How I saved $216k for Eventbrite
How I saved $216k for Eventbrite
Simple design decisions can save a lot of money for big companies.
(spoiler alert — the obvious and approved design did not make it to the end)
Eventbrite is a global events marketplace that serves event creators and event-goers in nearly 180 countries. Users browse through 1000s of events to book an event.
While looking for an event there are varied parameters that a user goes through to make a decision. These include - price, location and venue, time, category of event etc. After looking at the data we figured that venue/location was a very important parameter for the user.
"As a user, I cannot easily determine if the event is
close to me or far away. This makes it hard for me to decide if I should book the event"
[My role]
At Eventbrite I was a part of the 'Land & Browse team'. That meant, not only did I have to take care of the landing and SEO pages, but also identify wider opportunities such as this project. Often such opportunities had overlap with touch points like Listing page, Booking confirmation page etc. In such cases, I would collaborate with the other teams' PMs and PDs to get a thumbs up on the design intervention. However, the designs were researched and executed by me.
[consumer research shows]
Location tops for upper funnel ticket purchase decision followed by price
[My role]
At Eventbrite I was a part of the 'Land & Browse team'. That meant, not only did I have to take care of the landing and SEO pages, but also identify wider opportunities such as this project. Often such opportunities had overlap with touch points like Listing page, Booking confirmation page etc. In such cases, I would collaborate with the other teams' PMs and PDs to get a thumbs up on the design intervention. However, the designs were researched and executed by me.
[consumer research shows]
Location tops for upper funnel ticket purchase decision followed by price
[opportunity]
We started with shortlisting opportunities across a few surfaces including (but not limited to) search, listing, homepage, city browse etc.
We dived into the user journey, research insights and data analytics of all these surfaces.
[listing page data]
-Average 24 million users per month visit listing page
-Out of this 1.2% spend more than 3 mins on the page going through details
-Out of this, 68% book an event
-On the other hand, users who spend less than 2 mins, only 30% make an order
The idea is to increase the denominator of users who are looking for the distance or location of an event. Here we are going forward with the hypothesis that if a user is spending more time on the listing page they are scrolling down to the location section and spending time to understand the location (amongst other things).
If we can increase the 1.2% of users who are engaging with location/distance, then we will be able to harness the 68% CVR better
[current state]
As you see in the screenshots below, the user would have to scroll through the page to fine the venue and location. If the user was very interested in the event, they would drop off this page to check the location on Google maps. This would often lead to an abandoned journey.
[simple + obvious solution]
One might say, an obvious solution is to have an interactive map show up on the listing page (in addition to other touchpoints). That is exactly what I did too. Below is a screenshot of the design I proposed. This design got positive feedback internally and was approved to go live. (spoiler alert - this was not the final design!)
The idea is to increase the denominator of users who are looking for the distance or location of an event. Here we are going forward with the hypothesis that if a user is spending more time on the listing page they are scrolling down to the location section and spending time to understand the location (amongst other things).
If we can increase the 1.2% of users who are engaging with location/distance, then we will be able to harness the 68% CVR better
[current state]
As you see in the screenshots below, the user would have to scroll through the page to fine the venue and location. If the user was very interested in the event, they would drop off this page to check the location on Google maps. This would often lead to an abandoned journey.
[simple + obvious solution]
One might say, an obvious solution is to have an interactive map show up on the listing page (in addition to other touchpoints). That is exactly what I did too. Below is a screenshot of the design I proposed. This design got positive feedback internally and was approved to go live. (spoiler alert - this was not the final design!)
[BUT…]
Before this design could go live I decided to do some digging. It struck me— how come this has not been done before at Eventbrite? Considering this is an event platform, putting a map on the listing page seemed like an obvious element.
But, I also knew that a lot of the design and product teams were new at Eventbrite (1 year old or so), just like my team.
Upon asking around I figured something important. This information did not come from my design manager or lead or even the PM. I found out that—
However, when I got my design reviewed and approved, no one raised this issue because often such things get lost or buried. Some PM somewhere long back probably figured this would cost a lot of money and probably did not think it was feasible at the time.
Adding the map on the listing page will cost us USD $216k per month by calling the said google API
Adding the map on the listing page will cost us USD $216k per month by calling the said google API
However, when I got my design reviewed and approved, no one raised this issue because often such things get lost or buried. Some PM somewhere long back probably figured this would cost a lot of money and probably did not think it was feasible at the time.
[what next?]
I reviewed this case with my team— Product manager, Engineer manager and Design manager. This was to determine, are we in a position to spend $216k? From a product design perspective I was sure this solution would move the needle in a positive manner; but overall was this worth it?
[contextual problem solving]
At this point in time we knew that a complete redesign and rebrand of the product was underway. I checked with the team working on the redesign of how they were tackling this. They confirmed that they have budgeted for this in the new designs.
So keeping this in mind me and my team decided to approach the map+location element in phases.
Phase one would include a 'Show map' CTA that would hide the map behind the button. This would mean the API would not be called upon unless the user showed the intent of checking the location details.
Phase two would include a 'x miles away' tag along with map CTA. This obviously would need more feasibility from tech, as up till this point Eventbrite was not calculating 'x miles away'. We would spend more time with product to figure out if the 'x miles away could be calculated from the user. Maybe we would show 'x miles away' from city centre. But would that make sense to the user? This would all depend on the roadmap and how close to the redesign we would be at this point in time.
And finally, with the redesign of the website and app we would be able to show the map at all times
Out of which 72% users made an event booking
Out of which 72% users made an event booking
$216k per month was saved by not putting the map API open to all. This gave us enough time to plan for this for the eventual release
$216k per month was saved by not putting the map API open to all. This gave us enough time to plan for this for the eventual release
Dig deep
Often the obvious answer might be infront of you but as a product designer sometimes you have to think beyong design and think of a strategy that caters the business
Dig deep
Often the obvious answer might be infront of you but as a product designer sometimes you have to think beyong design and think of a strategy that caters the business
Phase-wise execution
Thinking of how to break down a simple design into phases is important to see better gradual results.
Phase-wise execution
Thinking of how to break down a simple design into phases is important to see better gradual results.
