How OpenTable improves UX by getting you in, out, and on your way - Mixpanel

How OpenTable improves UX by getting you in, out, and on your way

Justin Megahan

Creating the shortest path between you and a delightful meal

“It’s not about how much time you spend in our app,” says Alexa Andrzejewski, eschewing any notion that simply more user engagement is better. Alexa wants the right engagement. “We want to create the shortest path between you and a delightful meal.”

Exactly what that path is to your meal could be different for a hundred different reasons. At OpenTable, where Alexa is Mobile Experience Director, their top concern is that your trip is short and the meal is perfect for the occasion.

Sometimes that path means looking at a single restaurant profile page, sometimes it means looking at seven.

“In a high stakes situation – like if you’re going on a date or taking someone out for a business meal – you’re more likely to book a familiar restaurant.”

So finding and booking one specific restaurant for next Friday has to be easy. But so does finding just the right spot for tonight – even when you aren’t quite sure what that spot is. That’s the tricky part for Alexa, who came to OpenTable through the 2013 acquisition of Foodspotting, which she co-founded. Her team is constantly balancing the motivations and priorities of each and every person that picks up their phone and opens up OpenTable.

alt text

“Like high speed rail, OpenTable is not the destination in itself, but it should be the most delightful way to get there. You shouldn’t be spending a ton of time in our app. That’s not the goal. We want to get you from here to there in the fastest way possible, while making sure we do it in an enjoyable way. So every time you’re taking high-speed rail, you’re using us. And that means frequent, but short dips in and out of the app.”

It’s easy when the user knows exactly what they want. But what do you do when they don’t? Or if their first choice isn’t available in the time they need? They’ve got a decision to make, and OpenTable needs to get them not just options, but the right information they need to make that decision.

Anyone who’s spent half an hour on Netflix not deciding what movie to watch understands that the answer isn’t always more options.

“For a restaurant like Flour + Water, we could surface the most popular dishes, the scarcity of availability, or the transit time to get there–but different things matter more to different people, and even vary based on the occasion.”

alt text

But, then again, what helps me make my dining decision probably isn’t what will help you make yours.

“In creating the next generation of OpenTable, we want to better understand what matters most to a given user for a given occasion and surface the right restaurants and the right information at the right time.”

I live in a city and my regular source of transportation are my size nine Chucks. The difference between one mile and seven is pretty drastic. But my brother in southern Delaware would gladly drive an extra fifteen miles for a worthy beer selection. OpenTable’s looking to help us both.

“People need reasons, not just recommendations. Why should I try this restaurant vs. that restaurant? Recent additions to our app like Popular Dishes are aimed at helping people make these decisions. But in upcoming versions of OpenTable, we’ll be doing even more to personalize the decision-making experience.”

Thinking longer term

Taking an even longer perspective, Alexa’s team is also trying to find the right balance between meeting users’ immediate needs and setting expectations for what else they can get from the app.

Half of OpenTable reservations happen on the app. And app users tend to have different priorities than the web users OpenTable first built their business around in the late 90s. While those visiting are more likely to make reservations for an upcoming outing, those opening the app are more likely to be looking for a place to eat tonight, or even right now.

And that’s a behavior they’d like to encourage.

“During lunch and dinner, we’re testing defaulting to “Dine now’ on the app.”

When users realize that OpenTable can answer that immediate question of, “Where should we eat tonight?” they’re more likely to come back on a whim, again and again.

“In the short term, we want to make it as easy as possible for you to make their reservation for whenever you need it. But in the long term we’d like to educate the user that this is a thing, and see those same day reservations grow.”

Beyond a conversion

And that’s just getting the user to convert and book a reservation. So much of the experience, and what will determine whether the OpenTable experience was something worth returning to, happens after the booking and outside the app. Which means it’s only getting started with that conversion.

“Of course we look at things like the funnels. We’ll put them up on the wall and look at the biggest drop-off areas. Then we’ll brainstorm what we think the causes might be. But then we actually talk to our users to know what’s going on out in the world. We spend a lot of time on qualitative research. The important thing about designing an app like OpenTable, is remembering a lot of the dining experience happens outside the app.”

That’s what has helped turn OpenTable into a service that more than 16 million diners use each month to book a restaurant. Not just allowing you to make a reservation for you next dining occasion, but becoming the place you’ll return to again and again for the best dining experience.

And that’s something that a simple “bookings/sessions” conversion number isn’t always going to represent.

“Just looking at conversions is short-sighted. There’s so much beyond that. Data tells you what is happening and qualitative research can tell you why.”

alt text

Payments and demand testing

The new version of OpenTable Payments is a recent big step to expand their service and help create a better dining experience. They understood the frustration of waiting around after a meal to pay the bill, and knew they could help restaurants streamline by allowing diners to pay right from the OpenTable app. It seemed like a promising feature, but they always do. What Alexa needed to know before moving forward was, “Will people actually use it?”

“More and more we are getting creative about how we test to see user interest, without completely building it out.”

Before developing payments, they ran a demand test. After booking a reservation, a simple survey would trigger. “Would you be interested in paying for this meal with your phone?” If the answer was “Yes,” they’d follow up with a second question, saying, “We’re still working on this feature, but tell us why you’re interested.” If “No,” they’d ask why not.

This not only gave Alexa an idea of how many people they could expect to take advantage of payments, but it also gave her an idea who those people were, and how they would like to use it. And why they might not. Of all the reasons they brainstormed why one might not use it – not being rewarding enough or not remembering – the top concern was something that hadn’t even been considered: security.

These results informed the engineers how to best build the feature before they even wrote the first line of code. They knew there was a shift towards mobile payments in general, but they were able to focus on the users who would use it today, which, it turned out, were those using Apple Pay. And they were able to promote the security aspect of being able to make the payment in OpenTable, instead of having to hand a card over to the restaurant.

“What’s the headline you’d like to write?”

Having goals for the experiments OpenTable runs is important to Alexa. Otherwise, it’s too easy to backfill your goals with the results that you got. When she sits down at the beginning of a project, she always asks, “What is the headline you’d like to write? And what is the metric you want to move, and by how much?”

As they’re testing the new “Dine Now” option, the goal is to get users to make more same day reservations. But how much more is a success?

alt text

The problem with getting too focused on a conversion is that the memorable experience – actually sitting down and enjoying a meal out – happens well after the reservation has been booked. And if it’s not a pleasant one, the user isn’t as likely to use OpenTable again. Great dining experiences are why users come back again and again to OpenTable. That’s why Alexa and her team focus so much on the post-conversion experience.

Sure, they can send you a push notification reminding you of that upcoming reservation. But what about recommending a few dishes from Foodspotting? And when’s the best time to make those recommendations? Is it in the afternoon or in my Uber ride to the restaurant?

Or what about getting you to review your meal afterwards? Every month OpenTable users leave nearly half a million reviews, not only helping other users find a delightful meal, but also tipping OpenTable off to somewhere you’d probably be interested in booking again. When’s the right time to do that? 3 hours after the meal? The next morning?

“As we are creating more features that are helping you decide where to go, one of the things that we want to test is ‘Are we achieving that goal?’ But that’s one of those things that is difficult to test, because it’s not a straight up conversion thing. You might have booked a place, but we want to make sure you book the right place for the right occasion every time.”


Get the latest from Mixpanel
This field is required.