Launching a Gravatar feature on 1,480 conference badges

Networking was why people came, and we barely supported it

From May 2024 to March 2025 I was one of 77+ organizers of WordCamp Asia 2025, the WordPress community’s conference for Asia, held in Manila. I was on the operations team, so my job was making sure every other team hit its deadlines and clearing whatever was blocking them.

Everyone who registered filled out a short survey, and the biggest reason people gave for coming was networking. The event didn’t do much for that beyond putting people in the same rooms.

At the time I also worked at Automattic, the company behind Gravatar. Gravatar already held contact details for a lot of people, tied to their email address, and it needed more people actually using it. So the idea was pretty simple: put a QR code on every badge that opens that attendee’s Gravatar profile. Attendees get a fast way to connect, and Gravatar gets a launch in front of about 1,480 people at once. It was a product launch and a physical launch in one, because the feature and the printed badges had to work on the same day.

The WordCamp Asia organizers heard the pitch first and loved it, and on 9 October 2024 I pitched it to the Gravatar team.

What I had to work around

I didn’t own Gravatar’s roadmap, so whatever I asked for had to be small enough for their team to say yes. The pitch offered WordCamp Asia’s tech team for the implementation, keeping Gravatar’s involvement to a minimum.

The calendar was tight, because the operations team aimed to finish everything by December 2024, with a lot of people away over Lunar New Year in January. Gravatar’s build started on 11 December with a target of 20 December, and shipped on 6 January 2025, only weeks before the event.

Plenty of attendees had no Gravatar account at all, and a QR code that points at a profile that doesn’t exist is a dead end.

And there were a lot of groups to line up: the organizing team, the event producers, local team members, the Gravatar team and the venue production company that printed the badges. Most of my work at Automattic had been internal, so this was the first time I had to coordinate this many groups outside it, all alongside my day job on Automattic’s data team.

What I did

Through October and November I kept the pitch moving until the Gravatar team was in, and in late November I wrote the script that hashed every attendee’s email. Gravatar finds a profile by a SHA256 hash of the email address, so the hash could become a QR code that worked whether or not the account existed yet, and the badge carried a hash instead of an email address.

From 11 December, Gravatar’s team built the piece that handles missing accounts, with Jake leading the project, two engineers and a designer, and me as one of the stakeholders. What they built is a “claim your profile” page: when someone scans a code whose profile doesn’t exist yet, they land on a page where the owner of that email can claim it.

Once the feature was live, I worked with WordCamp Asia’s tech team and the production company to make it fit how the badges actually got made, and tested the whole process end to end. The production company generated its own QR codes, so the plan of Gravatar creating them didn’t work. We gave each profile URL a UTM parameter that ties every hash to the event, which also made the process repeatable for any future event.

With the comms team, I published a post on the WordCamp Asia site explaining to attendees how to use their QR code. Ronnie, who led Gravatar, also gave us custom NFC chips designed for the event, which anyone could link to their own profile and share with a tap even without their badge. I handled distributing them and explaining how to use them.

At the event I demoed the feature to dozens of attendees, volunteers and organizers. On day 2 I noticed most attendees had no idea they could scan each other’s badges, because the registration volunteers weren’t mentioning it, so I trained them during my shift and saw an immediate difference.

The calls I made

Build on Gravatar instead of a plain QR code

The alternative was a code that pointed to an event-only page or a contact card. I pushed for Gravatar because it already held a lot of people’s contact information, and because the same work could grow an Automattic product. The cost was that my event now depended on another team’s roadmap.

A claim page instead of temporary profiles

My original pitch suggested creating a temporary profile with basic registration details for anyone without an account, and someone on the Gravatar team suggested a minimal prefilled profile page too. We went with the claim page because the goal was the most impact for the least build time, and it only had to exist on the path where a profile was missing, so Gravatar could ship it in weeks. The trade-off was that scanning an unclaimed badge showed a claim page instead of that person’s details.

Use the production company’s QR codes instead of fighting for ours

I could have pushed the production company to change their process close to the event so Gravatar’s codes would work. Putting the event into the URL kept their workflow intact instead.

Fix the launch during the event

When I saw on day 2 that nobody was hearing about the feature, the easy option was to write it down for the retro. I retrained the volunteers on the spot instead, since the event itself was the launch and there wasn’t going to be a second one.

What happened

Across the event there were about 2,820 scans from about 1,480 attendees, so roughly two scans per attendee. Matt Mullenweg had announced it internally as the first live event to try Gravatar’s new QR code generator and thanked me for helping make it happen, and Gravatar’s team cited the work in their internal updates.

Once people understood the feature, they loved it. Many of the people I spoke to had no idea they already had a Gravatar account, or that it could hold contact details at all, and a lot of them were excited to update their profiles and start scanning.

Some of it didn’t work. Only a couple of sponsors scanned attendee badges, and because so few people understood the feature going in, many of the NFC chips we gave everyone went unused. I also don’t have a figure for how many people claimed a profile through the new page, which was the metric Gravatar’s team set for the feature.

Afterwards I wrote a recap with the numbers and the lessons, and reached out to the WordCamp Europe organizers about using the same system. They showed some initial interest, but I hadn’t heard back by the time I published the recap. In our operations retro, keeping Gravatar QR codes on badges went straight into the “keep” column.

What I’d do differently

I’d start the attendee marketing much earlier. We only emailed attendees about the feature one week before the event, which wasn’t enough time for many of them to read it, so next time I’d send several emails well ahead and put up clear signage on scanning and the chips.

I’d train the registration volunteers before the doors open, since the fix on day 2 worked right away.

I’d give NFC chips only to sponsors, speakers and organizers, so they feel like special swag, and I’d make QR scanning part of the sponsorship package, with raffles, lead capture and follow-ups built around it, so sponsors actually use it.

And I’d use the empty space on our badges to make the names and the QR code bigger.