Jul 2016 revised Nov, 2018
In March, 2016 I decided to take upon myself the project of creating an internal podcast at Shopify. When I signed in early 2013, there was 150 people on the team. In 2016 when the podcast started, Shopify was ten times that size with over 1,500 employees. When I started, you had a fairly good idea of what was going on at the company. Between lunch, team emails, our internal tool for sharing accomplishments, Friday demos and conversations in the kitchen—you had a decent picture of what people were working on. As the headcount continued to double, it became more and more difficult to maintain this overview. Finally, one day in late February, 2016 I was browsing the organization chart and decided I had to understand what more of these faces were actually doing.
Being in the middle of growth at this rate is extremely rare. Experiencing this in R&D first hand, and later, Site Reliability Engineering, I have learned much about how organizations evolve. You go from trusting a tight, small team of people with their own expertise, to trusting teams. You see people jumping around teams. Prioritization aligning across a department. Balancing hiring. Complete re-organizations. Sudden changes of direction involving 10s of people as priorities change. An increased focus on building tools and process to make everyone more effective. Projects become increasingly ambitious as the lowest hanging fruit has been plucked, requiring more cross-team communication and understanding the history of decisions that lead one up to the current state.
I wanted to understand how other departments have tackled this tremendous growth. I wanted to be able to appreciate the work that they do, which is often completely invisible to other parts of the organization if done well. I felt the best way to do that was to reach out to people, get a handful of resources and come up with a bunch of questions. Inviting people to lunch over a 3-page question sheet felt intense. However, if it would be recorded and shared with the rest of the company in some form, suddenly that’s not so weird. I didn’t want to transcribe it, because we already had plenty of text content internally. Second, this is not my full-time job, just a one hour a week side-project. Transcribing is incredibly time-consuming, as a colleague who’d done something similar in the past pointed out. Audio has become a prevalent medium in the past couple of years. Podcasts and Audiobooks becoming more and more popular. An internal-only podcast seemed like a great addition to all the videos and text we were already producing and consuming internally.
To get the initial content, I scheduled calendar events with four people. A week before the interview I’d ask them for resources about their role and projects: books, articles, videos, Podcasts, brain-dumps, whatever. This was a fantastic forcing-function to learn about areas of an organization I didn’t know much about. I learnt about business development, and got a completely new appreciation for it. Doing an interview with our government relations person forced me to learn much more about lobbying in Canada and Canadian politics. A day or two before the interview, I send my questions to the person being interviewed so they can note down key points to the questions. Since I don’t do any editing, it’s important to me people attempt to come in as poised as possible. It would also help me figure out if there was anything I missed. Because I didn’t want to end up in the rabbit-hole of recording equipment I decided that an iPhone would do. Today, my setup is a little more elaborate, but only after a dozen episodes did I invest more in this.
After having the first four interviews in the bank the next problem surfaced: hosting the episodes. There is a ton of great software out there to host your podcast. However, it’s all build with the reasonable assumption that your podcast will be public. This podcast would only be for the employees of Shopify, and would hold confidential information. Additionally, standard podcast software only supports unauthenticated endpoints. The last problem is that when an employee leaves the company, they cannot continue to receive new, confidential information on this stream. I built an extension to our Intranet that would give each employee a unique endpoint to subscribe to the podcast feed. If someone leaves the company the feed will automatically vanish. I thought it was paramount to not have another app to consume content. I believe strongly the content should be where my coworkers are already consuming it.
When it launched, it was a big hit! The first episode was downloaded by ~30% of the company. Since then, every few months. The number of listeners has climbed as it’s shared within the company and more people join the company. I have received a lot of great feedback on the podcast. Today, almost 30 episodes have been released.
- Shopify Capital
- Shopify Plus
- Individual Contributor Director
- HTTPS Storefronts
- Business Development
- Garage Team
- Government Relations
- VR & AR
- Online Marketing
- Trust, Security, & Compliance
- Black Friday and Cyber Monday Infrastructure
- Apps and Partners
- Engineering History with CEO Tobi
- Data and Machine Learning
- Support’s Involvement in Product
- Investor Relations
I have done public speaking in the past. Evaluating video recordings of public speaking and Podcasts are two completely different things. Public speaking is one-way communication. You can get a sense of the audience by attempting to assert the room, but that’s about the real-time feedback you get. Podcasts are completely different. It’s a conversation. Anything can happen. Outside of journalism, you don’t get an opportunity to evaluate the way you have conversations with people and interview them often. It’s helped me identify how to ask better questions to help people communicate their message as clearly and coherently as possible. In the beginning, I was fairly tied to my questions and their structure as it was new to me. Now, I’m more confident in running interviews and can jump around more. I still believe writing out the questions and doing plenty of research beforehand is valuable. It helps you ask the right questions, provide context for listeners and confidently go off-track and back.
Today, almost 3 years later after I started my first podcast, we have about 20 internal podcasts: the CEO’s own podcast, recordings of our weekly town hall, onboarding content, life stories of employees, training, and many others. It was a massive catalyst when it become just a few clicks to create a new, internal, secure podcast.
I highly recommend this to other companies that have hit the size where this makes sense. How many employees that means, I don’t know, and I doubt a magical number exists. Send some of your interesting employees an email, read something about their role, put a recorder in front of them and ask them questions! The trickiest part is secure hosting. Reach out to me if you have more questions about this aspect.