2km or not 2km, that is the question📍

Hey Everyone! 👋

It’s been a while, so we thought it was time to keep our awesome WalkInites up to date with what we’re up to and what we’re learning on the way.

Our goal for 2020 is to make the best restaurants out there super accessible and in the long run, help you get a table anywhere you want anytime you want.

While on that journey in the last few months we’ve probably experienced our most challenging time since the company started. We’ve had to work crazy hours, got some things wrong and not executed things exactly how we would have liked. That being said, we’ve also made more progress than ever before having seen our WalkIn community grow massively to over 125,000 downloads and over 40,000 monthly active users! It's been a pretty amazing experience and we’re massively grateful to all of you for being a part of the journey 🤩

Below we will talk about how we’re using our technology to make the restaurants even more accessible, whilst trying to drive a ‘Good Queuing’ mentality and ending no shows forever!

2km or not 2km, that is the question 🤔

When talking with the busiest and best restaurants in the game, a lot of them would be open to taking bookings but won’t because it either causes loads of no shows and/or dead table time.

For example, if a restaurant has a booking at 6pm and another at 8pm but the first seating leave at 7.30pm, they have 30 minutes where they can’t do anything with that table. Then when the 8pm party then doesn’t show up and they’ve held it for an additional 15 minutes the restaurant have already lost 45 minutes on one table alone.

Given restaurants might only be at peak business for 4 hours (say 6pm – 10pm) if they have 50 tables that’s 15,000 potential minutes to go around. As a result, if this happens once on each table during a busy service (which is pretty common) then 2,250 minutes or 15% of their potential utilisation is lost. This is why really busy restaurants tend to not take bookings, it just doesn’t make sense for them – it’s too risky.

Being aware of situations like the one above, when we enabled diners to join the queue via the WalkIn app we needed to be really careful we didn’t create a platform that created no shows, but genuinely drove incremental diners to restaurants.

We felt like if we can help make that happen for the busiest restaurants around we would really learn what works and what doesn’t as quickly as we possibly could. We had some seriously heated debates in the office that would go something like:

Amos: "We're in a great position with allowing our users to join queues, but we need to focus on stopping no-shows. No doubt our users are more reliable than bookings, but we need to get to a world where we stop no-shows completely."

Archie: thinks intensely.

Frazer: “The only way the booking systems have found to even remotely tackle this problem was by taking deposits up front...”

Luca: “No way, I wouldn’t use an app to join the queue or book if I get fined if I didn't show up. I have seen strike systems being used, where you get 2 or 3 strikes and you're blacklisted from the restaurant.”

Amos: “Hmm, but if you no show 3 times the damage has already been done for that restaurant. We need to find a way to be more proactive.”

Molly: "There is also a reality that sometimes people's plans change, so anything we do we need to be careful that it's not permanent."

Jake: “We have enough data on the backend to build a profile of users, we just need to decide how it can be used to prevent people not showing up...”

These solutions and others that we thought about just felt really punitive to diners, plus we’d already seen booking platforms try them and it still not solve the problem of no shows.

When we launched the app in January 2019, we set a hard cap on the distance users could join the queue from to 2km. We felt that was the best middle-ground between giving users a tangible benefit to walking up to door, but still protected the restaurants from people checking in from so far away they’d never show up.

This was a good starting point, but in reality we knew that this didn’t provide the experience that we wanted to enable. Really, you should be able to grab a table from wherever you are (within reason!)

We knew we had a big question to ask ourselves…

2km or not 2km?

We knew consumers wanted more (as many of them told us, we’re still crying about a 1* rating on the app store because we only allowed a 2km radius.) We also knew restaurants didn’t want no shows and 2km helped protect them from that.

As we were exploring our options we came across some data which blew us away and changed our thinking on 2km generally; some cheeky app users were changing their location on their device and joining the queue from outside the 2km radius!

Amazingly, our retention (the percentage of people who get seated after joining the queue) was 98% for these users which was 6% higher than our average retention for app users. Maybe distance wasn’t the reliability driver that we thought it was?

Obviously we had to take this with a pinch of salt, because if you’re keen enough to change your phones location information to join a queue you probably really want to eat there! It was enough though to get us looking into the right info.

Therefore we knew we had to allow people to join from further away, but we didn’t know where to start.

Cheeky users = great results 🤯

To be able to really use the data at our disposal we needed to implement new tools and create new tracking methods in the database to capture everything we needed.

This took us quite a while and for a while we got nervous we weren’t making enough visible upgrades to WalkIn but we managed to complete this work and the data we were able to make decisions off made all the difference.

We did some more research looking into our most legendary WalkInites who’d used the app over 10 times, we looked at one off users and everything in between. We looked at where diners were joining queues from and average return rates.

Here’s a cool image showing a heatmap of where some of our users are joining queues:

Checkin_Heat_Map

When we started really digging into our data, we found that there isn’t a strong correlation between distance of check in’s and their no show rate, or even how quickly they came back for their table!

What we found was diners who were historically reliable would also prove to be reliable going forwards, while others who no showed two times or more in the past were far less likely of turning up at all and often would show up later than other diners.

Essentially there were “Good” app users and “Naughty” app users. As a result we felt like if users are good and helping restaurants run efficiently by turning up on time we should find a way to reward them in a big way, then on the flip side we should discourage users from joining the dark side. (We each have Star Wars aliases on our team slack channel, yes yes turbo nerds we know).

What do you mean I can join from wherever I like? 🥁

After doing a bunch of research, running some undercover trials and running some A/B testing, we’ve just released our biggest feature set to date. Announcing Dynamic Radius!

Now, the distance you join queues from will be based on your rating as a diner. So all you awesome diners that show up on time and make life awesome for our restaurants will be able to join queues from further and further away each time you use WalkIn. Forget the 2km hard cap, now you can start checking in across town if you build up a good enough user rating! For naughty diners, if you don’t show up for your table, your distance will reduce 😭 Don’t worry though, we’ve factored in ways for you to increase your radius if you then start showing up on time!

Think of Dynamic Radius as an Uber style rating system but instead of drivers or restaurants subjectively giving you reviews, it’ll be purely tied to your reliability as a diner. By doing that we can keep it super fair and non-judgemental, just a rating based on your past reliability history.

We’re starting with a 5km cap on radius but if we see that people are hitting this threshold and it’s working well for restaurants we will increase this distance in future, potentially to no cap at all!

Finally, we’ve also put one small safeguard in place to include the restaurants current wait time in the calculation – if the wait time is really short we restrict the overall distance, as you can only get so far in such a short time. On the flip side though, when restaurants get bigger queues your rating will be in full effect.

Rating iphoneX

This is cool, I want to make it better!

If you like the idea of being able to join the queue from wherever you are as our community grows we’re looking for a community of users to give us feedback and suggest improvements for product builds going forwards. Either send us an e-mail at [email protected].

For more immediate feedback: you can speak to us by clicking the question mark on the header bar of the app and just let us know any suggestions/improvements!

Much ❤️ from the WalkIn Team