Profiles in UX Research: Canopy Tax
Canopy Tax launched in summer 2014 as a cloud-based practice management and tax resolution solution designed to help practitioners with workflow, client communications, document management, and client invoicing. In addition to facilitating management, Canopy offers a tax resolution solution that helps practitioners guide clients through the IRS collections process.
Product Team Lead Nate Sanders talked to Validately recently about how user recruiting and moderated remote testing have been integral to product development since the early days of the company.
What was the impetus for creating a cloud-based, SaaS tax solution for tax professionals?
Shortly before Canopy Tax launched, our founder, Kurt Avarell, was working as an attorney handling tax resolution for clients. The problem was, he found himself drowning in seven or eight different tools as he worked. They were all necessary, but they didn’t work together. Based on this experience, he saw the need for a product suite with all of this functionality in one place.
Also, tax software has remained relatively stagnant in recent decades, so there was plenty of room for innovation—particularly in the cloud.
Tell us about how you weave user research through development and iteration at Canopy.
Our main goal is to ship theories instead of hypotheses. Everything about our product development and discovery process closely mirrors the scientific method for this reason. This has been critical to us since day one.
The first thing we do when we’re pursuing a new idea is diving into user and market research. We spend a lot of time with our users understanding what “jobs-to-be-done” they have. We want to make sure we understand pains they experience in their work everyday, what possible product offerings would add value to their work, and what would make us competitive in a large market.
Once we’re confident we have a solid understanding of what will be valuable to our users, we dive into crafting a hypothesis. The product manager, engineers, and UX designers all collaborate together to drive towards the right solution. We then create high fidelity, and highly functional prototypes and validate our ideas with our users. Working off the feedback and observations we iterate through several rounds of hypothesizing and testing until we’re confident in the solution we’ve created.
When did you decide to utilize Validately’s moderated remote testing for this development?
We’ve used Validately for about a year now. Steve Cohn, Validately’s CEO, spoke at a conference around that time, and we decided to switch from Blue Jeans video conferencing to Validately’s more robust moderated remote testing tools.
Does this supplement in-person testing?
We haven’t really done a lot of in-person testing, to be honest, and we don’t feel we need to. We can get more user breadth and talk to more people using the video and audio engagement of remote testing.
You’re also using Validately for user recruiting. What does that process look like for you, and what kind of user are your targeting?
When we first started user recruiting, we Googled names of tax professionals and compiled lists of contact information. Then, we cold-called each one and asked them to engage in a testing session—while also trying to convince them we weren’t selling anything. That was rough, and it sucked up a lot of our time. After bringing on Validately, it only made sense that we’d utilize their user recruitment. We worked with them on a screener to be sure we were getting the right user base, then turned our focus to product improvement.
Once you have your users lined up, how do you structure your tests?
Before we ever start testing, the UX Designer will prepare a testing script for each session. The script details a series of task based scenarios that can deductively tell us whether the user can accomplish their goals with the solution we prototyped. We prep our users that we’re not going to help them very much, and that they might struggle with the tasks we give them. It’s crucial for us to be able to see where the user struggles, what’s painful, and what we’re doing exceptionally well. Throughout the entire session we ask the user to “talk aloud” and give us feedback on our prototype.
Do you conduct similar testing for live products?
Not a lot. Because we do everything we can to incorporate discovery and user testing throughout the entire development life cycle, we don’t have a lot of usability issues when the product ships. We measure twice, and cut once. We solicit and normalize feedback after every single product release, and fine tune everything we possibly can. We do some testing to focus on “shoelacing” the different features within our product suite to ensure everything within Canopy is intuitive and works together seamlessly.
Speaking of which, what is the balance of qualitative vs. quantitative feedback you’re receiving during user testing?
We really value goal directed design, and we feel it really defines our user experience strategy. Under goal directed design, you’re working to understand the user’s vision and desired end state, and then you make sure you do everything you can to ensure arriving at that end state is as delightful and enjoyable as possible. Given that approach, we focus a lot on observing how well the user can accomplish their goals, and how satisfied they are doing it. We balance that with qualitative feedback. Because Validately’s remote testing allows for both audio and video recording, we can easily note every reaction a user has.
What are some Canopy Tax features you have improved using moderated remote testing?
A core part of our application is completing and filing IRS tax forms. Filling out tax forms is something that really hasn’t changed for decades, and we wanted to enhance the entire user experience for these tasks. We used Validately to test and iterate through several different prototypes to find out what would be most valuable to practitioners as they completed these tax forms. Based on our findings we saw that practitioners were entering the same information multiple times, and that they needed an easy way to see how all the data they were entering flowed onto the IRS tax form. Our final solution cut down the amount of data entry for practitioners dramatically, and allowed them to quickly toggle between a web form and the IRS tax form.
What features of Validately were particularly helpful in delivering these improvements?
Capturing faces during video recording is huge for us. We can watch facial expressions—which is invaluable feedback that’s often lost when you’re just doing audio recording, or when doing in-person testing and the tester is not paying attention. When users are silent, these facial expressions indicate how they’re feeling and show us where product features are not intuitive or where certain features are a pleasant surprise.
The ability to flag moments within videos has been huge, too. It makes it easy to revisit specific user responses that indicate pain points. The only thing that we would like to see is the ability to tag or title flags, so we can then query them to see how often certain topics/tags appear in a testing session. [Editor’s note: This is a feature that is on Validately’s short-term development road map.]
For those who are looking to engage in user research, what recommendations or suggestions would you offer?
In short: measure twice, cut once. Don’t fall victim to the farce that user testing takes too much time, or it will slow you down– it’s a lie. Your job is to create products people want and love. What you come up with on your own isn’t enough. It’s a conjecture. Get outside and validate your assumptions. The product development team at Canopy has moved faster than any team I’ve ever been a part of because of this philosophy. With a lean testing tool like Validately, we have been able to uncover prototype pain points quickly and move from shipping guesses to shipping theories.
Product Team Lead | Canopy
Nate has been Product Team Lead at Canopy since January 2015. Previously, he worked as a UX designer for BambooHR. A UX designer for most of his career, Sanders is very much in favor of a design philosophy that centers on user experience, instead of clinging to inflexible waterfall development.