PROBLEM
Context
The editor is used by accounting and finance teams to produce reporting and presentation-ready documents. Tables are central to these workflows, containing multi-period comparisons, financial schedules, and summaries that must remain stable across pages and columns.
​
Unstable table behaviour
Tables ignored column and page boundaries, often overflowing past the layout instead of continuing predictably. They also misaligned after resizing and frequently lost structure after paste, import, or reuse across documents.
​
Ambiguous controls
The table menu relied on ambiguous icon-only actions with no labels or grouping, making it difficult for users to understand whether actions applied to a cell, row, or the full table.
​
Impact
Teams developed workarounds that slowed reporting cycles and reduced confidence in the output. Support tickets confirmed the deeper issue: users began avoiding tables entirely in complex documents.
​
Design challenge
Stabilise table behaviour and clarify interaction controls while preserving established authoring patterns, within a fixed delivery timeline.
PROBLEM
Legal and financial professionals manage large, structured datasets spanning dozens of interdependent entities.
Bulk updates such as jurisdictions, attributes, or ownership states required repetitive point-and-click actions across many items, often taking hours and increasing error risk.
​
While AI could compress this work into a single instruction (for example, “Move all US entities to the Cayman Islands”), speed without control undermines trust in regulated environments.
MY ROLE
I led the end-to-end design over a three-month engagement, working closely with a PM and two engineers. I owned the interaction strategy, defined table behaviour across all critical editing paths, and redesigned the table menu system. My focus was not only on fixing visible failures, but on eliminating undefined behaviour that caused instability to resurface elsewhere.
PROBLEM
Legal and financial professionals manage large, structured datasets spanning dozens of interdependent entities.
Bulk updates such as jurisdictions, attributes, or ownership states required repetitive point-and-click actions across many items, often taking hours and increasing error risk.
​
While AI could compress this work into a single instruction (for example, “Move all US entities to the Cayman Islands”), speed without control undermines trust in regulated environments.
KEY DESIGN DECISIONS
Preserve capability, increase predictability
Engineering initially proposed blocking table continuation across columns and pages to eliminate breakage. While simpler to implement, this would have removed a core capability for accounting and reporting workflows, where tables routinely span pages.
I pushed to preserve natural continuation and make it reliable. Tables now flow consistently across columns and pages, matching the behaviour of other layout elements. The solution aligned with patterns already established in a neighbouring module, reducing cognitive load and supporting future reuse.
​
Define behaviour across edge cases
The instability was not caused by a single defect. It emerged from undefined behaviours across many editing paths, including resizing, pasted data, merged cells, imports, and template reuse.
​
I created an edge-case matrix mapping input conditions to expected outcomes and worked with engineering to close each gap systematically. The framework was designed to be reusable, allowing the same behavioural definitions to be applied when tables are extended elsewhere in the product.
​
Clarify interaction scope
The existing table menu relied on ambiguous icon-only controls with unclear scope. With no labels or grouping, users struggled to understand whether actions applied to a cell, a row, or the full table.
​
I reorganised the menu into four clearly scoped sections, introducing labelled actions and redesigned icons to make the target of each action explicit. The interaction model and visual hierarchy were aligned with another editor module, creating a consistent mental model and reducing future design and engineering effort.
OUTCOME
Table continuation across columns and pages was stabilised and the menu redesigned with clearly scoped, labelled controls to reduce ambiguity. The work established explicit behaviour definitions for future table functionality, though release was deferred as other roadmap priorities took precedence.
TAKEAWAY
In reporting workflows, predictability is the product. Defining behaviour across edge cases and clarifying interaction controls restored trust in the editing experience without changing established authoring patterns.