THE RE-DESIGN OF: HOST PRICING TOOLS FOR TURO
By early 2018 Automatic Pricing, Turo’s intelligent pricing algorithm, had shown a steady decline in adoption rate from hosts. Because Turo believed that the algorithm sets better prices that lead to more bookings, it became crucial that we increase the amount of hosts who opt to use it.
We set out to research and uncover the friction points behind the Automatic Pricing feature. We pulled user feedback, conducted informational interviews, reviewed our tracking, analyzed competitors, and conducted a design audit of the current experience.
The pricing algorithm wasn’t always accurate
We had a strong signal when we saw an increase in cancellations and feedback from hosts saying they turned off Automatic pricing because the suggested price was either too high or too low for their market.
“The automatic rate that Turo picked is ridiculously low! It’s not worth the risk of my $130,000.00 car to be rented for such a low price!!!“
— Host cancellation reason
Armed with some assumptions, our small but mighty pricing team sought to answer one question:
HOW MIGHT WE HELP HOSTS BETTER MANAGE THEIR PRICING WHILE STILL INCREASING REVENUE FOR TURO?
Hosts have different strategies for pricing
Some hosts want to maximize utilization of their cars. We found this happens most often with a host who is using Turo’s platform as a business. They likely have more than one car and want their car(s) continually booked so that they are making money and not sitting idle.
Other hosts want to maximize earnings for each trip. These types of hosts want to make the most they can for each day their car is being driven. They are hyper sensitive to the value of their time, which may stem from their location and relative distance from airports or common delivery spots. Other hosts are hyper sensitive to the cost of maintaining their car.
Hosts want more control with Automatic Pricing
When we asked hosts why they opted out of Automatic Pricing, the number one complaint was not having control over determining their weekly and monthly discounts. Hosts also didn’t understand how the discounts could make the price drop below the minimum daily price they set.
Manual pricing is time consuming
The hosts that turn off Automatic pricing could set one flat price but struggled with knowing what to price to set. They often spent hours researching and checking similar listings in their markets prices and cross referencing their availability. They also have to spend time manually overriding their prices for certain days or on weekends with a pricing calendar that has a sub par user experience. All of this hinders their workflow and makes managing their vehicles a difficult task.
CURRENT DESIGN AUDIT
I took a look at the original flow and quickly realized there were some big problems about educating a host on the value of Automatic pricing.
Too much hidden information
With the current design, we were hiding a lot of important context for each setting inside the question mark tooltips and we were relying on a user to discover that information on their own.
Difficult to digest
Additionally, one of the most important questions “How much will I earn” was not easy to digest or understand.
Complex error states
Some of the form fields had validation logic that was contingent on other fields. For example, if a host set a minimum daily price and then added a discount that would push the daily price below the minimum, a poorly worded error would occur, often leaving hosts to scratch their heads.
We analyzed other market places like Airbnb, Etsy, Getaround, and thumbtack for ideas on how we could improve our messaging around pricing. Air bnb has a really great example of how to provide the right amount of context. They also force a host to determine pricing settings in the listing flow, which creates additional friction but sets a host up for success.
After a month of diving deep into the current issues with pricing, we discovered several areas we could focus on. We developed a plan that would pave the way for the next generation of our pricing.
Fix the Automatic pricing model
We knew that no matter what we did to improve the UI and UX of the pricing feature, hosts would not use it if we didn’t improve the logic behind pricing itself. The Data scientists set out to simplify the variables and incorporate more intelligence.
Give hosts with Automatic Pricing more control
We heard hosts loud and clear when they said they wanted control over their discounts. The second most common piece of feedback about the failings of automatic pricing was that hosts wanted the ability to override any day’s price on the pricing calendar. This is what shaped our new view on Automatic pricing vs manual pricing. We decided it would no longer be a binary choice, but that we would dedicate the next six months on blending the two together.
Give hosts who choose manual pricing more guidance
While we felt that adding back more control to Automatic Pricing would solve many of the complaints hosts had with the feature, we also knew that fluctuating prices aren’t for everyone. Some hosts still want to set one price and leave it. For those people, we wanted to provide more guidance on what Turo thinks is the right price. They can choose to ignore it, but will have piece of mind.
A majority of the designing process came down to organizing the form fields into a more logical order and introducing a less severe “nudge” state to the fields that would provide more guidance.
Some of the issues with the existing experience involved the position of Automatic pricing toggle. Turning it on and off would change the rest of the fields on the page, and yet it was positioned at the bottom of the page. We moved to correct this, and added the ability to set custom discounts in automatic pricing.
Desktop suffered from a similar organizational issue as IOS and Android did. Automatic pricing was moved to the top of the page, and more white space was added between sections to create separation. We also re-worked the layout of the form fields themselves to improve scanability.
In addition to re-organizing the content and adding more contextual and explanatory copy, we updated the UI and utilized some newer design patterns such as the white modals. This helped us give full focus to one setting and provide just enough context for that specific setting inline.
We were thrilled to see that by giving hosts more control over their prices, simplifying the algorithm, and changing the way we talk about pricing, we were able to increase adoption by 10% in a matter of a few months. The trajectory is still moving up.
Our vision for Pricing was to create more visual representations of Turo’s recommended price. We hope to surface more data to hosts so that they feel even more informed and empowered to manage their pricing and their business.