Selling a car with AutoTrader

I already went trough a personal experience of trying to sell a car with AutoTrader but didn't want to use that as the only source of information. So I conducted quick tests with two of my friends, telling them to 'sell their car trough AutoTrader'.

My goal was to evaluate the overall user experience, identify potential usability and other general problems.

Valentina

Laurence

Testing with friends

* This step helped me find a problem to solve for this presentation. In an ideal case I would have access to complementary insight (analytics, customer feedback etc.).

There it is, quite easy to spot. No problems here.

Hm, what's this?!

Nevermind, my model is Cruze... but what variant?

No idea about the trim or derivative either... I know it's a 1.8 engine but not diesel, maybe this one?

Urgh... that's not the one. How do I change this now? No that's not it... hm, I'll just choose the closest one...

Ok, what next? Oh there it is... 'Next step'...

Ok, mine's VW Golf 2012 BlueMotion.

Not sure about HSP though... which derivative?

Oh ok... the others get pre-filled then...

Where's the... ah there it is... 'Next step'.

Hm... I have no idea how much mine's worth. Is there no evaluation or anything to help me?

The car is not that old, I'll just put in a bit more than half of the original price. But could I get a better deal?

What's the difference between 'Continue' and 'Next step'? Oh, ok...

What's that? Hm... I guess a preview? Doesn't really look like one...

Hm, my Golf was around 15k when I bought it 3 years ago... How much would it be worth now? The milage is quite good...

Where could I get a genuine estimate?

VWs hold their price quite well... I'll put in 13k... see what happens

I have no idea what features my car has... I know it has electric windows, but other than that...

Ah ok, it has an alarm and air conditioning as well... it's quite hard to scan this list...

What should I write in 'Other details'? It seems quite important but I don't know where to start... I'll leave it empty.

Electric windows, air conditioning, alarm, alloy wheels, lots of airbags but not sure which ones...

It has full service history, not sure about the date of MOT though... can I choose month/year only?

Other details? What is this? I really don't know what to write... It's in excellent condition, kept in a garage... I'll write that.

'Add photos'... ok, let's go.

Hm, this is taking a while... probably my slow internet connection.

What size should the photos be?

Hm, managing these is a bit tricky.

I don't want to expose my private phone number. Can I skip this? I guess not...

'Contact by email', yes.

I don't want to enter my personal number... ah ok it's protected.

Oh! What's this? I thought this was free... I feel a bit... 'tricked'?

What's the difference between these anyway? '6 weeks advert', hm, I don't know. I just want to sell my car asap. Is this going to take that long?

Hm, that's not very cheap is it? No free option? The main difference is in running time of the ad...

There's nothing on this page that would convince me that my car will be successfully sold. I'll go for the cheaper option, see what happens.

Premium only has the extra desktop feature. I don't think that's worth it. I'll go for the 'Standard'.

I identified some usability problems and some more fundamental problems people had with this interaction/interface. 

Problems identified

• the users felt the process of creating their ad was unnecessary long and cumbersome,
• they didn't know what their car is worth (possibly leading to worse deals or harder sales),
• they were surprised about the pricing, even felt a bit 'tricked'. They didn't see the value they would get out of selling their car with AutoTrader ('how is this any better than other sites?').

* Problem identification would ideally be supported by further insights (analytics: how many people don't complete the ad? How many people go for which pricing plan?)

I believe that

creating a simpler and more transparent 'sell my car' onboarding process

for first-time car sellers

will achieve a higher rate of successful car sales and an increased product satisfaction.

My hypothesis

I will know this is true:

• when I'll see a decrease in uncompleted ads (quantitative),
• when I'll see an increase of successfully completed car sales (quantitative)
• through qualitative feedback (user testing) and quantitative customer feedback (NPS & SUS)

Subhypothesis: Empathy Map

I created a so called 'proto empathy map' based on my research. Ideally this would be done with the whole team on a whiteboard and sticky notes so it would remain easy to update (once we learn something new about the user).

Subhypothesis: Outcomes

After defining who I'm designing the solution for and why, I wrote down the outcomes that I would like to achieve with this new solution. These should be the answers to how I'll get there.

1. Simple, easy and transparent ad setup process,
2. reach out to a bigger audience,
3. get the best deal possible,
4. sell the car as soon as possible.

Subhypothesis: Features

Brainstorming for ideas for features. What could AutoTrader build to achieve those outcomes?

1. Simple, easy and transparent ad setup process,
2. reach out to a bigger audience,
3. get the best deal possible,
4. sell the car as soon as possible.

Initial sketches

After I defined my subhypothesis (the what, for whom and why) I went on to start producing the first sketches.

Converging

I combined two ideas into one in the second iteration round.

Higher Fidelity

The first wireframe based on the sketches.

The Next Steps

1. Create an InVision or HTML & CSS prototype.
2. Test the prototype (qualitative feedback).
3. Establish the SUS and NPS baseline score.
4. Build an MVP.
5. Test the MVP (qualitative tests + A/B test & customer feedback)
6. Repeat the measurement of SUS and NPS scores, compare to the baseline scores, analyse A/B test results.

Thanks

I enjoyed working on this assignment and would love to spend a lot more time bringing it to life.

Matej