Building Web Summit’s Dream Digital Experience with Cisco
As an events engineer, you pray for everything to go right, but you plan for things to go wrong. The good news is, with the correct infrastructure in place, you limit the impact.
Gall’s law says complex systems are built on simpler systems connected together, and that's how they evolve—events engineering in a nutshell. You have an incredibly complex scenario with thousands of people and millions of requirements from different profiles and interests—and you have to make everything work.
The trick is to break it down into actionable steps: focus on one thing at a time, address it in a repeatable way and make sure it adds to your end goal.
It takes a lot of preparation, but applying this logic to each step gives you a solid foundation that will pay dividends going forward.
And, at the end of the day, it's a live event, so issues will arise. You’ll have to roll with the punches and figure out solutions on the fly. I personally thrive on unpredictable scenarios, which is perfect for my role as Events Engineering Lead at Web Summit—the world’s largest tech event.
Before joining Web Summit, I’d always been involved in building businesses and startups, specifically within the tech sphere. I’ve spent most of my time working for myself. But as my career evolved, I realized I needed to join something larger than myself. I wanted to work on a project where I would be out of my comfort zone and would keep pushing me forward. I joined the Engineering team at Web Summit in 2016. Now, as Events Engineering Lead, my team is responsible for the technology we deploy at all our global events.
Every year, we organize four events worldwide, reaching over 100,000 people from 170 countries. At Web Summit, we have investors, over 2,100 startups, partners, and general attendees, all with highly specific initiatives, like women in tech or open source. There are many influential people presenting keynotes on anything from policy to actual deep dives into tech solutions. It's an intense mix of different interests, experiences, and backgrounds over four days. In Lisbon, our event occupies two venues, side by side, which are two of the biggest venues in Portugal.
Web Summit attracts over 60,000 attendees. It is all hands on deck for 170 Web Summit staff during the event. There are only five people involved in the preparation process for the network infrastructure of the entire event, and it is managed by two of us on the ground at event time.
Building a stable network at the event is our main objective. Obviously, at a tech event, the tech just has to work. Without a stable network, we’ll let down tens of thousands of attendees. With so many people gathered together, we have several high-density scenarios to think about. We may be broadcasting Stephen Hawking’s keynote speech, while still needing to provide secure connectivity to the people roaming around the venues. At the same time, we have our partner booths, startup stands, and even our internal staff network who all rely on a stable network.
But the first concern that always kicks off the event is our registration. If the network goes down on our registration system, that is a big problem. We need to provide connectivity for attendees to download our app, complete their profiles, and register via a QR code.
Simply put: Our network can’t be a secondary consideration like it is at other events. The digital experience is a major aspect of the event. With all of these challenges, having a seamless network experience is my team’s top priority, which is where our collaboration with big industry players like Altice and Cisco started.
An Early Win
We initially worked with Cisco on our flagship event in Lisbon. We worked closely with the venue and technical partners to bring all of our requirements together—a lot of them last minute requests and changes.
Our first big save from Cisco came early in the process. The night before the event, I was setting up the registration system, which was comprised of about 80 different computers for that area. The software we developed uses automation to orchestrate all those machines at the same time and make sure they have all the necessary information. But I started to see an issue where they were intermittently going offline and then online. This could have been a disaster at an event of our size. Registration is our attendees’ first experience at the event, so if we don’t nail it, then it’s downhill from there.
At 1 a.m., I raised the issue with the network teams from Altice and Cisco. Within ten minutes, I had seven or eight people onsite with me. We spent about an hour debugging until we found the issue. It was just a small configuration tweak, and suddenly, my work was done. If I was on my own I would have had to check each machine individually, which could have easily taken me four or five hours.
Running First-Class Events
Our experience in Lisbon was spectacular. Instead of fixing issues of availability or connectivity, I could finally be onsite, interacting with people to see how our network was performing. By taking routine tasks off my plate, our technical partners allowed me to get back to my job: creating the best digital experience for attendees.
The level of staff support from the network teams onsite is incredible. Cisco staff make sure everything is taken care of and constantly communicate with the team to address and anticipate any issues. This really put my mind at ease because I know that if an issue should arises, it won’t spell disaster for our event—I know we’ll be ok.
To Everest and Back
To give you a sense of the scale of our network in Lisbon, we laid enough fiber cable to reach the peak of Mount Everest eight times—that's 80,000 kilometers. We had over 2.2 million wifi sessions accumulated over the duration of the event and around 45 terabytes of traffic. At any one time, we would have upwards of 35,000 devices connected.
At the end of the first day of the event—which is our busiest day—I asked for a report on both venues to see if there were any significant network issues. Even with an event of this magnitude, the only problem we had that entire day was a report that one single network cable had been badly clipped and wasn't working. That was it. At an event with tens of thousands of people, it’s almost unbelievable how flawlessly we pulled everything off.
On the day after the event, I was having lunch in a shopping mall about 300 meters away from one of the venues. I checked my phone to find I had better wifi quality using our event network than the mall's. I took a picture and sent to all of our team, saying, "Guys, this is how awesome our network is."
Looking forward, we are actively trying to understand how we can use the model that worked so well in Lisbon to deploy networks at our other events. We have an array of peripheral events we organize as well as partner events, speaker dinners, pub crawls, and more.
But the problem is that each event is unique. Different venues have varying connectivity availability: some will have none, some will have excellent resources—and everything in between. But we know with a partner like Cisco, we’ll be prepared for whatever comes next.
Together with Altice and Cisco, we’ve created a rock-solid foundation for a successful event. We now have an understanding of how to deploy this in a way where we can focus on that strategy and the experience, and not have to worry if it’s going to work. We know we can’t stop the unpredictability of events, but we can try to organize some of it.