Enterprise Resource Planning System for Utilities
United Systems & Software
THE INITIAL TRIGGER
As part of an effort to pivot towards being the leading ERP product for utilities in the Mid-west and the South, United systems embarked upon the journey of redesigning its legacy product suite.
The new suite of applications needed to be standards-compliant and help cities and utilities modernize their core business processes using automation and simplicity.
THE CHALLENGE
Utilities currently have a highly manual and time consuming workflow for their daily operations and our aim was to reduce the time and effort that users spent on these tasks. I did this by:
1) Improving information architecture, navigation and taxonomy.
2) Making the platform compliant with accessibility standards.
3) Streamlining design patterns that make conducting daily operations more accessible to different kinds of users
4) Automating processes where necessary; like generating reports, test data entry, sending notices etc.
​
However, one of the biggest challenges with the redesign was that we couldn't be very radical with the redeisgn and alienate 5000+ current users. We had to make sure our redesign struck a delicate balance between optimizing current processes and aligning with current user mental models.
BEFORE REDESIGN
Information heavy and cluttered
Little of consideration of user's tasks and workflow
Difficult navigation patterns
Desktop app before redesign
Mobile app before redesign
THE PRODUCT
​
Backflow was one of our core products of our suite that brought in hundreds of thousands of dollars of revenue each year. Our application let users manage their customers' utility accounts, payments, compliance tests, meter readings, devices and work orders.
MY ROLE
I embarked on the research initiative to understand how our application was being used currently. I did this in collaboration with the Sales, Customer Service and Marketing teams. After modeling the product's present state, I created a design vision for the product's future, and served as researcher, designer and product manager to ensure the success of the redesign.
AFTER REDESIGN - IMPACT
28%
increase in new customer adoption
20x
faster task completion
40%
reduction in customer support calls
10x
faster output by Dev due to set design standards
Web app after redesign
Mobile app after redesign
DISCOVERY AND DESIGN PROCESS
OVERVIEW OF METHODS USED
FINDINGS
​
I conducted interviews with users and talked with the different teams within the organization to get as much of a complete picture of why the product suite is the way that is currently. I used conversations with users and different teams within the organization to determine the main user pain points and organizational limitations. The major challenges that emerged from those conversations were as follows:
Challenge 1: Backflow's breadth of users
The product was trying to be all things to all users. This resulted in a content-heavy product, unintuitive navigation and features that lacked clear purpose or hierarchy.
​
To tackle this issue, I conducted persona research to create models of our main users and used those as a guide to create more personalized experiences.
Challenge 2: Lack of intentionality
​
In order to cater to each client's individual needs, the product had accumulated features created for one-off clients and scenarios. This had resulted in a lot of "fluff" consisting of screens that didn't serve a purpose anymore.
​
To tackle this issue, I conducted a content audit of 150+ screens to determine what content was being used often, what could be repurposed and what can be purged. I coupled this with discovery interviews where I listened to users and other stakeholders to understand what everyone wanted the new web app to achieve.
Content Audit that I conducted to analyse the current state of the product.
Challenge 3: Wasted time to get to important and frequent tasks
​
After conducting a heuristic analysis, I found that it took over 10 seconds for users (and over 17 seconds for new users) to navigate to their target location/action. This was because of an unintuitive navigation and information architecture. Confusing and technical naming also played a part in this.
​
To tackle this issue, I conducted multiple rounds of card sorts to improve the information architecture and taxonomy to reduce the time and clicks taken for a user to finish an important task.
Example of a test participant's card sort activity board.
OUTCOME
​
Using the above mentioned research insights, I followed 2 solution patterns that I used as an overarching guide to redesign the product. Shown below are just some instances to demonstrate how these solution patterns were used.
Solution Pattern 1: Simplify process to get usable data
System admins struggled to find specific items because of poor search. They're usually in a rush and it's like finding a needle in a haystack to find a specific result. Most of our customers had over 10,000 users in their system. Even after getting to a specific result, users sometimes realize they navigated to the wrong item and now need to start that entire process again of finding what they're looking for.
Users now have a robust search system where they can filter, sort or search by keyword to find what they're looking for. The tabbed navigation and custom-made categories help them narrow down their search even further. The "Side Peek" allows them to view and edit quick details about their selected item without erasing their search parameters. If the choice was correct, the user can expand their selection to take more detailed actions.
Solution Pattern 2: Balancing Personalization vs Customization
​
Personalization
Our average user didn't have the desire or time to figure out how to optimize our platform for their gain on an every day basis. So, we personalized their experience based on our previously defined user roles. These personalizations were important considerations while showing them catered data interpretations, assigning them permissions or showing them a specific view of a screen.
Concept of a dashboard design that provides personally relevant interpretations of data instead of making the user do the heavy lifting of figuring out how to use the data. Users are provided insights about how they can optimize resource allocation.
Customization
Research indicated that different utilities had different workflows that were decided by their administrators. This is where they needed customization and control over certain defaults setup options. The goal with the new designs was to make the experience less overwhelming for regular users by reserving these intricate customization settings for admin-type users only by housing them in their own separate section, out of regular view for general users.
Admin-level users have to ability to personalize how tests are grouped on the main page by defining groups based on attributes like due dates, hazard type etc. This makes it easier for them to recognize the tests that require attention without having to filter them each time.
Solution Pattern 3: De-emphasizing noise (and showing more of what's necessary)
​
Research showed that our users spent almost 75% of their time filling out forms for creating and updating compliance tests, creating new billing accounts and creating/updating work orders. In a typical workflow, users didn't use more than half the inputs and options available which indicated a lot of clutter and opportunity for streamlining processes.
Users are able to quickly find what they're looking for using tabs and clearly demarcated sections. The 3-column layout provides a cleaner layout and better scannability. There is clear indication of editable fields vs text-only fields. Users can also view highlights of the form at the top.
Learnings
1. Since this was my first time redesigning a legacy product suite, it didn't strike me how different it would be than creating a product from scratch without any backlog or technical restraints. Working with a legacy software means working with systems that are archaic and resistant to change. I had to maintain a delicate balance of strategically and slowly suggesting improvements without disrupting the technical ecosystem already in place. As a result of this, I had to tone down my designs from being radical to being achievable in the given constraints.
​
2. Building my own relationships with other teams was pivotal with to the success of this project. Since I was the first and only designer at my company, I really had to work to build relationships with experts from other teams and carve out a place for design within the workflow of the organization.
​
3. Aligning stakeholders on questions like "what problems we're trying to solve?", "what metrics of success are we using?" looks straightforward on paper but really isn't. My biggest learning was that I needed to get all stakeholders (dev, manager, CTO, customer support, qa, marketing" in the same room rather than having 1-1 conversations with them. 1-1 conversations were a good starting point to get people on your side but true alignment only happened when everyone had the same expectations of the product.
(this project is still ongoing so please reach out to me @janhavishah99@gmail.com to talk further about the project)