Function Point Productivity Software
Function Point (FP) is a Software as a Service company that provides an all-in-one financial and workflow management solution for creative agencies.
I was brought in as the first designer to the company. Prior to my joining, FP did not have a Product team. The software grew overtime based on client feature requests that were designed and implemented by the Development team. The company suspects poor UX is a contributing factor to customer churn. My role was to work with the VP of Client Experience, who is a subject matter expert, to envision and embark on the product redesign journey.
My Role: UX Designer
- User Research
- Concept / Beta Testing
- Writing User Stories
- Post-release Analytics
The first item we set out to do was to understand our users and how they were using the product. Through a series of user interviews and on-site user shadowing sessions, we identified 5 main user roles: Owner, Account Executive, Traffic Manager, Accountant and Production Staff.
We’ve also observed that roles within small creative agencies are very fluid. Depending on the agency, a staff member can perform tasks across multiple roles. E.g. a Digital Producer often manages the client account and keep track of all project work associated with the client as thus play the role of Account Executive and Project Manager. Because of this, we created very light weight user personas and opted for using user mental model and workflow diagrams to map out user needs and intent.
1. Work pace at a creative agency is very fast.
Project priorities and deadlines can change several times in a week. Users, especially Account Executives and Project Managers, felt it was too labor intensive to keep project information up to date within FP and so they don’t do so. Information in FP does not reflect true project status.
2. Fragmented & Inconsistent workflows lead to bad data.
Because the FP system was built over time based on a series of client feature requests, there are numerous workflow in the system that are not easy to discover and also not well integrated. Data need to be updated in multiple places and it is very easy to miss something and lead to inconsistent data.
3. Desire for flexible business intelligence.
An “all-in-one” solution is important because the Owner and Accountant wants holistic business intelligence. The data they are looking for is stored within the system. However, there were no centralized views where you can see everything.
Product Redesign Strategy
Re-imagine Schedule & Tasking
Scheduling and Tasking is a central part of agency workflows. This was the first area we redesigned and the focus of this case study.
Analysis of the old Schedule module
Project Managers indicate they can easily be spending 2-3 hours a day in the Schedule module. Frequently, they will create the schedule at the beginning of the project for the client, but the schedule never gets updated because the module is difficult to use. We need to transform the schedule into a tool that is quick and efficient to up date.
Old school “View” and “Edit” mode: The schedule was built 7 years ago based on the premise of a “view” mode and an “edit” mode. The “Edit” page consists of an large HTML form where all information are loaded in as input fields and thus loads slowly. Loading was so problematic that, based on user feedback, they had to place in 3 different edit modes: “Outline”, “Managed” and “Full” to load more or less fields. In reality, users would much prefer a spreadsheet-like experience where they can quickly edit information inline.
Complexity of the old schedule structure: Tasks within the Schedule was built based on a “parent” and “child” task model that is difficult to adjust. Users frequently unintentionally convert a child task into a parent and this resulted in a lost of information because less information is stored on a parent task level as compared to a child task. The old structure also supports schedules up to 12 levels deep. However, users rarely create schedules more than 1-2 levels deep.
Step 1: Simplify schedule structure
To avoid confusion between “parent” and “child” tasks and to allow us to expand on future task functionalities, we opted to simplify the schedule down to 2 levels consisting of Tasks and Task Groups. This means we needed to migrate our existing customers’ schedules into this new structure.
Through analysis of customer data, we discovered agencies typically use 3 types of schedule structures: flat list of tasks, simple 2-level trees and more complex multiple-level trees between 3-5 levels deep. Therefore we came up with 3 corresponding transformation schemas and conducted concept tests with customers that have each type of schedule structure.
Next, we created supporting documents for the rollout, including interactive walkthroughs to explain the schedule transformation process to customers. Finally, we implemented a staged rollout plan.
Step 2: Make schedules easy to edit
Once we’ve simplified the schedule data model, we’re in position to redesign the schedule interface to facilitate ease of editing. Project Managers often indicate they would like the system to be as easy to use as a spreadsheet. They want to simply click into a cell and start typing.
Through the use of code-base prototypes, we investigated different ways for providing a spreadsheet-like experience and how various interactions affected the perceived speed of editing. We’ve built the new Schedule Grid View as a 2-axis scrolling grid that allows for easy inline editing.
Step 3: Enhance data visualization
Next, we added a new Gantt view to the schedule help Project Managers better visualization project dates across time. This view also allows them to quickly move items on the timeline and to share the updated schedule information to the team and their clients.
in customers who use schedules
avg schedule views per customer