Tax software purchase path

Project overview

In 2019, Intuit decided to introduce a new pricing structure for existing customers of ProConnect Tax Online – a cloud-based professional tax software. The new pricing structure would rely on personalized pricing derived from data points such as prior year purchase history. This was a shift away from the previous purchase experience, which consisted of tiered bundles. As a result, a new in-product shopping cart experience was needed to reflect the change.

The business drivers:

1. Positioning the product appropriately against competitors
2. Preserving the business unit’s ROI in anticipation of tax pros migrating from desktop to the cloud 
   

Of course, all businesses need to make money, but I decided that we needed to add in some customer experience drivers to ensure that we were delivering the best user experience.

The customer experience drivers:

1. Ensuring our customers understand the new pricing model
2. Communicating the value we’re delivering to the customer
   

Research

I worked with a researcher to come up with a learning objective. Our objective was to test purchase path prototypes with real customers across the 3 different segments – acquisition, retention, and migration (those moving from desktop to the cloud) users – to identify usability issues, understand their customer purchase journey, and ultimately to provide a better experience. To accomplish this, we conducted user interviews, then followed up with prototype testing. We tested with 10 users across the 3 segments.

Design

I wanted to make the experience of purchasing as realistic and encompassing of the end-to-end journey as possible. After interviewing internal stakeholders, I decided to prototype two different purchase path experiences, and to begin each user flow at the most likely entry point into the product. For acquisition targets, this was a marketing email including a new customer discount. For existing retention customers, this was an email reminding them to take advantage of their early renewal discount. All users ended the flow at the purchase confirmation screen.

Acquisition flow

Retention/migration flow

User experience by customer segment

Wireframes

Testing

I expected to learn mostly that customers who were accustomed to the tiered pricing model would not like the new personalized pricing model, but I was shocked to learn that most users didn’t even notice that there was change! This is probably in part due to the fact that our customers often only purchase once or twice a year, so they are not overly familiar with the experience. Another reason could be the artificially of the test itself – the customers were in the same headspace as they would be in a real purchasing decision. Instead, our most critical findings had nothing to do with personalized pricing, but were instead far more foundational to the purchase experience itself.

My researcher and I plotted the insights on a matrix with the axes “High to low effort” and “High to low criticality to success” based on customer pain points. The top 4 heavy-hitters were the insights in the upper right-hand corner of the matrix (High criticality of success, Low effort):

1. Define individual and business returns
2. Describe what else they are getting for the price
3. Include option to call in the purchase experience 
4. Expose higher volume for pricing

Revised prototype

1. Define Individual and Business returns

The top-most finding was by far the most shocking: Users did not know what they were buying.

• 5 of 10 participants said they would not continue with their purchase because they did not know what they were really getting for the price
• 3 of 10 questioned if e-file was included, 1 of whom was a current customer
• 4 of 10 were not sure what “business” or “individual” returns entailed.

My researcher and I noticed this pattern early on. Users began to ask questions like:

• "Did 'individual returns' include states, or just federal?
• "What business return types were included? 1120S, 1065?
• "What about gift returns? How are those classified?”

Tax pros needed much more information within the shopping cart to even begin interpreting the value they were getting.

2. Describe what else they are getting for the price

Through in-depth interviews with prospects, I learned that most tax software is not priced on a per return basis, but rather on an annual or unlimited license. Because ProConnect Tax Online does not offer these pricing options, it is natural for the customers to think of the cost purely in terms of the tax return itself, rather than the other benefits offered by the software. In order to influence the mental accounting without changing the pricing structure entirely, I decided to include what else was included in the price. 

3. Include the option to call

Many of the large firms would not have proceeded after the “Select Returns” page. They are accustomed to working with a sales rep and negotiating a price, rather than put thousands of dollars down online. While we did have a phone number in the experience we tested, more work had to be done to a) ensure that they were directed to the right sales rep, and b) make sure the option was prominent yet un-intrusive.

3. Expose higher tier pricing

In the original testing for acquisition users, the highest pricing tier we showed was for 200 returns. Many firms that did a high volume of returns scoffed at the low number, and thought that the threshold made us look less than professional. After bringing this Voice of Customer to Marketing, they agreed to showcase one extra tier, up to 300 returns. While this was likely still low in the eyes of the larger firms, it was an improvement nonetheless.

Success measures

Data collection

The only real baseline data point we had to work with was cart abandonment on Select Returns, which was 33% during the 2017 extension season. We originally said that we wanted to decrease cart abandonment as part of this refresh, however I quickly realized that our experience might actually increase cart abandonment, in part due to the change of the price itself, and also due to our added experience drivers that might push them to call Sales rather than complete the purchase in product. Instead, I decided to start measuring the experience by the reasons behind cart abandonment.

While it is difficult to determine success in the absence of baseline data, I pushed for the collection of data as a starting point. So now, any future experience changes can be measured with more accuracy.