A cautious reopening led to an explosion of pent-up delivery orders that initially overwhelmed the restaurants.
By Charles Bililies as told to Chris Mohney
Zagat is teaming up with TechTable for the 2020 TechTable Summit on Tuesday, December 15. This one-day live virtual event will feature discussions with a blockbuster roster of restaurant industry influencers and hospitality leaders. Get your tickets here.
San Francisco was the first major city to lock down in the middle of March. We devised a system modeled after the United States Armed Forces DEFCON system, where DEFCON 5 is normal and DEFCON 1 is nuclear war. For us, if we needed to make decisions, everyone knew what each DEFCON level meant—everything from normal operation at DEFCON 5, to reducing indoor dining, to takeout and delivery only, to fully closed at DEFCON 1. We went from DEFCON 5 to 1 in a matter of three days.
We ended up keeping the restaurants closed for 115 days. We took this very seriously. We took the health and safety of our staff very seriously. We kept the entire team on through the end of March. Starting in April, we were continuing to let people use their vacation and sick time. We kept the entire management team on the entire time. We got PPP money, and we used a considerable amount of our own financial resources to keep everyone on through the beginning of June, when, unfortunately, we had to lay off the hourly workforce—about 150 people, basically after our PPP ran out.
Then we hit a point where we really had no choice but to start to reopen unless we wanted to lose the restaurants entirely. The scariest part was that by that time we were dealing with exponentially worse conditions than when we first closed. Certainly we had learned an awful lot. We had created our own COVID playbook after reading so many other plans and piecing together elements we could adapt for our own operation.
So at the beginning of July, we started the reopening process—exclusively with the management teams, about 30 individuals who were not only the most senior, but also the most experienced. Almost our entire management team had actually risen within the organization from line-level roles.
We reopened in the middle of July to overwhelming consumer demand. It was truly insane, basically breaking all of our hardware and software we had in place to execute to-go and delivery. After day three, we had to shut the systems down and revert back to an in-person ordering system that could better tolerate the order velocity. We were getting hundreds of orders at the same time, and we only had the capacity to produce so much food so quickly. People were waiting hours just to pick up food, because the line to pick up the orders was so long.
We’ve always had a little bit of an outlier problem with our tech, whether it’s point of sale, delivery, and so on, just because of our volumes. An average Souvla, pre-COVID, would serve up to a thousand guests a day through multiple channels. When you’re trying to funnel even half of that volume through one channel, they were not necessarily designed to deal with that order velocity. Thankfully, we happened to have a very, very strong partnership with Square. We have a dedicated Slack channel with the Square engineering team. We’ve broken so much technology over the years that we’ve gotten good about being vocal, about being proactive, and about being collaborative with our tech partners so that they can make these fixes.
Everyone likes to eat lunch at noon and dinner at seven. Normally, a line is the best way to throttle that demand. You can only process so many orders in person. But what if you don’t have any ability to throttle that digitally? Our system was designed to take in a predetermined amount of orders every 15 minutes. After you exceed that order number, it moves into the next time slot. So if you ordered at noon and you were one of the first to order, your food would be ready at 12:15. If 30 other people also ordered, it’s going to say your order might be ready at 12:30 or 12:45. What they didn’t account for was so many people ordering at once that it got to the point of, “Your order will be ready tomorrow.” That literally happened.
We were jamming all of that pent-up demand through two locations as opposed to all five. In the last three months, we’ve phased in the rest of the restaurants. Now we have all four of our brick-and-mortar restaurants open, as well as our delivery-only restaurant. We’ve hired back about 90 of the 150 line-level employees that we had to lay off. We are still not at full operations in terms of the hours, but we will phased that in before the end of the year.
All that’s what we spend a lot of time working on. We never wanted to publicly position ourselves as too tech-gimmicky. At the same time, we wanted to give our guests the freedom and flexibility as far as how best to order Souvla. On the back end, we automate as much as we can so that we can have all our data and all of that information flowing as seamlessly as possible.
As of now, we are only doing to-go and delivery. We are not doing outdoor dining or indoor dining, even though both are allowed to in San Francisco. We will not be reopening our dining rooms until the end of this year at the earliest. I still feel that both outdoor and indoor dining are simply far too dangerous, especially for our teams. We’re in a really, really good place right now culturally, and I think a lot of that is because we’re not doing outdoor dining, and because we’ve got really, really robust health and safety protocols in place, including biweekly in-house COVID testing.
So much of what made Souvla embody that “fast fine dining” moniker is the in-restaurant dining experience—the fact that you could come in for a leisurely lunch, or come in the evenings and the restaurants are nice enough to bring a date. You come in and have a bottle of wine, it’s served with stemware, the lights are down, there’s candles on the tables. It looks and feels like a full-service restaurant. Obviously, that element is on a hold for the foreseeable future, but we very much want to bring it back.
We’ve now stabilized to the point where we’re going to be able to ride this out, and we’re going to be fine. That’s very much a testament to the team, the product, and all of the effort that we put in while we were closed just to get us back to this point. For us, and for basically every other business, 2020 will always have an asterisk next to it. But now it’s really about starting to think about the future.
At the start of 2020, we were planning to add two new restaurants to the portfolio. We were very vocal and public about bringing Souvla to New York City, and, obviously, that is on hold. How do we move forward positively, as opposed to just treading water for the next year? I have a responsibility to my team to continue to provide opportunities, whether it’s partnerships or new restaurants and developments. So I’m looking at next year as a way to get back to growing and expanding the brand, but doing it in a little bit of a different way—like for example, in the form of interesting collaborative partnerships that may involve brand extensions. We just launched a line with Williams-Sonoma that was a lot of fun to work on. We’re thinking about other ways that we can bring Souvla into people’s homes that may not live in San Francisco.
We’re going to start looking a little bit more in our backyard, inside the Bay Area, as opposed to going to a new market that involves me getting on an airplane. I used to fly two or three times a month, traveling all over. I haven’t been on an airplane since the beginning of March. I had always believed Souvla to be an urban neighborhood brand. I think that there is an opportunity with the rest of America—those that are leaving cities and pursuing life in the suburbs—that we might be able to follow that a little bit. And even though we’re not doing outdoor dining right now, we’re going to want that with any new restaurant. Instead of having 40 or 50 seats inside and 10 seats outside, maybe that number gets flipped.
That New York Location was going to represent a monumental leap that we spent three years working on. Now it’s no more big leaps, but small, positive steps.