Booking.com Insurance

The team and I were involved in the inception of this MVP, and we found out that customers who book non-refundable accommodations do it out of choice or necessity due to availability.

Themes such as flexibility and user control became increasingly important, especially during COVID-19 pandemic, as we found out during our research.

How might we provide more flexibility to our non-refundable accommodation bookers?
Short answer: myself and my team focused on delivering an insurance product that would solve the pain points mentioned above.

I participated in weekly sprints with direct stakeholders: consultants, product managers, researchers & data scientists; our goal was goal to bring an insurance product in the accommodation area.

During the sprints I contributed with mapping the competitors and benchmarked the lifecycle of insurance products (sale & post-sale), including how other verticals such as Flights or Car Rentals are integrating insurance in Booking.com.

At this stage it was all about learning and observational analysis, things like similar insurance products, additional insurance offerings and single trip products arose as topics from this stage.

Using the results of the sprint in combination with our research findings, I could create, review and validate the first prototypes. After ideating, I concluded with two options, a single transactional screen & a progressive disclosure funnel.

I facilitated a design critique to capture feedback from all the stakeholders, and then I tested both versions (1 and 2) with potential customers.

The key learnings we got helped me to understand the level of user comprehension and enabled me to tweak the prototype until I was satisfied with the responses I was getting from potential users.

We took some wrong turns on our way to the MVP release, managed to steer towards the right path thankfully to research and data backed decisions.

Backed by research (pivot 1)
I started first with an assumption that users would prefer to purchase an insurance after they complete their accommodation booking. It turned out that in our studies, users were more likely to purchase an insurance during booking their accommodation.

The data is right (pivot 2)
‍Our second assumption was to start the MVP on mobile, however our data indicated that our target group is bigger if we switched to web first.

I figured this holistic way was the right approach because it was also an opportunity to learn and be open about what are all the ways are users interacting with accommodations, be it amending it, cancelling it, or simply viewing the details of their policy.

I used the jobs to be done (JTBD) framework  to clearly outline each action a user must take in order to reach their goals. Later, these evolved into user scenarios offering greater visibility about our potential customers by mapping each goal, task, and expected outcome.

Here’s an example of using JTDB below:

I used the scenarios in combination with high-level data to craft user flows that included each step of the user tasks. These were critical both to map out our MVP scope but also to  get approval (content-wise). These helped me to visually communicate the intention with the insurance provider and hence, reach some agreements.

I facilitated design critiques to get feedback on the prototypes and after implementing the feedback, I’ve tested it with users.