“It’s not about trying to do everything and instead giving significant opportunities to our partners, who in some cases will solve problems in new ways that we might not have thought of.”
For our latest ‘What I’ve Learned…” interview at OpenChannel, we sat down with Blair Beckwith of Shopify. Joining Shopify in 2012, Blair is one of the few people that has grown a leading developer ecosystem from the ground up. He is currently Developer Relations Lead at Shopify and was a Developer Advocate helping to engage developers around the Shopify platform in the early days. He is also a member of the Strategic Advisory Council for the Small Business Web, a trade association for companies selling cloud software for small businesses with over 1000 member companies, and former organizer at Startup Weekend, helping startups to build, network and learn. We were excited to sit down with Blair and hear about what he’s learned along the way…
OpenChannel: Few people have seen and managed the growth of a developer ecosystem like Shopify’s, can you talk about what the ecosystem looked like when you started and where it is now?
Blair Beckwith: When I joined Shopify, we were interesting in that we provided our open API but we also provided a marketing opportunity for developers to get what they built into the hands of customers. When I started in early 2012 we had about 80 apps in the Shopify App Store.
We had some larger integration partnerships and a lot of smaller developers as well. These smaller developers were usually Shopify merchants or partners building custom bits of functionality and then deciding to productize that as an app in the app store.
If a partner was making $1,000 a month it was seen as a big success back then. Now we’ve got about 1600 apps in the app store and some of those smaller developers have turned into large companies in their own right.
We’ve got partners who are selling millions of dollars of apps that are built exclusively for Shopify and still have those large integration partnerships as well. Back in 2012 an integration might have been a small part of a developer’s business but now Shopify can often be the largest driver of customers. That’s been a fun transition to watch.
OC: What contributed most to the growth in the number of successful apps?
Blair: At a product strategy level, we’ve done a good job of being clear internally about what we want to build versus what we want to leave up to the developer community. That said, we still have work to do on the transparency side and communicating that out to our community.
We often say, “Shopify is going to build what most merchants (Shopify customers) need most of the time.” We want to make the essential things easy for merchants, and everything else possible through the API.
We’ve been able to identify opportunities for third-party developers to capitalize and that’s been one of the biggest drivers of platform success. It’s about not trying to do everything and instead giving significant opportunities to our partners who in some cases will solve problems in new ways that we might not have thought of. For example, we launched a product called Shopify Shipping that’s really great for the majority of merchants but a company that focuses full time on shipping may build functionality that serves different segments, whether that be higher volume shippers or shippers in geographies we haven’t tackled.
OC: What do you think are the best ways to communicate your roadmap and be transparent with developers?
Blair: I’ve seen some really interesting examples like Slack with their public Trello board. That’s awesome to see.
We try not to be prescriptive and instead share more of our thinking behind it. When you tell your developers to build something you can open yourself up to problems. We might tell a developer to build in one space in 2013 and then that space might become strategically important to us in 2015. You run into trouble there when you have to go back and tell developers “I know we told you to build this, but actually now it’s something that we need to own.”
Instead we try to say “These are the areas that we care about and that means we’re going to be building out a lot of functionality ourselves in that space. – so if you want to be safe, it’s maybe best to stay away from those areas.”
But I don’t think developers should look at those spaces as being completely off limits either. The Shipping example is a great one; we are going to focus on building the best Shipping solution for a typical Shopify merchant. If you’re going to build something targeting the same segment, then that might not be the best idea. If you’re going to solve unique problems in a unique way that serves a different segment or a different need, that can be a good choice.
OC: If you were starting a developer community today, what would you focus on?
Blair: One of the biggest things that has contributed to our success is showing the business case for the API. People have a perception that third party developers just want to build really cool things. That’s definitely part of it but a lot of developers have higher aspirations. Framing the business opportunity is really important.
That relies on there being a business opportunity and making sure the platform fundamentals are in place. A lot of developers and potential partners understand money. They want to see a path to money through your API.
OC: What are the most important metrics you track and why?
Blair: One of the key metrics that we track is active developers. It’s a really important metric if you do it right. We’ve spent some time coming up with a definition for what an active developer is.
