User communities have invited commercial activity for centuries. Even the most valuable company in the world today began by selling to the ‘Homebrew Club’:
During January and February 1976 Jobs started to badger Wozniak about the possibility of making and selling some printed circuit boards so that others could build their own versions of the computer. Wozniak had not contemplated doing anything apart from handing out schematics of the machine to any Homebrew members who were interested. “It was Steve’s idea to hold them in the air and sell a few.” Jobs entertained the notion of a fleeting, informal venture that would be more of a partnership between friends than a proper company… Jobs’s thoughts about the possible market were limited to a few friends, members of the Homebrew Club, and one or two stores.
Return to the Little Kingdom, Mike Moritz
It’s not surprising that more enterprise products are landing in the curious embrace of communities today (or is it the other way around?).
How do these communities form?
As founders when we introduce the product to early users, we can usually identify a small number of really happy users who love the product, the superfans. Superfans form the early community of users around the product and often evangelize it to others, building their own reputation and adding credibility to the product in the process. With their success attached to the product, superfans can be our closest allies in closing the feedback loop for the product and in growing adoption through network effects.
What do superfans look like?
Superfans are generally aware of the problem that the product solves and actively seek out solutions to address the problem. They are willing to pay retail price for the product, are vocal about how well it works, and would be disappointed if it were killed. As influencers among fellow practitioners or champions in their organizations, superfans really shine in articulating the pain that the product addresses and how to evaluate it compared to alternatives.
Bootstrapping a community
Beyond tapping friends and personal networks, here are a few ways to bootstrap superfan communities. We’d love to hear from you about other ways that you’ve seen…
1. Find a stage
Seeing and hearing from the creator of a product establishes a human face that people instinctively trust more than a website or marketing video. As scary as it may seem to get on a stage, it’s a high leverage way to reach a large audience. Here’s Ryan Dahl talking about first presenting Node.js:
I quit my work and I worked on Node for six months, basically straight. I was completely convinced that this would be a thing. I wrote very nicely to the JSConf people and begged them to give me a slot where I could present it at JSConf EU… I was extremely scared, presenting this thing that I’d been working on for six months.
Working in Public, Nadia Eghbal
Arguably an even better way to ‘be on stage’ is to have a customer speak about how they’re using the product. Alex Solomon, CTO and co-founder of PagerDuty, describes how their booth got mobbed and they saw an instant uptick in adoption after a customer mentioned them in their keynote session at the Velocity conference in 2010.
2. Create your community
Embracing or bootstrapping your own community that’s invested in the problem space can be a powerful force in growing product awareness and adoption. A small group of early adopters who initially self-select into the viewpoints and norms of a social group that can eventually grow into a much larger force. Tristan Handy, CEO and founder of Fishtown Analytics, describes how the dbt community has grown mainstream:
The dbt Slack community continues to be the largest, most engaged, and most authentic community of professionals working in the modern data stack. The group has 3x’ed since this time last year, from roughly 2,000 to roughly 6,000 humans around the world. That growth is fantastic! It means that the dbt viewpoint is reaching more people than ever, and it also means that we each benefit from the collective wisdom of an increasingly large, diverse, group.
Four Years In: From Misfits to Mainstream, Tristan Handy
Enabling prospective customers self-select into the ‘club’ has shifted the trajectory and accelerated engagement velocity with customers who are already convinced about the product’s value, especially compared to customers who start their product trials cold.
Communities are emerging across social sites and applications such as GitHub, HackerNews (HN), Reddit, Dev.to, Twitter, Medium, Twitch, YouTube, TowardsDataScience, and so on. Helping users find community members at public discussion forums has become table stakes today. As founders get more creative with engaging and growing communities, here are a few patterns that we have observed:
- Write & Curate Relevant Content – Apart from a basic SEO friendly website, most startups host a blog with content that’s relevant and differentiated to attract prospects into the community. Others take this further with curated articles and newsletters (Data Science Roundup by DBT).
- Practitioner & How-to Videos – Short how-to videos on YouTube are a great way to help users discover the product, show how it works in practice, and help them get started on it quickly (see PulumiTV).
- Live Events & Conversation – YouTube, Twitch, and Zoom have been great for hosting live coding events, talks, and meetups. Podcasts and interviews with leading industry practitioners adds validation and taps into the communities of these influencers (see Nobl9’s channel).
- Hubs & Repositories – Some companies are going further by creating ‘hubs’ by publishing a gallery of recent projects and models (HuggingFace, Weights & Biases) that enable practitioners to learn and get started with prior work from other leading practitioners.
