Enterprise B2B Desktop and Web Application

I would like to discuss my case study on one of the modules I worked on Equinix Fabric & Network Edge.

Comprehensive Network Port Ordering and Management Module

I have specialized in Designing consumer systems, order management, and connectivity to various network service providers. My expertise ensures enhanced network performance, reliability, and scalability. I excelled in leveraging advanced interconnection services like Equinix Fabric and Network Edge, enabling seamless, secure, and high-performance connectivity across diverse platforms and regions. 

ROLE & CONTRIBUTION

UX DESIGNER 

UX design for a new wizard powered by new back-end process--reducing average order fulfillment time from 45 days to less than 24 hours (in one reported case): from task flows to low and high fidelity mockups. 

UX RESOURCES 

• Design system: new and sparse

• Visual designer

• External design consultancy

STAKEHOLDERS, NON-UX 

• Product owner, product manager, technical product manager, remote engineering team

• Users: external enterprise and business customers

Project Goal

The goal of this project was to shorten the time between submitting an order for an internet exchange port in one of our data centers to activation of that ordered port; business analysis had established that Equinix's average order to activation time was 45 days.

In order to do that, we needed:

1. A new online ordering process that collected all the information that the fulfillment process would need, from information to create new accounts along the way to ways to collect signed documents, steps that, previously, required separate manual processes following an initial ordering form submission.

2. New back-end processes to minimize the need for any manual handling, whether to ask for additional information or to manage the signing of documents or additional authorizations.

My responsibility as the UX Designer was focused on 1) providing an online ordering experience that allowed our customers to place their orders easily, completely, and accurately, and 2) providing the new back-end business processes everything they needed to fulfill the order in much less time than our average order to activation time prior to this project.

Summary of Process

Ideally, a UX practice is grounded in ongoing observations and data about user behavior in the actual context of their social, task, and tool ecosystems.

Owing to an unfortunate Equinix policy, our project could not get direct access to users, so we drew heavily from other parts of the User Insight Pyramid (see next section) to understand our users and their needs as we went through our project's process steps, each done in collaboration with cross-functional stakeholders:

1. Understanding The Problem(s). Once an opportunity for improvement has been identified, create and develop product, user, and domain requirements with key stakeholders

2. High-Level Designs and Abstract Models. Digest into flows, maps, and abstract models

3. Low Fidelity Designs. Quick and low investment to explore before committing to designs. Output: sketches, mockups, wireframes, and prototypes

4. Higher Fidelity Designs. Using design systems and other high-fidelity tools - create, iterate, and polish. Output: screens, specs, clickable prototypes

5. Develop Application. Work with development to catch and smooth out any wrinkles

6. Release the product to benefit users!

7. Look For Usage Data And Feedback. Watch for data indicating how the design is doing

Design Guidance: Triangulating To Understand Users

Cognitive Science at UC San Diego is an interdisciplinary approach to understanding how we think--using any discipline, evidence, or level of analysis as part of a holistic effort to triangulate and learn everything we can about cognition.

In the same way, understanding our users can be approached from many levels, in addition to direct observation in their natural environments using your specific tool to do exactly the tasks of specific interest to your project. While such a situation has the high potential to provide the most directly relevant design inspiration, guidance, and constraints, other approaches can also provide great design input.

I often think of these sorts of input in terms of what I call a User Insight Pyramid.

User Insight Pyramid 

Use what's available to glean whatever design insight you can.

The Nielsen Norman Group suggests something in a similar spirit in their September 10, 2021 video, "Triangulation: Combine Findings from Multiple User Research Methods" (opens in a new window).

Learning About Our Internet Exchange Port Ordering Users

Designing

I usually approach designing at a few different levels:

1. High level workflows and abstract models

2. Low fidelity designs for key screens and others as necessary

3. Increasingly higher fidelity designs to whatever level enables the easiest collaboration with our visual designer and/or developers

Step 2: High Level Design And Abstract Models: Foundations

I build abstract models (content and navigation models, described in Constantine and Lockwood's Software for Use and task flows as the foundation of any designs with any task complexity.

Step 3: Low Fidelity Prototypes

Low-fidelity designs are quick and low-investment to better support exploring possible designs without premature commitment to any particular design. Output: sketches, mockups, wireframes, and prototyping.

Wireframing The Task Flows

With the TPM's process map, the content models, and the navigation map, it was possible to enumerate a list of customer scenarios, goals/tasks, and "use cases" that we intended to include in our project's release.

This allowed me to draw out a progression of screens and states that a customer would encounter as they went through the process to place an order corresponding to each of those scenarios, what we sometimes referred to as a task flow.

An overview of some of task flows through our IX port ordering wizard. 

Recognizing New Design Pattern Needs

Hopefully, most of the time, what you need is covered by the design system. If not, a new design pattern may be needed.

Identifying New Problems

Abstract models and task flows enable us to identify new problems for which we don't yet have solutions.

It saves design and project time to identify multiple common problem instances in overview rather than trying to design screen by screen.

This allows me to allocate more time up front to find a generalized solution rather than multiple idiosyncratic ones, each costing unforeseen time and consistency for the UX.

Step 5: Develop Application

Working With Remote Development Team

While working across big time zone differences can be a challenge, we used our reasonable overlap in work times to coordinate using video conferencing with all stakeholders, adjusting our frequency to the needs of the project.

Iterations: Tuning

The back and forth between UX and development allowed both the design and implementation of the application to benefit from continued iteration and tuning.

Sometimes, that involved adjusting either the design or the implementation to accommodate a constraint here or there, or as unanticipated issues were identified.

Step 6: Release