OC: Can you share your definition of what makes an active developer?
We look at it first like a traditional funnel. To get started as a developer you register for your partner account. You then you create an API key. Then you create a test Store to test your app. You make an API call with the client you’ve created.Then you submit the app and it’s published on the App Store. That middle step of making an API call is what we’ve settled on as our true active developer number.
We’ve done a good job of making it incredibly easy to create a developer account. It’s so easy that sometimes a merchant who’s not technical will end up creating an API key by mistake not even knowing what that means. That’s why we’ve settled on actual API activity as the key metric.
The entire funnel is important as well because active developers in and of themselves doesn’t necessarily serve any business goals. It does indicate activity but every step of that funnel is important and we do different things to drive activity at each stage of that funnel.
OC: You said “things to drive activity.” if we pick making an API call, what’s an example of an activity you can do to drive the activity (an API call) that you want?
Blair: Not every API is immediately intuitive, though we wish they all were, so we put a lot of effort into our documentation and tutorial-style content. The biggest driver for that step is making sure we’re providing the right content and making sure that we’re surfacing it at the right time.
You can build the most beautiful documentation but if developers aren’t finding it when they need it, then that’s a big problem. Contextual documentation is important, for example when they create their API key, giving them a path forward from there is helpful. One thing that we don’t do a good enough job with yet is giving developers a shop that looks and acts like a real shop – it has existing data like orders and customers.
There’s no point in telling a developer, “Why don’t you ping the orders end-point?” if all they get back is a list of no orders. Something we need to do a better job with is creating a simulated environment because making those first API calls result in something interesting means they’re more likely to stick around.
OC: How do you allocate time and resources on your developer relations team?
Blair: We’re lucky from a company structure standpoint that we’re an API-driven company. When our R&D team is building features for our merchants, they’re doing it with our own APIs. So when it comes to releasing new APIs and and covering what our developers need in a platform, we have the support of the entire Shopify organization behind us.
We’ve got tens of thousands of developers versus maybe a thousand when I started five years ago. When some of our smaller partners had kids I knew their names. It was so small and I could keep the entire community in my head.
Now the community is a lot bigger, you still want to deliver the best developer experience but you have to do it in more scalable ways. That’s something we’re still figuring out. I look at large developer platforms, even all the way up to Apple. I don’t want to call them a completely faceless developer organization but that’s the way it appears to 99% of the developers. That’s something that I don’t ever want to be.
We segment our developer community into verticals. If you’re building something in the email marketing space, we have a developer advocate that’s dedicated to that. It allows them to build deeper relationships and domain expertise that is valuable to both Shopify and our partners in that domain. That’s been really good for us when it comes to scaling the platform.
We also empower the rest of the organization to help developers. Sometimes a developer will contact our broader customer support organization. It would be easy for our support team to tell them that we don’t support developers through those channels, but we try to help them understand the developer community and some of the problems that they’re facing. We probably have one of the largest customer support organizations around. With our developer support team being around 20 people, having a huge support organization that supports our developer community is huge.
OC: Shopify held Unite. Why did you do a developer conference first and what are the most important things when running a developer conference?
Blair: Unite is really interesting. I’ve involved, spoken the last two years and been involved in the content creation. I’m proud that we did a developer conference first and we did that for a few reasons.
When you’re talking to developers you’re talking to people who can have a big impact on the platform. Sharing things with merchants is important but If I share things with a thousand merchants, that’s as far as it goes. Whereas, if I share things with a thousand partners, those partners are going to interact and build for many merchants and form a solution for our entire merchant base.
For a conference it’s easy to say, “meet in person and drive action” but it should be drilled down more. It’s fundamentally about building trust. It’s about sharing what you can of your roadmap, where we think Shopify is headed and trying to surface some of those opportunities that we talked about earlier.
To build trust there are two very simple things. You need to announce features and then you need to build those features, but you have to do both. Trust is easy to talk about but there are fundamental things you can do and having Unite is one of those things.
OC: Beyond the App Store, how do you showcase developer success?
Blair: We produce great developer community content and have a partnership blog. We do a few posts a week and do some developer showcases to highlight partners. We put out video content as well and even at Unite we kicked off the conference with a video that highlighted some partners and their stories.
