Community building

Build & grow your data science community

Communities help us discover new capabilities, connect with others, and solve problems faster. They help us get unstuck and prevent us from reinventing the wheel.

people passing blocks in a line, illustration

First things first

It doesn't have to be perfect right away.

Try new things and see what works best. Everything isn’t set in stone. Experiment to see what works well.

Identify your goal

The best communities start with a simple question: what do you want people to get out of this? Whether it's connecting colleagues, meeting other data people in your neighborhood, or solving shared problems, your goal shapes everything else.

Start small

You don't need a big audience to get started. Find a core group of five people, and be intentional about including diverse perspectives from across the business from day one. Hold one event and build from there.

Be consistent

Communities don't grow overnight, but they do grow when people know what to expect. Pick a cadence, whether it's once a month or once a week, and stick to the same day and format. Consistency helps turn a one-time attendee into a regular.

Getting Started

Community isn’t one thing

Many organizations and groups take a multi-pronged approach. It's not just a Slack or Discord server, it's the events you run, the way you reach people across teams or across a city, and everything in between.

A few examples, adaptable to either setting:

  • A casual social gathering (coffee shop, brewery, a walk around the city, etc.) with no agenda beyond spending time together
  • Monthly community meetups with 1 presentation or with 3 lightning talks

     

  • A TidyTuesday built around a shared dataset (a company dataset internally, or a local/open dataset for a regional group)
  • A monthly spotlight on a community member, featuring a short "10 questions" interview

     

  • A curated monthly newsletter for the community
  • An annual conference or meetup day (Shiny Day, for example) with a live workshop and a half-day of groups presenting their own apps and use cases
  • A central repository or website holding all the material: learning pathways, recorded talks, and more

     

  • Think about who you already know, or kind of know. A great talk at a conference, a presentation at your own company, someone from a Data Science Hangout, a blog post that stuck with you — if someone's work resonated, tell them. It's a great opening to ask if they'd want to help build something new together.
  • LinkedIn and similar tools make it easy to find people with shared interests — search your area (or company) for members of groups like the R Project for Statistical Computing or the Python Developers Community.

     

  • Regis James, Associate Director of AI and Data Science at Regeneron thinks about community building the way an engineer thinks about a pipeline: identify members through the company directory, reach out once with a warm standardized message, and onboard responders. The result is an engaged community that keeps growing organically.

     

Many communities meet virtually these days, but in-person data meetups are picking up again. You don't need your own office to host one. Here are a few places to start:

  • Local companies hiring data scientists: many will host your group for the night in exchange for a chance to promote open roles
  • Colleges in the area: often have event space they're willing to share
  • Co-working spaces: great partners, and they like showing the community what it's like to work there
  • A member's own office: ask around; more people have access to open space than you'd expect
  • Coffee shops or breweries: good for casual, no-agenda social gatherings

Consistency lets you standardize. Whether it's office hours or a monthly meetup, create an invitation template, a logo, and a format you reuse every time. It takes a lot of the planning and scheduling off your plate. Once you have a fixed cadence (the same day and time each week or month), people can slot themselves in around their own schedule, rather than you rearranging each event around a specific speaker's calendar. Planning gets easier once you're a few months ahead.

We've seen consistency work for a range of community initiatives:

  • Posit's Data Science Hangout runs every Thursday at 12pm ET. A different data science leader co-hosts each week to answer audience-led questions with no presentation, just Q&A on things like "How do I talk to communicate more effectively with business stakeholders?" or "Should I have a centralized or decentralized data science team?"
  • John Deere hosts bi-weekly office hours for tools like Power BI, Tableau, and Posit Team, giving beginners a place to bring questions and connect with power users.
  • AstraZeneca runs a Lunch & LeaRn at lunchtime, alternating between European and US time zones.

Consistency alone won't fill the room right away. It takes time to catch on, and it works best alongside the other tips here.

large conference audience looking at a stage

Can we help you get started?

Tips for Building Engagement

Don't wait for volunteers, invite them!

Two things that trip up new community builders: waiting for volunteers and trying to please everyone. Communities that thrive aren't the ones waiting for people to raise their hands. They're the ones with someone behind the scenes personally inviting, welcoming, and planning. It's a bigger job than it looks, but it's also what makes all the difference. Libby and Rachael at Posit have learned most of what they know about community building by just getting started and actually doing it.

Libby Heeren, Host of the Data Science Hangout

Think of virtual events as someone coming over to a party at your house. When someone walks in the front door, we don’t expect them immediately be the life of the party. Similarly, it is going to take time for someone to start leading conversations and/or presenting at a community group.

Welcome them in and show them around. Welcome people into our virtual space the same way: say hello, introduce yourself, let them know how they can ask questions, explain what the space is and how they can interact with others, etc.

Start a conversation to introduce them to someone else. Try to introduce people who have shared interests, bring people into the conversation, or simply give permission for conversation to happen in the chat as well

Engagement grows when people feel comfortable. It's the small intentional gestures that turn a one-time attendee into a regular.

  • Reach out early: "We're planning out the next few months of events, would you be open to sharing your work on x, y, z? I'm always happy to chat through some ideas too!" 

     

  • Offer a series of short, lightning talks. Lower stakes make it easier for someone to say yes, especially if it's their first time presenting.

  • Ask "What's something you're proud of lately?" or "What's something new you've learned this month?" It gets people talking from a place of what they've learned, not what they're missing.

  • Make room for people who aren't ready to present yet. Slack, Teams, or Discord channels can give a lower-pressure way to start contributing and meeting others.

Remember that every event could be someone's first time there. It’s important to explicitly tell people what to expect and that all are welcome. For example:

  • Let people know this is an inclusive space, open to all regardless of their experience or background
  • Make it clear how to ask questions and give people an option to ask anonymously
  • Explain ahead of an event that people are free to join to just listen-in as well and don’t necessarily have to participate

     

  • Be aware of terms/lingo that new attendees may not know. Share definitions in the chat when acronyms come up.

A welcoming environment may look a lot different to you as the organizer than it does to a new attendee. Are there people that have joined who may be open to giving you feedback? Try connecting with a few members to learn from them about their experience. For example: How did they find out about the group? Did it take them a few months to feel comfortable joining after learning about it?

Additional community resources

How to start your own Data Science Hangout

Rachael Dempsey & Libby Heeren

Running a polyglot data science community

Melissa Van Bussel

Novices to experts: building an engaged community

Natalia Andriychuk

R User Group community explorer

Ben Ubah

Community lessons growing up in a bar

Rachael Dempsey

Building an internal data community at work

Chris Engelhardt, Dooti Roy, James Laird-Smith