-wide
• Removed a key dependency on the older Glow architecture
• Enabled the decommissioning process for the legacy platform
• Improved support for keyboard-driven workflows
• Greater visibility into import outcomes before execution
• Reduced navigation friction between related tasks
• Clearer terminology and workflow structure

08

Reflection

One of the most valuable lessons from this project was that feature-parity initiatives do not have to result in feature-parity experiences.

Projects like ADAW can easily become exercises in replication, where success is measured by how closely the new version matches the old.

Instead, I treated the migration as an opportunity to challenge assumptions, revisit long-standing workflow decisions, and improve the experience wherever possible without disrupting existing customer workflows.

The result was a solution that not only met the technical requirements of the migration but also delivered a more efficient and understandable experience for the people using it every day.