Some people see it as a little bit fluffy, but that kind of aspirational style content is important. It comes back to the point from earlier about showing what success looks like on your platform.
OC: What’s mistakes have you made or what mistakes do you see other companies making when manage their community?
Blair: I’ll tackle both of those. We’ve been lucky in that we haven’t had any like super crippling missteps. When you look at the industry, it’s littered with stories of platforms that forget how their developer community helped them get to where they are. it’s really a shame.
Not to point fingers but companies like Twitter took a bit of an anti-developer approach. The developer community built really great things and then Twitter decided they wanted to own those things. They had strong business reasons for that, so you shouldn’t crucify them but it’s a shame what happened. I jokingly say that working in developer relations at Twitter must be one of the hardest jobs in the world.
When the platform was growing and it was just me on the developer relations team, there were some decisions about what we wanted to developers to build versus what we didn’t. We made the best decisions given the information we had.
There are definitely some decisions I wish I could take back but those are the moments that define a developer relations program. When you have to make difficult decisions that goes against an earlier decision, it’s hard because developers are relying on you to be a stable platform for them to build their business.
How you handle that is going to dictate the type of platform you become. I’m lucky to have had relatively few mishaps but those that have happened have been impactful on my developer empathy and keeping developers at the forefront of our thought across Shopify.
OC: What developer community do you really admire and why?
Blair: I don’t want to say Twilio but I have to give them a shout out. They’re often held up as the pinnacle of developer community success in this field and even their conference Signal has turned into a type of developer mecca.
We limit ourselves when thinking about developer communities. All communities, whether developers or not, are interesting to look at. There’s one in particular, do you know the Instant Pot? It’s a pressure cooker.
OC: Instant Pot? No (as we wonder where Blair is going with this, maybe he’s hungry…)
Blair: You’re going to think this is random. It’s a pressure cooker that got a lot of media coverage and somehow has built up this thriving online community of instant pot enthusiasts.
They’re across various forums and have active groups on Facebook, it’s just incredible. Our admiration of them started off as a bit of a joke but now we’re in awe at what they’ve managed to do.
People will post on the Facebook group “I’m not sure what to do with my Instant Pot so I’m thinking of selling it.” The community will immediately jump on that thread and be like “Oh, have you tried this?” or “Try this recipe, it’s what got me to keep mine.” It’s a thriving community of really helpful enthusiasts.
When I look at our own developer community, that’s what I want to see. One of the metrics we look at in our forums is our answer rate. How many forum threads are getting answers? In the Instant Pot community I bet that is near 100% and it’s all done by the community themselves. Our forum answer rate is close to 100%, but a lot of that is done by Shopify where we’re answering questions.
That has been a goal for us because that’s how you scale a developer community, you get developers involved and answering questions and helping eachother grow because they understand the value of the community and ecosystem and they get the concept of a rising tide lifting all ships. A key thing for scaling is making it so you don’t have to do everything yourself. Instant Pot has got that nailed down, it’s incredible.
We even have an instant pot community at Shopify now. They’ve got a Slack channel where they do a competition every week. They all cook the same recipe and it’s been hilarious to watch.
OC: That’s the best answer we’ve ever gotten for that question (and apologies for the lack of faith earlier). What’s something that you’re really good at or love doing that most people don’t know?
Blair: I stumbled on leather working as a bit of a hobby and love it. When you work in tech you sit at a desk all day. Our team is lucky, we get to go out and talk to people but a lot of our work is still online doing things with bits and bytes. There’s just this desire for me to build something with my hands.
Maybe that’s my post tech end game, to start a workshop and sell leather wallets. So far it’s on a smaller scale like wallets but one day I’d love to make a pair of boots.
There’s also something about things the things you’ve built getting better with age. I love the idea that they’ll be my boots in 10 years and they’ll still be mine in 20 years. People build a connection with their stuff and I’d rather the stuff I build a connection with be something that’s going to last.
That’s something that we can miss out on in tech, seeing something that you built appear here in the real world. That’s what I love about Unite, we built it, you can see it, you can touch it.