3. Invest in Live Conferences
As the community grows, live conferences work well to bring the community together regularly, educate practitioners, and showcase customer success stories. As these events gather scale, they demonstrate the network effects of a healthy community led go-to-market strategy.
Amy Barone, who led Marketing Events and Engagement Programs for Tableau (2007-18) describes how they shifted focus from sponsorships at trade shows to sponsoring their own hosted events with ‘large-scale conferences, segment-focused summits, product launch events and roadshows, and user groups’:
We knew we needed to get our brand and product in front of people in a meaningful way. In 2008, we squeezed 187 people into a conference room at a hotel, and launched our very first customer conference. Ten years later, our annual conference attracted nearly 20,000 customers.
All these events, especially our hosted events, have been integral in building our brand and community and have played a huge role in growing at the scale and level and pace that we did.
Tableau developed different event types based on their audience goals and sales objectives. For example, an ‘executive roundtable’ helped influence executive buyers where there might already an open opportunity. Educational events and workshops were meant to educate practitioners and aren’t evaluated sales goals. Large scale events targeted every stage of the sales cycle from awareness to accelerating proof-of-concepts, to expanding usage.
This year as live conferences have gone wholly virtual, I’ve been pleasantly surprised by the low travel and lodging investment we now need to have meaningful 1×1 conversations at the event or after.
4. Build an Open Source Community
Companies such as Elastic, Confluent, MongoDB, Databricks, and Grafana have built strong commercial products alongside their open source communities. Open source communities are worth covering separately because the success of these communities depends on the broader open source strategy around the product. This requires establishing tenets around the governance model as well as a transparent approach around commercialization.
A well-defined governance model establishes the processes and cultural norms around the project and clarifies questions such as who the final decision makers are and whether external code contributions welcome. As the Linux Foundation notes, a plan for commercialization helps build a sustainable model for innovation and maintenance:
Long-term, sustainable projects arise from feature-rich developer communities whose code is productized in commercial products that profit businesses which in turn reinvest back into the projects. The goal is to enable a virtuous ‘projects build products that generate profits that get reinvested back into the project community’ lifecycle.
Mike Volpi has talked about the first wave of open source commercialization with RedHat offering enterprise-grade support subscriptions for their Linux distributions. The second wave was led by software such as Hadoop that was developed primarily within a single company (Yahoo) and commercialized by Cloudera and Hortonworks. Their business model was built around unlocking enterprise features for a charge while the core product remained free and open source. As customers have moved to the cloud, open source business models have shifted to offering software-as-a-service (SaaS) where the economic value of the product accrues to the company that carries the pager and owns operating the software.
Most companies today straddle these business models even they continue to shift towards the SaaS model. Kubernetes, one of the most active projects on GitHub, was started in 2014 by Craig McLuckie and Joe Beda. Craig and Joe went on to found Heptio, a Madrona portfolio company that was acquired by VMware for $550 MM in late 2018. Heptio combined key dimensions of these models with Heptio Kubernetes Subscription (HKS), offering production deployments of Kubernetes with 24×7 support, including hot fixes and tools. Craig noted in blog post:
“We believe that Kubernetes is a seed technology that enables businesses to transform their approach to IT, significantly reducing infrastructure costs and simplifying operations. While the technology is incredibly powerful, we have seen developers struggle to get up and running quickly. Heptio’s early focus is on making Kubernetes more accessible to developers running apps on-premises or in the public cloud. In the future we plan to work closely with the open ecosystem to advance the platform, and deliver the features enterprises need to run Kubernetes at scale.”
Pulumi, another Madrona portfolio company, offers a modern way of creating, deploying, and managing infrastructure as code. As founders, Eric Rudder and Joe Duffy believed that open sourcing the core technology was table stakes for their developer customers. It encourages contributions, helps developers understand how it works, and gives customers confidence that they will continue to have the source code for the product regardless of what happens to the company. They also knew from day one how the revenue opportunities would unfold and found that the open sourcing the product aligned well with a SaaS model (learn more on Building an Open Source Business from Scratch featuring Eric and Joe in conversation with Soma, MD at Madrona).
Open sourcing the core product has traditionally offered developers (a) the ability to use the software without charge, and (b) access to the code.
It’s clear that the ability to start using the software quickly with production-grade reliability and without hitting paywalls has been clear adoption drivers for enterprise products.
It’s unclear whether customers today care about free access to the code. I believe they do. As enterprise systems become more interdependent, stakeholders will demand deeper transparency for the software they’re relying on for business-critical functions. Sharing how the product works builds trust, which is ‘the #1 thing we’re selling’ (Jeff Lawson, CEO of Twilio). In that sense, code may be the currency of trust in enterprise software.