Principle-led product design: how we’re bringing more confidence to car buying

Ben Smith
Ben Smith
Dec 6, 2019 · 7 min read

Within the Consumer Experience team at Auto Trader, everything we do is designed to make car buying easier and help buyers find the right vehicle as quickly as possible.

Over the past six months, we’ve been working on changes to our adverts and price indicator feature — a system designed to create trust and transparency between buyers and sellers in the automotive marketplace. This article shares our journey, and the principles that guided us to our outcome.

The problem(s) we set out to solve

Our consumer research has regularly identified a lack of price transparency as one of the key pain points in the car buying journey.

Three years ago, we introduced a price indicator system to help consumers understand if a car has been priced fairly in relation to other similar cars for sale on our marketplace. We added ‘Low’, ‘Great’ and ‘Good’ price indicators to our adverts, but no indicators for ‘High’ priced cars. In hindsight, this felt like a missed opportunity to add greater transparency.

Further research also showed us that price isn’t the only factor consumers consider when choosing between cars. We believed we had additional data we could expose to help consumers make decisions and understand a car’s true value.

We set ourselves three objectives for this project:

Objective 1: Give consumers confidence that the price they pay is fair by providing greater transparency to our price indicator system.

Objective 2: Help consumers understand the price in context of other factors that make up the deal, so they can judge for themselves if this represents good value.

Objective 3: Drive more connections between buyers and sellers; the ultimate measure of how well our product is performing (and a proxy metric for sales).

Using research to define our principles

We were able to draw on proprietary research, including our Auto Trader Market Report and Car Buyers’ Report to help shape the consumer problems we wanted to solve. We also had the fantastic support of Idean, Contentsquare, and Endless Gain, who helped with our qualitive and quantitative research.

From our research, we developed five guiding principles:

1. This isn’t ONLY about price

The highest priority consumer jobs-to-be-done relate to deciding if the price is fair, if the car is reliable, and if the seller is trustworthy. Content relating to these jobs should be organised together and structured in order of priority. This involves focussing on information architecture and content design.

2. Insights can build trust

We’re in the privileged position of having more data on cars than any other website in the UK. Our research suggested that by surfacing our data in a way that is valuable, unbiased and insightful to consumers, we could build consumer trust and enable users to make smart, informed decisions.

3. Price, mileage and specification are connected

When consumers are shortlisting cars, they often compare the price, mileage, specification and optional extras. There is usually a trade-off: Is it worth paying more for a lower mileage car? Is the car with the panoramic roof worth the extra money? Our valuations adjust for mileage and spec, but we want to make it easier for consumers to understand the differences and trade-offs when they are in the shortlisting phase of their buying journey.

4. Price is explicit, value is implicit

Mileage and spec are examples of data points that explicitly influence the value of a car, which our machine-learning algorithms can model. There are other factors that don’t explicitly influence the valuation a car but do influence consumers perception of whether or not it’s a good deal e.g. paying a premium to buy from a franchise retailer, a full-service history, or if the car has had one/few owners. We want to help consumers find and understand the implicit value in our adverts.

5. Prioritise what’s important to consumer, not what’s important to us

Looking at behavioural data, we observed that some content positioned further down our adverts was getting a high amount of engagement, suggesting we were making it hard to find the content they’re looking for. Going forward we will be prioritising content consumers want to see (e.g. car description and spec) not what we want to promote (e.g. our new products).

Collecting our early thinking around the relationship between the car, the seller and the price, and how user percieves value.

Working in ‘phases’

At Auto Trader, we work in cross functional, agile teams. As a company, we look to get validated data from our live audience as quickly as possible. Some ideas we’ll build and test in days (even hours), and other projects require more time up-front to develop and test the proposition before we commit anything in code.

Reflecting on this project there were four ‘phases’, with plenty of overlap between them.

Phase 1: Proposition development:

We spent 4–6 weeks reviewing our existing research and working with Idean to help us develop the consumer proposition. We did consumer testing and held a workshop with retailers to test our early thinking and give them the opportunity to input into the design process. It was in this phase that we developed the above principles that have guided the rest of the product development.

Phase 2: Product design and testing

The output of the proposition development was a direction and well-formed idea, but we went through a number of rounds of testing and iteration to get this to work in our product and design system.

We tested flat Invision prototypes, using UserZoom to get rapid feedback on our evolving design. We got into a really good cadence of getting consumer feedback and iterating our design and content on an almost daily basis. We measured ‘how easy’ consumers found our designs, and compared this against our competitors, so we had some quant data to evidence the changes we were making.

Phase 3: Development and live testing

At Auto Trader, we favour early delivery and fast iteration based on user feedback. When product changes were small, and mostly front-end UX, we broke the work down into small, logical parts, and released regularly based on what we were learning from how consumers were interacting with it.

Alongside these UX-focused changes, we also built some new features which required us to work with teams around the business and make changes further up the tech stack. Examples of this included making changes to our price indicator system and building the new mileage indicator.

Phase 4: Go to market

As the largest UK automotive marketplace (over 75% of time spent on automotive classifieds is spent on Auto Trader), we have to balance the needs of buyers and sellers — a lot trickier than it sounds!

Over the last few years, we’ve introduced a number of features that were designed to make car buying easier for consumers, but these features were initially unpopular with retailers (our customers). Examples of this include our dealer reviews, changes to our search algorithm, and the first iteration of price indicator.

For this project, we knew it was important to bring retailers with us on the journey, so they could understand the impact of these changes from the start. We hosted a workshop with them in the very early stages and showcased our designs with retailers at our regular Masterclasses. When we were ready to take the product to market, we hosted a number of full-day training sessions with our RDS Tribe (sales & support), who are on the front-line of handing conversations with our retailers and making sure they understand the value of any changes we intrroduce.

So… what did we build?

A new (more transparent) price indicator system

We’ve included ‘Fair’ and ‘High’ price labels and introduced a visual ‘dial’ to help consumers better understand how these indicators work as a scale. We are also exposing the £ variation from our market valuation to add further context to the indicator and give consumers full price transparency (addressing one of the biggest consumer pain points).

Greater transparency around mileage and spec

We’ve introduced a new mileage indicator, showing consumers the variance from the average mileage of similar vehicles on Auto Trader. We also highlight to consumers where a car has additional features above what typically comes as standard, and what those features are. These are both examples of us surfacing smart insights and revealing the data that sits behind this.

Help users find implicit value

We’ve made it easier to find the service history information, the number of former owners, the benefits of a manufacturer approved scheme (where applicable), plus retailer ratings and awards. These are things that we learned consumers place value on, but aren’t explicitly included in our valuations.

Organise content around consumer jobs-to-be-done

We’ve reorganised the content on the FPA and created a clear information hierarchy that gives consumers confidence in the price, the car and the seller.

Clearer hierarchy of calls-to-action

We’ve simplified the number of calls-to-action on the page and created a better hierarchy around our primary CTAs that deliver leads to retailers, and secondary CTAs (like viewing the location or visiting the retailer website).

Our new advert, designed to give consumers greater trust in the car, the price and the retailer selling it.

And what were the results?

There are two primary metrics we use to measure the performance of our core journey; advert views and leads delivered to retailers (leads can be emails, phone calls, online chat etc.). We report on these in numbers in absolute terms to the city, but as the consumer product team we focus on the metrics per session and conversion through the funnel.

We’ve been testing these changes for the last two months, and have seen no significant change in ad views per session, and a 6.8% increase in the conversion from advert views to leads. We’re super pleased with this, and we think it is a strong indication that we’ve been successful in giving consumers greater confidence when they are on our marketplace, which ultimately will drive more sales for our retailers.

This project was led by the Consumer Buying Experience Tribe but has involved multiple teams from across the business coming together to deliver this change. We’ve made changes to our retailer facing systems, the machine learning that powers our valuations and price indicators, and our data feeds and APIs. The coordination of the tech work, sales activity and consumer and trade marketing has been managed by our fantastic delivery team — a great example of how we work collaboratively at Auto Trader to lead the digital future of the UK automotive marketplace.

Fancy joining the team? We’re hiring!

Auto Trader Workshop

Ben Smith

Written by

Ben Smith

Auto Trader Workshop

Design, UX and Product Thinking from Auto Trader

More From Medium

More from Auto Trader Workshop

More from Auto Trader Workshop

How to create an ownable suite of icons

More on UX from Auto Trader Workshop

More on Product Design from Auto Trader Workshop

More on Product Design from Auto Trader Workshop

Product design lessons from ‘Mrs. Armitage on Wheels’

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade